Digital PDFs

AA-H225D-TC

2000

293 pages

Original

9.6MB

Document:	DECnet-RSX Network Generation and Installation Guide Sep85
Order Number:	AA-H225D-TC
Revision:	0
Pages:	293
Original Filename:

OCR Text

Networks· Communications

.. .. .... .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .... .. .. .. .. .. .. .. .. .. .. .. ..
-----DECnet-RSX
Network Generation
and Installation Guide

~D~DD~D

wore

DECnet-RSX
Network Generation
and Installation Guide
OrderNo.AA-H225D-TC
September 1985

This manual tells you how to generate and install DECnet-RSX
software on properly configured RSX-11 M, RSX-11 M-PLUS, and
RSX-11 S systems.

Supersession/Update Information:

This is a new manual.

Operating System and Version:

RSX-11M V4.2
RSX-11 M-PLUS V3.0
RSX-11S V4.2

Software Version:

DECnet-11 M V4.2
DECnet-11 M-PLUS V3.0
DECnet-11S V4.2

AA-H225D-TC
First Printing, September 1985

The information in this document is subject to change without notice and should not be
construed as a commitment by Digital Equipment Corporation. Digital Equipment Corporation assumes no responsibility for any errors that may appear in this document.
The software described in this document is furnished under a license and may only be used
or copied in accordance with the terms of such license.
No responsibility is assumed for the use or reliability of software on equipment that is not
supplied by Digital or its affiliated companies.
Copyright © 1985 by Digital Equipment Corporation

The postage-prepaid Reader's Comments form on the last page of this document requests
the user's critical evaluation to assist us in preparing future documentation.

The following are trademarks of Digital Equipment Corporation:
DEC
DECmate
DECnet
DECUS
DECwriter
DIBOL

~DmDDmD

MASSBUS
PDP
P/OS
Professional
Rainbow
RSTS
RSX

RT
UNIBUS
VAX
VAXcluster
VMS
VT
Work Processor

Ethernet is a trademark of Xerox Corporation.

This manual was produced by Networks and Communications Publications.

Contents
Preface

Introduction
1.1
1.2
1.3
1.4
1.4.1
1.4.2
1.4.3
1.4.4
1.4.5
1.5
1.5.1
1.5.2
1.5.3
1.5.3.1
1.5.3.2
1.5.3.3
1.6

Making Preliminary Decisions ....................................... 1-1
Performing a SYSGEN .............................................. 1-2
Performing a PREGEN ............................................... 1-2
Performing a NETGEN ............................................... 1-2
NETINS.CMD ........................................................ 1-3
NETREM.CMD ....................................................... 1-3
NETCFE.CMD ..................... , ................................. 1-4
NETCFG. TXT ....................................................... 1-4
CETAB.MAC ............................................... , ......... 1-4
Post-NETGEN Considerations ....................................... 1-4
Installing DECnet .................................................... 1-4
Testing a Newly Installed Network .................................. 1-4
Reconfiguring and Tuning after NETGEN ............................ 1-5
Configuration File Editor (CFE) Commands ......................... 1-5
Network Control Program (NCP) Commands ........................ 1-5
Virtual Network Processor (VNP) Commands ....................... 1-5
Reading Paths ........................................................ 1-6

Planning DECnet
2.1
2. 1.1
2.1. 2
2.2
2.2.1
2.2.2
2.2.3
2.2.4
2.3

RSX-ll Configuration Requirements to Support DECnet-RSX
Software ............................................................. 2-1
Requirements for the Host System ................................... 2-1
Requirements for the Target System ................................. 2-2
Network Configuration Guidelines .................................. 2-3
General Guidelines .................................................. 2-4
Ethernet Configuration Guidelines .................................. 2-5
Areas and Area Configuration Guidelines ............................ 2-7
Guidelines for Specifying Routing Parameters ...................... 2-10
Selecting DECnet Components ..................................... 2-11

Contents-1

2.3.1
2.3.1.1
2.3.1.2
2.3.1.3
2.3.1.4
2.3.1.5
2.3.1.6
2.3.1.7
2.3.1.8
2.3.1.9
2.3.1.10
2.3.1.11
2.3.1.12
2.3.1.13
2.3.1.14
2.3.1.15
2.3.1.16
2.3.1.17
2.3.1.18
2.3.1.19
2.3.1.20
2.3.1.21
2.3.1.22
2.3.1.23
2.3.1.24
2.3.1.25
2.3.1.26
2.3.1.27
2.3.1.28
2.3.1.29
2.3.1.30
2.3.1.31
2.3.2
2.3.2.1
2.3.2.2
2.3.2.3
2.3.2.4
2.3.2.5
2.3.2.6
2.3.2.7
2.3.2.8
2.4

Getting Ready for RSX-11 M/M-PLUS NETGEN
3.1
3.1.1
3.1.1.1

Contents-2

Optional DECnet Components ..................................... 2-14
MiccocodeLoader(MLD) ........................................... 2-14
Event Logger/Collector (EVLlEVC) ................................. 2-15
Direct Line Access Controller (DLX) ................................ 2-15
Network Control Program (NCP) ................................... 2-15
Network Management Device Driver (NMDRV) .................... 2-15
Event File Interpreter (EVF) ........................................ 2-15
Digital Communications Process (DCP) ............................ 2-15
Ethernet Protocol Manager (EPM) .................................. 2-15
NETFOR.OLBandNETLIB.MLB .................................... 2-16
Network Information and Control Exchange (NICE) ............... 2-16
Event Logging Receiver (EVR) ...................................... 2-16
Network Display Utility and Server (NTD/NTDEMO) .............. 2-16
Link Watcher (LIN) ................................................. 2-16
Network Verification Program (NVP) .............................. 2-16
Down-line System Loader (DLL) .................................... 2-16
Up-line System Dumper (DUM) ............... " ............. '" .... 2-17
Console Carrier Requester (CCR) ................................... 2-17
Host Task Loader (HLD) ............................................ 2-17
Satellite Task Loader (SLD) ......................................... 2-17
Network File Transfer Utility (NFT) ................................ 2-17
File Transfer Spooler (FTS) ......................................... 2-17
File Access Listener (F AL) ........................................... 2-17
Command File/Batch File Submission Task (MCM) ................. 2-18
Remote Terminal UtilityRMTACP(RMTIRMTACP) ................ 2-18
Remote Terminal Host ACP and HTDRV (RMHACPIHTDRV) ...... 2-18
Network Command Terminal Server Process (NCT) ................ 2-18
Remote Command Terminal Host (RTH) ........................... 2-18
Terminal Communications Utility (TLK) ........................... 2-18
Listen Utility (LSN) ................................................. 2-18
Phone Communications Utility (PHO) .............................. 2-18
Remote Task Control Utility (TCL) ................................. 2-19
Optional PSI Components .......................................... 2-19
Data Link Mapping (DLM) .......................................... 2-19
Trace Capture Task (TRA) .......................................... 2-19
Trace Interpreter Task (TRI) ........................................ 2-19
X.29 Terminal Access ACP with HTDRV (X29ACP) ......... , " .... 2-19
KMX Microcode Dumper (DUK) .................................... 2-19
KMX Microcode Dump Analyzer (KDA) ............................ 2-19
PSI High Level Language Library (PSIFOR.OLB) .................... 2-19
PSI Macro Library (PSI.MLB) ........................................ 2-20
The Next Step ...................................................... 2-20

RSX-11M-PLUS System Requirements .............................. 3-1
PreSYSGENned RSX-11M-PLUS Systems ........................... 3-1
Hardware CSR Selection ............................................. 3-2

3.1.1.2
3.1.1.3
3.1.1.4
3.1.1.5
3.1.2
3.1.2.1
3.1.2.2
3.1.2.3
3.1.2.4
3.2
3.2.1
3.2.1.1
3.2.1.2
3.2.1.3
3.2.2
3.2.3
3.2.4
3.2.5
3.2.5.1
3.2.5.2
3.2.5.3
3.2.5.4
3.2.5.5
3.2.5.6
3.2.5.7
3.2.5.8
3.2.5.9
3.2.5.10
3.2.5.11
3.2.5.12
3.2.6
3.2.6.1
3.2.6.2
3.2.6.3
3.3
3.3.1
3.3.2
3.3.3
3.3.4
3.3.5
3.3.6
3.3.6.1
3.3.6.2
3.3.7
3.3.8
3.3.9
3.3.10
3.3.11

System Selection ..................................................... 3-2
Customizing the Start-up Procedure ................................. 3-2
Device Reconfiguration ............................................. 3-2
PREGEN ............................................................. 3-3
SYSGENned RSX-IIM-PLUS Systems ............................... 3-3
Including Support for DECnet, PSI, LAT ............................. 3-3
Including support for LA T Terminal Servers ......................... 3-3
RSX and DECnet Communications Devices .......................... 3-3
Queued File Access Requests ........................................ 3-4
RSX-IIM/S System Requirements ................................... 3-4
Determining Partition Layout ....................................... 3-5
SmallRSX-IIM Systems ............................................. 3-8
Medium RSX-IIM Systems .......................................... 3-9
LargeRSX-llMSystems ........................................... 3-10
Pool Usage Considerations ......................................... 3-11
Avoiding DECnet Loading Problems ............................... 3-11
Special Considerations for NMVACP ............................... 3-13
RSX-IIM/-IIS SYSGEN Considerations ........................... 3-14
Including Support for DECnet ...................................... 3-15
RSX and DECnet Communications Devices ........................ 3-15
Standard Function System (RSX-IIM only) ........................ 3-15
Network Command Terminal Support ............................. 3-16
System-controlled Partitions ....................................... 3-16
Checkpointing Support (RSX-11M only) ........................... 3-16
Terminal Driver Support ........................................... 3-17
Queued File Access Requests (RSX-11M only) .......... ~ ........... 3-18
Large or Small Executive ........................................... 3-18
Editing SYSVMR.CMD .............................................. 3-19
Adding CEXPAR .................................................... 3-19
Adding Other Partitions ............................................ 3-21
RSX-llS System Considerations ................................... 3-22
Down-line Loading to a DMP IDMV Tributary ...................... 3-22
SupportforSLD .................................................... 3-22
Including NETPAN (Up-line Panic Dump) .......................... 3-23
PREGEN ............................................................ 3-24
Who Must Perform PREGEN ....................................... 3-24
TwoTypesofPREGEN ............................................. 3-25
Overview of Standard PREGEN Operation ......................... 3-25
Overview of Small Dual-disk PREGEN Operation .................. 3-25
Copying Distribution Media to Disks ............................... 3-26
Starting PREGEN ................................................... 3-27
Disk Distribution Media ............................................ 3-27
Magnetic Tape Distribution Media .................................. 3-28
Understanding the PREGEN Dialog ................................. 3-29
Standard PREGEN .................................................. 3-31
PREGEN for a Small Dual-disk System .............................. 3-34
Sample Standard PREGEN .......................................... 3-36
Sample Small Dual-disk PREGEN ................................... 3-38

Contents-3

Using the RSX-11 M/M-PLUS NETGEN Procedure
4.1
4.2
4.2.1
4.2.2
4.3
4.4
4.4. 1
4.4.2
4.4.3
4.4.4
4.4.5
4.4.6
4.5
4.5.1
4.5.2
4.5.3
4.6

NETGEN Dialog: NET
5. 1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
5.10

Defining the Target and Remote Nodes .............................. 6-1
Defining the DECnet Communications Components ................ 6-8
Defining the DECnet Network Management Components .......... 6-12
Defining the DECnet Satellite Support Components ................ 6-17
Defining the DEC net File Utilities .................................. 6-22
Defining the DEC net Terminal and Control Utilities ............... 6-28

NETGEN Dialog: PSI
7.1

Contents-4

Initialization Questions .............................................. 5-1
Defining the Target System ......................................... 5-11
Defining the System Lines .......................................... 5-16
Defining the CEX System ........................................... 5-28
Defining the Communications Executive Support Components .... 5-31
Defining the System Management Utilities ......................... 5-35
Defining the CEX Products ......................................... 5-39
Completing the CEX System Definitions ........................... 5-40
Building the CEX System ........................................... 5-43
Generation Cleanup ................................................ 5-44

NETGEN Dialog: DEC
6.1
6.2
6.3
6.4
6.5
6.6

NETGENParameterDefaults ..... ,................................... 4-1
StartingNETGEN .................................................... 4-1
Running NETGEN on an RSX-11M/11M-PLUS System .............. 4-2
RunningNETGENonaVAX-11 RSXSystem ........................ 4-2
NETGENOrganization ............................................... 4-3
NETGEN Operating Modes and Saved Response Files ................ 4-5
Question and Answer Mode ......................................... 4-5
Dry Run Mode ....................................................... 4-5
Restore Mode ........................................................ 4-5
Component Mode ................................................... 4-5
Saved Response Files ................................................ 4-6
Duplicating Network Generations .......... '" ...................... 4-6
NETGENDialogFormat ............................................. 4-7
Explanatory Text .................................................... 4-7
Response Formats ................................................... 4-7
Responses for End-of-section and Break Questions .................. 4-9
Abnormal NETGEN Termination ................................... 4-10

Specifying Target Network Characteristics .......................... 7-1

7.2
7.3
7.4
7.5
7.6

Configuring a Pregenned RSX-11 M-PLUS End Node
8.1
8.2
8.2.1
8.2.1.1
8.2.1.2
8.2.2
8.3

Preliminary Considerations ......................................... 8-1
Starting the NETGEN Procedure ..................................... 8-1
NETGEN Dialog Format ............................................. 8-2
Explanatory Text .................................................... 8-2
Response Formats ................................................... 8-2
Responses for End-of-section and Break Questions .................. 8-3
The Pregenned RSX-11M-PLUS Dialog ............................. 8-5

Installing DECnet
9.1
9.1.1
9.1.2
9.2
9.2.1
9.2.2
9.3
9.3.1
9.3.2
9.4
9.4.1
9.4.2
9.4.3

Specifying Local DTEs ............................................... 7-5
Specifying Incoming Call Destinations .............................. 7-9
Specifying Logical Names for Remote DTEs ........................ 7-14
Specifying Required Components .................................. 7-15
Specifying Optional Components .................................. 7-18

Steps Prior to Network Installation .................................. 9-1
Use of the Configuration File Editor (CFE) ........................... 9-2
ExaminationofNETCFE.CMDandNETCFG.TXT ............... , ... 9-2
InstallingDECnetonRSX-11M/M-PLUS Systems ................... 9-3
Modifying NETINS.CMD ............................................ 9-6
Using VMR to Install Network Tasks ................................. 9-8
InstallingDECnet-llSSystems ...................................... 9-8
Installing Tasks Using VMR Commands .............. , ............... 9-9
Loading the Network UsingVNP Commands ....................... 9-10
Installing a DECnet-11 S System on a VAXNMS System ............ 9-11
Establishing VAXNMS Privileges ................................... 9-11
Installing Tasks Using VMR Commands ............................ 9-12
Loading the Network Using VNP Commands ....................... 9-13

Installation Testing
10.1
10.1.1
10.1.2
10.1.3
10.2
10.2.1
10.2.2
10.2.3
10.2.4
10.2.5
10.3
10.3.1

Local Node Software Testing ....................................... 10-4
Preparing to Run the Test .......................................... 10-4
Running the Test ................................................... 10-5
NTEST. CMD Failure Conditions and Handling .................... 10-10
Node Level Hardware Loopback Circuit Testing .................. 10-13
Preparing to Run the Test ......................................... 10-13
Running the Loop Node Test ...................................... 10-14
Restoring the Node ............................................... 10-16
On-Starting Substate Error Handling ............................ " 10-17
Line and Circuit Counter Error Handling .......................... 10-19
Remote Node Software Testing ................................... 10-21
Preparing to Run the Test ......................................... 10-21

Contents-5

10.3.2
10.3.2.1
10.3.2.2
10.3.3
10.3.4
10.3.4.1
10.3.4.2
10.4

Running the Test .................................................. 10-21
Using the NTEST.CMD Procedure ................................ 10-22
DECnet Utility Tests ............................................. , 10-22
Non-Ethernet Error Handling ..................................... 10-24
Ethernet Error Handling .......................................... 10-26
Problems Connecting to the Ethernet ............................ , 10-26
Problems Connecting to the Adjacency ........................... 10-27
Summary .......................................................... 10-27

Testing the RSX-11 PSI Installation
11.1
11.1.1
11.1.2
11.1.3
11.2
11.3

The Installation Checkout Procedure .............................. 11-1
Preparing to Run SCPXTS .......................................... 11-1
Running SCPXTS .............. '" .................................. 11-2
Error Messages ..................................................... 11-2
Verifying X.29 Installation ......................................... 11-4
Verifying Data Link Mapping Installation ........................... 11-4

Sample NETGEN Dialog

How to Include NETPAN (Up-line Panic Dump) in an
RSX-11 S System
B.1
B.1.1
B.1.2
B.1.3

Rebuilding CETAB, CEX, and DDMS
C.1
C.2
C.2.1
C.2.2
C.2.3

Logical Device Assignments ......................................... C-1
Rebuilding Commands .............................................. C-1
NET Components ................................................... C-2
DEC Components ................................................... C-2
PSI Components ..................................................... C-3

DTS/DTR Test Programs
D.1
D.1.1
D.1.2
D.1.3
D.1.4
D.2

Contents-6

How to Include NETPAN in an RSX-11S System ..................... B-1
Preparing for SYSGEN ............................................... B-2
SYSGEN Phase I Considerations ................................... " B-2
SYSGEN Phase II Considerations .................................... B-3

Types of Tests ....................................................... D-1
Connect Tests ....................................................... D-1
Data Tests ........................................................... D-2
Disconnect Tests .................................................... D-2
Interrupt Tests ...................................................... D-2
Operational Characteristics ......................................... D-3

D.3
D.3.1
D.3.2
D.3.3
D.3.4
D.3.5

DTS Command Syntax .............................................. D-4
Connect Test ........................................................ D-5
Disconnect Test ..................................................... D-5
Data Test ............................................................ D-5
Interrupt Test ....................................................... D-6
DTS/DTR Error Messages ............................................ D-7

9-1
10-1

Typical NETINS.CMD for an RSX-IIS System ....................... 9-9
NTEST. CMD Sample Listing ........................................ 10-6

2-1
2-2
3-1
3-2
3-3
3-4
4-1
5-1
5-2
5-3
5-4
5-5
5-6
5-7
6-1
6-2

A Sample Ethernet Configuration .................................... 2-6
A Multiarea Network ................................................. 2-8
Partition Layout after SYSGEN (CEXPAR added) ..................... 3-6
Partition Layout for Small RSX-IIM Systems ........................ 3-9
Partition Layout for Medium RSX-IIM Systems .................... 3-10
Typical Partition Layout ........................................... 3-12
Sections Within the NETGEN Process ............................... 4-4
NET - Section 1 - General Initialization ............................. 5-2
NET - Section 2 - Define the Target System ........................ 5-11
NET - Section 3 - Define the System Lines ......................... 5-17
NET - Section 4 - Define the CEX System .......................... 5-28
NET - Section 5 - Define the Comm Exec Support Components ... 5-31
NET - Section 6 - Define the System Management Utilities ......... 5-35
NET - Section 8 - Complete the CEX System Definitions ........... 5-40
DEC - Section 1 - Define the Target and Remote Nodes ............. 6-2
DEC - Section 2 - Define the DECnet Communications
Components ......................................................... 6-8
DEC - Section 3 - Define the DECnet Network Management
Components ....................................................... 6-13
DEC - Section 4 - Define the DECnet Satellite Support
Components ....................................................... 6-17
DEC - Section 5 - Define the DECnet File Utilities .................. 6-22
DEC - Section 6 - Define the DECnet Terminal and Control
Utilities ............................................................. 6-28
PSI - Section 1 - Specify Target Network Characteristics ............ 7-2
PSI - Section 2 - Specify Local DTEs ................................. 7-5
PSI - Section 3 - Specify Incoming Call Destinations ............... 7-10
PSI - Section 5 - Specify Required Components .................... 7-15
PSI - Section 6 - Specify Optional Components .................... 7-18
Installation Procedure for an RSX-IIM/IIM-PLUS System ......... 9-4
Installation Test Flowchart. ........................................ 10-2
Local Node Software Test .......................................... 10-3
Node Level Hardware Loopback Circuit Test ..................... 10-12

Examples

Figures

6-3
6-4
6-5
6-6
7 -1
7- 2
7 -3
7 -4
7 -5
9-1
10-1
10-2
10-3

Contents-7

10-4
10-5
10-6

Remote Node Software Test ...................................... 10-20
Node Level Loopback Tests Using DEC net-supplied Software .... 10-29
Node Level Loopback Tests Using User-supplied Software ........ 10-30

1-1
1-2
2-1
2-2
2-3
3-1
3-2
3-3
4-1
8-1
10-1
B-1

SummaryofNETGENOutputFiles .................................. 1-3
Reading Paths ........................................................ 1-6
Target Disk Capacity'" ............................................... 2-2
Supported Communications Devices ................................ 2-3
Component Selection Summary for NETGEN ...................... 2-12
Approximate Sizes of Key Components .............................. 3-7
CEXPAR Base Addresses and Sizes .................................. 3-20
PREGEN Files ....................................................... 3-24
SummaryofNETGENResponses .................................... 4-8
SummaryofNETGENResponses .................................... 8-3
Utility Availability onDECnet Nodes ............................. 10-22
Reboot Addresses for the M9312 ROM ............................... B-4

Tables

COrJtents-8

Preface

This manual tells you how to generate, install, and test DECnet-llMVersion 4.2,
DECnet-llM-PLUS Version 3.0, and DECnet-llS Version 4.2
This manual is for both DECnet-only users and DECnet users w:ho purchase a license
for the RSX-ll Packetnet System Interface (PSI) product. The PSI product allows your
DECnet-RSX node to access a Packet Switching Data Network (PSDN) using the X.25
and X.29 protocols.
Intended Audience

This manual is intended for an advanced user or for a Digital software specialist responsible for generating and installing a DECnet-RSX node on an RSX-IIM, RSX-IIMPLUS, or RSX-llS system. The DECnet-MicrolRSX Installation Guide, explains how
to install the network software for a DECnet-Micro/RSX node.
Structure of This Manual

This manual is organized into eleven chapters:
Chapter 1

Outlines the tasks you must perform before, during, and after a network
generation (NETGEN).

Chapter 2

Tells you how to configure your RSX system to support DECnet and contains information that will help you select the appropriate devices and
DECnet components.

Chapter 3

Discusses the steps that must be taken to prepare your RSX-I1M/MPLUS system for NETGEN. These steps include specifying partition layout, performing a system generation (SYSGEN), and performing a preliminary kit merge procedure called PREGEN.

Preface-1

Chapter 4

Tells you how to use the DECnet RSX-IIM/M-PLUS NETGEN procedure to generate the network software.

Chapter 5

Presents the first part of the RSX-IIM/M-PLUS NETGEN dialog (called
NET).

Chapter 6

Presents the second part of the RSX-IIM/M-PLUS NETGEN dialog
(called DEC).

Chapter 7

Presents the third part of the RSX-IIM/M-PLUS NETGEN dialog (called
PSI), for users who have purchased a license for the RSX-l1 PSI product.

Chapter 8

Provides information needed to configure a pregenned RSX-IIM-PLUS
end node.

Chapter 9

Tells you how to install and load DEC net on an RSX-I1M, RSX-I1MPLUS, or RSX-lIS system.

Chapter 10

Describes the test procedures that should be performed to check the
operation of the network software after installation.

Chapter 11

Describes the test procedures that should be performed to check the
operation of PSI software after a PSI installation.

This manual also contains four appendixes:
Appendix A

Presents a sample NETGEN dialog.

AppendixB

Tells you how to include NETPAN (Network Up-line panic dump) capability for an RSX-IIS system.

AppendixC

Tells you how to rebuild components that cannot be rebuilt using component mode (CETAB, CEX, DDMs).

Appendix D

Describes the DECnet test sender (DTS) and the DECnet test receiver
(DTR) utilities that are used for the network software testing described
in Chapter 10.

Associated Documents

Users of this manual should have the following related manuals of the document set:
•

Introduction to DECnet

•

DECnet-RSX Guide to User Utilities

•

DECnet-RSX Programmer's Reference Manual

•

DECnet-RSX Network Management Concepts and Procedures

•

DECnet-RSX Guide to Network Management Utilities

Preface-2

Users of this manual should have the following manuals available for system generation
(SYSGEN) procedures:

•

RSX-11M System Generation and Installation Guide

•

RSX-IIM-PLUS System Generation and Installation Guide

•

RSX-11 S System Generation and Installation Guide

Users who have purchased a license for the PSI product should also have the PSI product manual set available for reference.
Acronyms

The following acronyms are used in this manual:

AUX

Auxiliary process

CCR

Console carrier requester

CEX

Communications Executive

CFE

Configuration File Editor

DCP

Digital Communications Process

DDM

Device driver module

Dll

Down-line system loader

DlM

Data link mapping process

DlX

Direct Line Access process

DUK

KMX microcode dumper

DUM

Up-line system dumper

ECl

End Communication layer

EPM

EthernetProtocol~anager

EVC

Event collector process

EVF

Event File Interpreter utility

EVl

Event listener process

EVR

Event logging receiver process

FAl

File Access Listener

FTS

File Transfer Spooler

HlD

Host Task Loader

KDA

KMX microcode dump analyzer

LAB

LAP-B protocol process

LAT

Local Area Transport
Preface-3

LCP

LAT Control Program utility

LIN

Link Watcher

LOO

Loopback tester

LSN

Listen utility

MCM

Command file submission task

MIR

Loopback mirror

MLD

General Microcode Loader

NCP

Network Control Program

NCT

Network control terminal task

NDA

Network Crash Dump Analyzer

NETACP

Network ancillary control processor

NETFOR.OLB

DEC net high level language library

NETLlB.MLB

DECnet MACRO-11 library

NFT

Network File Transfer utility

NICE

Network Information and Control Exchange

NMDRV

Network management device driver

NMVACP

Network management volatile Ancillary Control P~ocessor

NTD

Network Display utility

NTDEMO

Network Display Server

NTINIT

Network initializer

NTL

Network loader

NVP

Network Verification Program

NW: device driver

PHO

Phone communications utility

PLI

Packet level interface

PSDN

Packet Switching Data Network

PSI

Packetnet System Interface

PSIFOR.OLB

PSI high level language library

PSI.MLB

PSI MACRO library

RCP

Routing control process

RMHACP/HTDRV Remote terminal host Ancillary Control Processor/HT: device driver

Preface-4

RMT/RMTACP

Remote Terminal utility/Remote Terminal Ancillary Control Processor

RTH

Remote command terminal host process

SlD

Satellite task loader

TCl

Remote Task Control utility

TlK

Terminal communications utility

TRA

Trace capture task

TRI

Trace interpreter task

VNP

Virtual Network Processor

XPT

Routing layer process

X25ACP

Ancillary control processor for NW

X29ACP

X.29 Terminal Access Ancillary Control Processor with HTDRV

Conventions Used in This Document

The following graphic conventions are used in this manual:
Convention

Meaning

Monospace
type

Monospaced type is used in examples to indicate user input (in red ink) and system output (in black ink).

UPPERCASE

Uppercase letters in command formats indicate actual values to be entered (you
can enter them in either uppercase or lowercase).

italic

Italic type in command formats and system displays indicates a variable for
which either you or the system must provide a value.

KEY

Indicates that you should press the specified key. CTRLlx indicates that you
should press the CTRl key at the same time as the key specified by x. Unless otherwise specified, every command line is terminated by pressing the < RET>
key.

All numbers are decimal, unless otherwise noted.
NOTE

Generally you can abbreviate command keywords to the first three
characters or the number of characters that make the keyword unique.

Preface-5

1
Introduction

This chapter introduces you to the steps you must perform before, during, and after network generation (NETGEN). These steps include making preliminary planning decisions, performing a SYSGEN (if necessary), performing a PREGEN (if necessary), performing a NETGEN, and understanding post-NETGEN considerations.
SYSGEN is a procedure that establishes the RSX operating system environment.
PREGEN is a pre-NETGEN procedure that converts the distribution media (disk or magnetic tape) into the form required by NETGEN. NETGEN is an indirect command file
procedure that generates the software for a DECnet node.
The steps involved in NETGEN preparation and the NETGEN itself vary depending on
the type of distribution kit you received. A reading path through this manual for users
of each kit is given at the end of this chapter.

1.1

Making Preliminary Decisions

The content of your kit reflects a basic decision that you have already made. You have
chosen to build
•

a full function DECnet node that can route messages from one node to another

•

a configurable end node that can communicate with other nodes but which cannot
route messages from one node to another

•

a pregenned end node that is, to a great extent, pre-defined

Before performing a NETGEN, you must make some preliminary decisions, as follows:
•

Decide what communications device(s) you want for your target system (the system on which the DECnet node will run).
1-1

•

Decide what optional DECnet components (as described in Chapter 2, Table 2-3),
you want for your target system.

•

Decide what library you want to use if your target system supports the memory-resident FCS library or the supervisor mode FCS library.

You may also need to specify the partition layout for your system. On RSX-11M-PLUS
and large RSX-11M systems, the partition layout supplied by your RSX system generation usually proves adequate. For other systems, you may need to modify the layout.
Section 3.2 explains this procedure.

1.2

Performing a SYSGEN

If you have not already generated and/or installed your target system, you must first do

so (see Sections 3.1 and 3.2). Even if you have previously performed a SYSGEN for the
node, it may be necessary to perform the SYSGEN again to build support services for
DECnet. An RSX-11M-PLUS SYSGEN by the user is never required for DECnet RSX11M-PLUS pregenned end nodes.

1.3

Performing a PREGEN

The PREGEN procedure (see Section 3.3) converts the distribution media (disk or magnetic tape) into a form that is directly usable by NETGEN. You may have one to three
distribution media depending on the type of kit you ordered and whether or not you
have purchased a license for the PSI product. If you have more than one distribution
medium, PREGEN selects certain parts of each and copies them to the working generation disk used by NETGEN. This minimizes or eliminates the swapping of disks later. A
PREGEN is never required for RSX-IIM-PLUS pregenned end nodes.

1.4 Performing a NETGEN
To perform a NETGEN, you engage in a dialog with the NETGEN procedure, basing
your responses to displayed questions on information developed during the preliminary, decision-making phase. The NETGEN procedure creates a version of DECnetRSX that is tailored to your needs. The NETGEN procedure generates output that is
used to install the network. The type of NETGEN you use depends on the type of kit
you purchased. The NETGEN procedure for RSX-IIM/S and RSX-IIM-PLUS nodes is
described in Chapters 4, 5,6, and 7. A shorter NETGEN procedure for RSX-I1M-PLUS
pregenned nodes is described in Chapter 8.
Table 1-1 lists the disk areas where the files generated by NETGEN reside on the target
disk. Group code x refers to the group code for NETGEN output that you specify in
NETGEN. The dduu: for listing and map files may represent the NL: device or any disk
in the system. You define the devices in NET, Section 1, Question 3.00. For
preSYSGENned RSX-IIM-PLUS systems, no map or listing files are produced. A
description of these NETGEN files is provided in the subsections that follow.

1-2

DECnet-RSX Network Generation and Installation Guide

Table 1-1:

Summary of NETGEN Output Files
UIC on the Target Disk
DECnet-11 M

1M-PLUS

NETLIB.MLB (MACRO library)
NETFOR.OLB (High-Level Language library)

[1,1]

NETHLP.ULB (help file for NCPNNP/CFE/NFT)

[1,2]

[x, 1]

[x,l]

NETGEN's copy of RSXMC.MAC

[x,10]*

[x, 10]

Object files, assembly and task build
command files

[x,24]*

[x,24]

Listing and map files

dduu:[x,34] *

dduu:[x,34]

Tasks and processes

[x,54]

[x,64]

Files
DECnet-11S
Libraries

Command files and work files
NETINS.CMD
NETREM.CMD (not for RSX-llS)
NETCFE.CMD
NETCFG.TXT
Saved response files
Prefix files (used to specify conditional
assembly parameters)

CETAB.MAC (permanent database)
NETGEN's copy ofRSXIIM.STB or RSX11S.STB

... Not in pregenerated RSX-IIM-PLUS end node systems

1.4.1

NETINS.CMD

The Network Installation command file is the file invoked to install DECnet after performing a NETGEN. Chapter 9 describes this command file and how it is used.

1.4.2 NETREM.CMD
The Network Remove command file is the file invoked when you want to turn the network off and remove network tasks. NETREM.CMD is not generated for RSX-llS
nodes.

Introduction

1-3

NOTE
The NETREM.CMD procedure will unload the HT: device but cannot
unload the HT: device's data structures. Any subsequent reloads are
restricted to networks with the same number ofHT: devices. In order
to load a network with a different number of HT: devices, you must
reboot the operating system.

1.4.3

NETCFE.CMD

The Network CFE command file is an indirect command file that can be used as input to
the Configuration File Editor (CFE). This file provides a record of the permanent
database configuration (CETAB.MAC) as created by NETGEN. It can also be used as a
template for making changes to the permanent database. In this case, you edit the command file to reflect the appropriate changes and then use the file as an indirect command file to CFE.

1.4.4 NETCFG. TXT
The network configuration file provides information in text form on the configuration
of devices, processes, and CEX products on the target system.

1.4.5 CETAB.MAC
The CETAB.MAC file contains the permanent database used to load the configuration
defined for the target node. It is one of the most important outputs of NETG EN.

1.5

Post-NETGEN Considerations

Post-NETGEN considerations include installing DECnet, testing the newly installed network, reconfiguring and tuning the network, and regenerating as required (see Chapter
4).

1.5.1

Installing DECnet

To install DECnet, you execute the NETINS.CMD file. In many cases, this indirect command file is all that is necessary to bring up the node. For a more complete description
of installing DECnet and b~inging up the node, including editing the NETINS.CMD file,
see Chapter 9. For a detailed explanation of the NCP commands necessary to bring up a
node, see the DECnet-RSX Network Management Concepts and Procedures.

1.5.2 Testing a Newly Installed Network
The newly installed network should be tested by running the tests described in Chapters 10 and 11.

1-4

DECnet-RSX Network Generation and Installation Guide

Chapter 10 describes procedures for determining whether or not
•

node software is operational

•

node and communications hardware are functioning properly

•

node is able to communicate with a remote node

Chapter 11 is for users ofRSX-11 PSI. It describes procedures for
•

verifying whether or not RSX PSI software was properly generated and' installed

•

verifying that X. 29 software was installed correctly

•

testing data link mapping (DLM) circuits

1.5.3 Reconfiguring and Tuning after NETGEN
After the DEC net node is up and running, you can add features or change parameters to
improve performance. On RSX-IIM/M-PLUS nodes you have three options: you can
perform a new NETGEN; you can perform a partial NETGEN; or you can use one of the
network utilities described in this section. On pregenned RSX-IIM-PLUS end nodes,
you must use the network utilities or perform a new NETGEN.
For example, to add a line to RSX-IIM/M-PLUS nodes, you must perform a new
NETGEN. To add or change an individual component (such as TLK) on RSX-I1M/MPLUS nodes you can perform a partial NETGEN. You can modify many network parameters on all types of nodes using CFE commands, NCP commands, or VNP commands.
A short description of these utilities is given in the next three sections. For more information on using these utility commands, and on tuning the network, see the DECnetRSX Guide to Network Management Utilities.

1.5.3.1 Configuration File Editor (CFE) Commands - CFE is used to change the
permanent database of the network; it allows you to edit the configuration file
CETAB.MAC. This file retains the master copy ofa wide variety of network parameters
and is read each time the network is loaded. Changes made using CFE do not take effect
until the network is reloaded. CFE can change many parameters that NCP (as described
in the following) cannot change.
1.5.3.2 Network Control Program (NCP) Commands - You can modify a wide
range of network parameters using NCP commands. These commands affect the running system (the volatile database) only. The next time the network is loaded (for example, after using the NCP command SET SYSTEM), all parameters assume their previous
values as recorded in the permanent database.
1.5.3.3 Virtual Network Processor (VNP) Commands - VNP supports a subset of
NCP commands that allows you to change the system image file for RSX-ll systems in
much the same way that VMR can be used to perform MCR operations on the system
image. As with VMR, any changes you make will not become effective until the next

Introduction

1-5

time the system is booted. VNP is especially useful for RSX-llS systems because the
RSX-llS NCP utility is a limited subset of the full NCP utility.

1.6

Reading Paths

The reading path that you should take through the remainder of this manual depends
on the target operating system, the type of kit you are using, and the distribution media
on which you received it. Reading paths based on kit type are shown in Table 1-2. The
type of media with which you will be working influences your reading path within
Chapter 3 and is not shown in Table 1-2.

Table 1-2:
Kit Type

Reading Paths
Target System
RSX-IIM/S

DECnetnode
DECnetnode
w/PSI
Pregenned
DEC net end
node

1-6

RSX-IIM-PLUS

2,3,4,5

6,9,10

2,3,4,5

6,7,9,10,11

6, 7, 9, 10, 11
2,8,9,10

DECnet-RSX Network Generation and Installation Guide

2
Planning DECnet

This chapter tells you how to configure your RSX system to support DECnet. It also
gives you some guidelines on configuring your network and selecting components for
incorporation into your DECnet node.

2.1

RSX-11 Configuration Requirements to Support DECnet-RSX
Software

An RSX system that supports DECnet must consist of a certain minimum software/hardware configuration. In addition, the requirements for the host system (the system on
which you are performing NETGEN) may differ from the requirements for the target
system (the system on which the DECnet node will run). In many cases, the same system serves in both capacities.

2.1.1

Requirements for the Host System

The host system can be an RSX-IIM Version 4.2, an RSX-IIM-PLUS Version 3.0, or a
VMS system running VAX-II RSX Version 2.0.
To generate the software for the target node, the host system must provide a certain
minimum amount of disk space. One RLOI disk drive (or larger), in addition to the disk
used for the operating system, satisfies this minimum. With a large disk (such as an
RP05 or RM80), one disk is usually sufficient to store the operating system and all files
necessary to perform NETGEN. See Section 3.3 for PREGEN requirements.
The host system must also provide sufficient target disk capacity for tasks, maps, and
listings as indicated in Table 2-1.

2-1

Table 2-1:

Target Disk Capacity *

Type of System

Tasks

Maps and Listings

DEC net/PSI

4,500 blocks

3,000 blocks

DECnet-only
routing or end node

4,000 blocks

2,900 blocks'" ...

Pregenned
RSX-I1M-PLUS
end node

4,000 blocks

... If you will not be generating all tasks, and if maps and listings are stored on separate
disks, the required capacity of the target disk is reduced correspondingly .
...... About 1,300 for RSX-llM-PLUS.

2.1.2

Requirements for the Target System

The target system must be RSX-11M Version 4.2; RSX-IIM-PLUS Version 3.0; RSXlIS Version 4.2.
The target system must be a mapped system with a minimum of 16K words of memory
dedicated to the DECnet software. Additional memory is required for user-written network tasks and any DECnet utilities.
The target system must have at least one communications device not in use by the Executive to satisfy the minimum hardware requirements. If you have an RSX-IIM-PLUS
system, see section 3.1.1.4 for instructions on reconfiguring system devices so that
they can be used by DECnet. Table 2-2 lists the communication devices supported for
DECnet and PSI.

2-2

DECnet-RSX Network Generation and Installation Guide

Table 2-2:

Supported Communications Devices

SUPPORTEDFORDECn~

UNIBUS

Q-BUS

Asynchronous Line Support

DLll
DZ11
DHU11
KMC111DZ11

DLV11
DZV11 or DZQ11
DHV11

Synchronous Line Support

DMP11
DUP11
DUll
DMC11
DMR11
DVll
KMC111DUPll

DMVll
DPVll
DUV11

Parallel Line Support

PCLll
Ethernet Line Support

DEUNA

DEQNA

SUPPORTED FOR PSI
UNIBUS

Q-BUS

DUP11
KMS11-BD

DPV11
KMS11-PX

2.2

Network Configuration Guidelines

In generating your node, several guidelines must be observed to assure proper network
routing operation. These guidelines fall into one of three categories:
•

General guidelines - for all network nodes (Section 2.2. 1)

•

Ethernet guidelines - for all Ethernet nodes (Section 2.2.2)

•

Area guidelines - for all nodes in a multiarea network (Section 2.2.3)

Planning DECnet

2-3

Most of the guidelines that are presented in the following three sections are mandatory
rules that must be followed to configure a network that will function correctly. Other
guidelines are included that, while not mandatory, will aid in creating a more useful network.

2.2.1

General Guidelines

The following guidelines apply to all Phase IV nodes:
•

All nodes in the network must have a unique networkwide address. If two
nodes have the same address and the path costs to the nodes are different, messages
will always be sent to the node with the lower path cost. If the two nodes have the
same path cost, the result is not predictable. In both cases, any node can receive
messages from the identically addressed nodes but can send messages to only one
of them.
The text accompanying question 02.00 in Section 1 of the DEC portion of the network generation procedure tells you how to specify a node address. NETGEN also
gives you the opportunity to assign a node name and a node ID to the node being
generated. See questions 01.00 and 03.00 in Section 1 of the DEC portion of the
generation procedure.

•

All routing nodes within an area should have the same value specified for the
highest node number parameter. For Phase III nodes this may not be possible due
to the expanded capability of Phase IV. For more information on areas and Phase
III restrictions, see Section 2.2.3. If this guideline is not followed, a node with a
greater value specified for the highest node number parameter would be able to
send a routing message that a node with a lesser value for this parameter could not
handle. When the routing node with the lesser value receives the routing message,
a partial routing update loss occurs. Nodes whose routing information was in the
part of the routing message that was lost are unreachable.
The highest node number parameter is established by your response to Question
05.00 in Section 1 of the DEC portion of NET GEN.

•

2-4

All nodes should use the same large buffer size. The value specified for the large
buffer size can create a situation in which a message segment cannot be forwarded.
If, for example, the large buffer size specified for an intermediate node is smaller
than the segment size agreed upon by the source and destination nodes, the intermediate node will not be able to forward it. For this reason, all routing nodes in the
network must have a large buffer size at least as large as the largest segment size in
use in the network. This guideline is not mandatory for end nodes because they
will never act as intermediate nodes. However, end nodes should never use a large
buffer size that is greater than that used by the routing nodes in the network.

DECnet-RSX Network Generation and Installation Guide

NOTE
On DECnet-RSX you must add the value 18 (the size of the routing
layer and end communications layer protocol overhead) to the segment size value when setting the large buffer size. The NETGEN
procedure and CFE will check this automatically.
More information about determining the large buffer size and message segment size
is given in the DECnet-RSX Network Management Concepts and Procedures manual. You specify the size of the large data buffer in response to question 02.00 in
Section 8 of the NET portion of the network generation procedure.

•

All nodes should reference other nodes by the same name. If, for example, a
node is known as BOSTON by one node, all other nodes should also refer to it as
BOSTON. Although the concept of node names is a local one, you should make
every effort to use a common node name list on all nodes. As networks increase in
size, the bookkeeping associated with network maintenance is simplified if all
nodes reference remote nodes by the same logical name. A command file is provided to aid you in establishing a common set of node names (see the description of
NETNODES.CMD in the DECnet-RSX Network Management Concepts and Procedures manual).
Remote node names are specified in response to Question 07.00 in Section 1 of the
DEC portion of NET GEN.

2.2.2

Ethernet Configuration Guidelines

When generating a routing node attached to one or more Ethernet circuits, you must
follow a set of guidelines to ensure proper node operation. These guidelines are listed
below. The DECnet-RSX Network Management Concepts and Procedures manual
explains how to set these parameters using the Configuration File Editor (CFE).

•

The total number of Ethernet end nodes with which the node will communicate, must not exceed the value specified for the Maximum Broadcast
Nonrouters parameter. For example, in the network shown in Figure 2-1, suppose that node FINANC has the parameter set to 20 and node PURCHS has it set to
10. In this network, the total number of end nodes on Ethernets 1 and 2 cannot
exceed 20 (FINANC's value), and the total number of end nodes on Ethernets 2 and
3 cannot exceed 10 (PURCHS's value). If this guideline is ignored, the end nodes
that would be reachable at any given time would be unpredictable.

•

The total number of routers on all Ethernet circuits of a node cannot exceed
the Maximum Broadcast Routers parameter for the node. This value should be
equal to the sum of all circuit Maximum Router values for the node (see next guideline). On DECnet-RSX nodes, the Maximum Broadcast Routers parameter is automatically calculated during NETGEN and set to the sum of all the circuit Maximum
Router values specified.

Planning DECnet

2-5

•

The number of routers on any single Ethernet circuit cannot exceed the value
specified for the Maximum Routers parameter for that circuit. In Figure 2-1, if
node FINANe had the Maximum Routers parameter for Ethernet 1 set to 2, the number of routers allowed on Ethernet 1 would be limited to 2, in addition to the executor node. The sum of all individual circuit Maximum Routers parameter values
should not exceed the Maximum Broadcast Routers parameter for a node (see preceding guideline).

•

The number of routers on a single Ethernet should be kept to a minimum. Routing nodes generate traffic among themselves (especially when nodes are in the process of becoming unreachable). To keep this traffic to a minimum, the number of
routers on a single Ethernet should not exceed 10.

ETHERNET 1

NODE
FINANC

ETHERNET 2

NODE
PURCHS

ETHERNET 3

TW0224

Figure 2-1:

2-6

A Sample Ethernet Configuration

DECnet-RSX Network Generation and Installation Guide

2.2.3

Areas and Area Configuration Guidelines

DECnet-RSX Phase IV implementations can support networks consisting of many
more nodes than Phase III implementations could support. Networks with thousands
of nodes can now be supported using the concept of areas. If your network consists of
more than 1023 nodes you must use the area network features of Phase IV. Even if your
network consists of fewer than 1023 nodes, you can, in many cases, benefit by splitting
your network into areas. A network containing three areas is shown in Figure 2-2.

Planning DECnet

2-7

•

Level 2 Routing Node

Level 1 Routing Node

End Node

TW0297

Figure 2-2:

2-8

A Multiarea Network

DECnet-RSX Network Generation and Installation Guide

In designing a network that implements the area concept, the network manager partitions the network into smaller networks called areas. Partitioning can be based on physical proximity, commonality of function, or both. If you choose to partition your network into areas, you must view your entire network on two levels. Level 1 is
equivalent in many ways to older Phase III/Phase IV networks. Levell routing nodes in
Phase IV networks with area support, however, must maintain a routing database only
the size of the area of which they are a member. All nodes in the same area have the
same area parameter value.
In a Phase IV networks consisting of more than one area, at least one of the nodes in
each area is also a member of the level 2 network. This node (called a level 2 routing or
area routing node) must maintain two routing databases; one for the local level 1 network and one for the networkwide level 2 network. Each node in the level 2 network is
a member of the networkwide level 2 network and one of the level 1 networks.
Communication between nodes in the same area occurs exactly as it occurred in Phase
III. When the source and destination nodes are in the same area, the level 1 routing
nodes forward the request using their local routing databases.
When access to a node outside a local area is desired, the source node sends the request
to the nearest level 2 router. The level 2 routing node will route the packet to a level 2
router at the correct destination area. The packet is then routed from the level 2 router
to the destination node in that area.
In designing a network that supports area routing, several guidelines must be followed
to ensure proper network operation.
•

Every level 1 routing node must be in one area only. It is illegal for a level 1
router to have a circuit connected to a node outside its area.

•

Each area (level 1 network) must be physically intact. All nodes within an area
must be connected in some way to all other nodes within the same area. The nodes
need not be physically adjacent; however, all nodes in the area must be connected
by a physical path that lies totally within the area. This path can involve level 2 routing nodes in the same area.

•

The level 2 network must be physically intact. All level 2 routing nodes must be
connected in some way to all other level 2 nodes. This connection must involve
level 2 nodes only. Level 1 nodes or Phase III nodes cannot form a part of the path
between level 2 routing nodes.

•

All Phase III routing nodes are treated as level 1 routing nodes, with some
important restrictions:
•

All Phase III nodes (both routing nodes and end nodes) can communicate only
with nodes within the same area as themselves.

•

All Phase III routing nodes can support communication only with nodes for
which they can maintain routing information. Phase IV DECnet networks can
have more nodes than can Phase III nodes (Phase Ill's limit was 255). For this

Planning DECnet

2-9

reason, any Phase IV node with an address above 255 is "invisible" to all Phase
III nodes in the area. This is a good reason to configure a network into areas,
each of which does not consist of more than 255 nodes.

•

A Phase III node cannot be in the path of two high address Phase IV nodes.
Because of the Phase III address limitation, a Phase III node must not be in the
path of two Phase IV nodes with addresses higher than the Phase III node can
handle (255).

•

A Phase III node must not be in the path of any interarea communication.
Because Phase III nodes do not know about areas, they cannot handle interarea
packets correctly.

Areas should reflect expected traffic flow. Observing this guideline will greatly

improve overall network performance. Routing within an area is less costly than is
routing between areas. Try, therefore, to include within the same area nodes that
are logically or functionally related. Area bounds should reflect a combination of
geographical and logical node location.
•

Areas should not be used to enforce protection. Although at first glance the con-

cept of areas seems to offer penetration protection across areas, the DNA architecture does not specify protection guarantees for areas. No assumption about such
guarantees should be made when setting up areas. Areas were not designed to be
used as an accounting or security tool.
Additional information about areas and area parameters can be found in the DECnetRSX Network Management Concepts and Procedures manual.

2.2.4

Guidelines for Specifying Routing Parameters

If you are generating a routing node, you must specify certain routing parameters in

response to questions that NETGEN displays. The values of some other routing parameters are calculated for you by the NETGEN process.
Routing parameter values that you must specify are conditioned by whether your network consists of a single area or multiple areas.
•

If your network consists of a single area, you must specify the highest node number

in the network.
•

If your network consists of multiple areas, you must specify the highest node num-

ber in the area in which the node that you are generating will be located.
•

If you are generating a level 2 routing node, you must specify the maximum area
parameter.

The highest node number determines the size of the routing database. Any node with a
number exceeding the highest node number value is unreachable. In networks without
area support, the highest node number is for the entire network; in networks with area
support, this value is for the area.

2-10

DECnet-RSX Network Generation and Installation Guide

The maximum area parameter defines the size of the node's level 2 routing database.
Routing parameters calculated for you by NETGEN are:
•

Maximum hop count

•

Maximum path cost

•

Line/circuit cost

•

Maximum area hop count

•

Maximum area path cost

NETGEN assigns a maximum hop count of 10, a maximum path cost of 1022, and calculates a line/circuit cost based on device type and speed. For level 2 routing nodes,
NETGEN assigns a maximum area hop count of 10 and a maximum area path cost of
1022.
You can change these values using CFE, NCP, and VNP commands. For more information on routing parameters and how to change them, see the DECnet-RSX Network
Management Concepts and Procedures manual.

2.3

Selecting DECnet Components

This section presents information that will help you select optional components that
perfornl DECnet and PSI functions. Certain components are always built even though
they may not be required. For a description of all required components, refer to the
explanatory text accompanying the NETGEN question descriptions in Chapters 5, 6, 7,
andS.
Table 2-3 identifies the individual components that you can select for your system.
The table indicates
•

components that are required (RQD)

•

components that are always built (A/B) but which are not required

•

components that are optional (OPT)

•

components that are never used (N/U)

Notes are provided to clarify exceptions for some components.
RQD and N/U components do not require you to make a decision. However, you do
need to decide whether or not you want to select certain optional components, or use
components that are always built but not required. The information in this chapter
should help you make these decisions.

Planning DEenet

2-11

Table 2-3:

Component Selection Summary for NETGEN

Component

System
-11M
-11M+

-11S

PreGEN

Communications Executive Components
Communications Executive (CEX)
Auxiliary Process (AUX)
Device Driver Modules (DDM)

RQD
RQD
RQD

Note 1
RQD
RQD

RQD
RQD
RQD

Note 1
RQD
RQD

RQD
RQD
Note 2
OPT
RQD

RQD

RQD
RQD
Note 2

RQD
RQD
RQD

OPT
Note 4

AlB
AlB
AlB
AlB

OPT

N/U
AlB
AlB
AlB
N/U
N/U

RQD
RQD
RQD
Note 6
Note 7
OPT

NET Communications Executive Support Components
RQD
Network Initializer (NTINIT)
RQD
Network Loader (NTL)
General Microcode Loader (MLD)
Note 2
Network Event Logger (EVL/EVC)
OPT
Direct Line Access Controller (DLX)
Note 3
NET System Management Utilities
Network Control Program (NCP)
Network Management Device Driver (NMDRV)
Network Management Volatile ACP (NMVACP)
Configuration File Editor (CFE)
Network Crash Dump Analyzer (NDA)
Virtual Network Processor (VNP)
Loopback Tester (LOO)
Event File Interpreter (EVF)
DECnet Communications Components
End Communication Layer driver (ECL)
Routing Layer process (XPT)
NetworkACP (NETACP)
Digital Communications Process (DCP)
Ethernet Protocol Manager (EPM)
NETFOR.OLB and NETLIB.MLB

N/U
Note 2
OPT
Note 3

AlB
RQD

RQD
RQD
RQD

AlB
Note 5

AlB
AlB
OPT

RQD
RQD
RQD
Note 6
Note 7

AlB

(continued on next page)

2-12

DECnet-RSX Network Generation and Installation Guide

Table 2-3 (cont.):

Component Selection Summary for NETGEN

Component
DECnet Network Management Support Components
Network Information and Control Exchange (NICE)
Event Logging Receiver (EVR)
Network Display utility (NTD)
Network Display server (NTDEMO)
Link Watcher (LIN)
Loopback Mirror (MIR)
Routing Control Processor (RCP)
Network Verification Program (NVP)
DECnet Satellite Support Components
Down-line System Loader (DLL)
Up-line System Dumper (DUM)
Console Carrier Requester (CCR)
Host Task Loader (HLD)
Satellite Task Loader (SLD)
DECnet File Utilities
Network File Transfer utility (NFT)
File Transfer Spooler (FTS)
File Access Listener (FAL)
Command File/Batch File Submission task (MCM)
DECnet Network Terminal and Control Utilities
Remote Terminal utility RMTACP (RMTIRMTACP)
Remote Terminal Host ACP HTDRV
(RMHACP IHTDRV)
Network Control Terminal task (NCT)
Remote Command Terminal Host (RTH)
Terminal Communications utility (TLK)
L~sten utility (LSN)
Phone communications utility (PHO)
Remote Task Control utility (TCL)
Required PSI Communications Components
LAP-B Protocol process (LAB)
Packet Level Interface (PLI)
NW: device driver (NW)
Ancillary Control Processor for NW (X25ACP)
PSI Device Driver Modules (PSI DDM)

System
-11M
-11M+

-11S

PreGEN

OPT
OPT
OPT
OPT

AlB

AlB
AlB

N/U
AlB

Note 8
OPT

Note 8

N/U

AlB
AlB
AlB
AlB
AlB
AlB
N/U
AlB

OPT
OPT
OPT
OPT

N/U
N/U
N/U
N/U

OPT
OPT
OPT
OPT

N/U

OPT

N/U

OPT
OPT
OPT
OPT

N/U
N/U
N/U

AlB
AlB
AlB

OPT
OPT

AlB
AlB

OPT
OPT
OPT
OPT

OPT

OPT
OPT
OPT
OPT
OPT
OPT

OPT

AlB
AlB
AlB
AlB
AlB
AlB

RQD
RQD
RQD
RQD
RQD

N/U
N/U
N/U
N/U
N/U

OPT
OPT
OPT
OPT
OPT

N/U

OPT

N/U

OPT

(continued on next page)

Planning DECnet

2-13

Table 2-3 (cont.):

Component Selection Summary for NETGEN

Component

System
-11M
-11M+

-115

PreGEN

Optional PSI Components
Data Link Mapping process (DLM)
Trace Capture task (TRA)
Trace Interpreter task (TRI)
X.29 Terminal Access ACP with HTDRV (X29ACP)
KMX/KMY Microcode Dumper (DUK)
KMX/KMY Microcode Dump Analyzer (KDA)
PSI high level language library (PSIFOR.OLB)
PSI MACRO library (PSI.MLB)

Note 9
OPT
OPT
OPT
OPT
OPT
OPT
OPT

N/U
N/U
N/U
N/U
N/U
N/U
N/U
N/U

LAT Components
LATprocess
LAT Control Program (LCP)

N/U
N/U

OPT
OPT

N/U
N/U

OPT
OPT

Notes:

Part of the RSX-11M-PLUS Executive.

MLD is required for all nodes with devices that need microcode to be loaded.

DLX is required if you specify certain operations (see Section 2.3.1.3).

NMDRV is required for RSX-11S only if EVL is specified.

Supplied on kit, but not moved by NETGEN procedure.

DCP is required for all nodes if you specify devices that require software DDCMP support.

EPM is required for all nodes if you specify Ethernet devices.

RCP is required only for routing nodes.

DLM is a required component for nodes that use DECnet over a Packet Switching Data Network. For all other nodes, DLM is never used.

2.3.1

Optional DECnet Components

The following paragraphs describe optional components for a DECnet-RSX node.
Note, however, that some of these components are selected for you as a result of
another selection you have made. For example, when you specify DLL for your target
system, DLX is automatically generated. On pregenned RSX-IIM-PLUS end nodes,
only the satellite support components are optional.

2.3.1.1 Microcode Loader (MLD) - The Microcode Loader (MLD) loads devices
with the proper microcode. Currently these devices include the UNA and the KMC/
KMS devices. KMC devices are KDP and KDZ. KMS devices are KMX and KMY.

2-14

DECnet-RSX Network Generation and Installation Guide

2.3.1.2 Event Logger/Collector (EVL/EVC) - The Event Logger (EVL) creates
records of network events, such as circuit state changes, node state changes, and
unusual network occurrences. Records of these events are sent to the event collector
(EVC). Specify event logging to facilitate tuning the network and isolating problems.
If you specify EVL, EVC is automatically included. It collects network events and sends

them to the proper terminal, file, or monitor task.

2.3.1.3 Direct Line Access Controller (DLX) - The Direct Line Access Controller
(DLX) is necessary for services that must bypass normal network data channels. DLX is
required if you specify down-line system loading, up-line system dumping, NCP loop
line/circuit testing, console carrier requestor support, or user-written programs that
use the DLX QIO interface.
2.3.1.4 Network Control Program (NCP) - The Network Control Program (NCP) is
the primary program for controlling network operations and is required for RSX-IIM
and RSX-IIM-PLUS nodes. NCP is optional for RSX-IIS nodes. NCP allows the RSXlIS network manager to perform on-line network management functions at a local terminal, or at a remote terminal using RMT .
2.3.1.5 Network Management Device Driver (NMDRV) - If you specify event logging for an RSX-llS node, the network management device driver (NMDRV) is always
built. NMDRV is required for RSX-IIM/-IIM-PLUS nodes.
2.3.1.6 Event File Interpreter (EVF) - The Event File Interpreter (EVF) provides
many of the same services for network event logging that the RSX-ll Report Generator (RPT) provides for error logging. EVF allows you to get a formatted report from
the event file you set up using the NCP SET LOGGING command. The EVF command
interface allows you to specify which events you wish formatted based on such parameters as the date on which the event occurred, the event type, and the event entity. EVF
is not available for RSX-11 S nodes.
2.3.1.7 Digital Communications Process (DCP) - The Digital Communications
Process (DCP) is the DDCMP line protocol process that provides an error-free data path
over a physical device channel. DCP is required if you specify devices that require soJtware DDCMP support.
2.3.1.8 Ethernet Protocol Manager (EPM) - The Ethernet Protocol Manager (EPM)
controls access to the Ethernet. EPM is required if you specify devices (DEUNA,
DEQNA) that support access to the Ethernet.

Planning DEenet

2-15

2.3.1.9 NETFOR.OLB and NETLlB.MLB - The DECnet high level language library
(NETFOR.OLB) allows user-supplied tasks, written in FORTRAN, COBOL, and BASICPLUS-2, to interface to the DECnet software. The DECnet Macro library (NETLIB.MLB)
allows user-written MACRO-II tasks to interface to the DECnet software. If you want
to code tasks using these languages, request the language libraries.
2.3.1.10 Network Information and Control Exchange (NICE) - The
Network
Information and Control Exchange (NICE) allows a privileged user (usually a system or
network manager) at any other node in the network to perform network management
functions on the node you are generating.
2.3.1.1'1 Event Logging Receiver (EVR) - The event-logging receiver (EVR)
receives event notification messages from other nodes and logs those messages to
either a local terminal, a file, or a monitor task. Specify EVR if you wish the local node
to receive events from other nodes.
2.3.1.12 Network Display Utility and Server (NTD/NTDEMO) - The Network Display utility (NTD) allows you to monitor the current state of a node. NTD sends
requests to the Network Display Server task, NTDEMO, to obtain information from a
local or remote RSX node. NTD requests and displays the information on the local
node.
NTDEMO provides the information to the NTD that requested it.

2.3.1.13 Link Watcher (LIN) - The Link Watcher (LIN) determines what is to be
done with an automatic service request. LIN is required to perform down-line system
loading, up-line system dumping, and loop circuit testing. LIN is always built for RSX11M-PLUS nodes, is optional on RSX-IIM nodes, and is not available on RSX-llS
nodes.
2.3.1.14 Network Verification Program (NVP) - The Network Verification Program (NVP) verifies accounting data for incoming connect requests against the system
account file. NVP is generated for all RSX-IIM and RSX-IIM-PLUS systems that have·
multiuser protection.
2.3.1.15 Down-line System Loader (DLL) - The down-line system loader (DLL)
resides on the host node and is used to down-line load remote systems. The host node
must be directly connected by a physical link to the system to be loaded and must be
able to access the target system's image file. DLX is automatically included if you specify OLL.

2-16

DECnet-RSX Network Generation and Installation Guide

2.3.1.16 Up-line System Dumper (DUM) - The up-line system dumper (DUM)
resides on the host node and is used to dump the memory from a remote system up-line
to a DECnet-11M or DECnet-11M-PLUS system. DUM services dump requests issued
by LIN in response to a dump request message received from an adjacent system.
2.3.1.17 Console Carrier Requester (CCR) - The console carrier requester (CCR)
provides access to console functions on an Ethernet remote system. If you have an
Ethernet line on your host node, you can specify CCR. CCR is not available on RSX-11S
nodes.
2.3.1.18 Host Task Loader (HLD) - The Host Task Loader (HLD) resides on a host
node and is used to load task images into a target RSX-11S node's memory. HLD communicates with SLD to down-line load a task, up-line checkpoint a task, or down-line
overlay a task segment. HLD is used only on the host node and is not available on RSXlIS nodes.
2.3.1.19 Satellite Task Loader (SLD) - The satellite task loader (SLD) resides on an
RSX-11S satellite node and is used to communicate with HLD on a host node. SLD
requests loading support from HLD for RSX-11S tasks. You must specify SLD for the
RSX-11S node in order to down-line load task images, up-line checkpoint a task, or
down-line overlay a task segment. SLD is available only for RSX-11S nodes.
2.3.1.20 Network File Transfer Utility (NFT) - The Network File Transfer utility
(NFT) allows users to access files on other nodes in the network. NFT operations
include file transfer, file deletion, directory listing, and commandlbatch file submission. NFT communicates with FAL on the remote node.
2.3.1.21 File Transfer Spooler (FTS) - The File Transfer Spooler (FTS) allows users
to access files on other nodes in the network. Spooled access indicates operations may
execute immediately, operations may wait for a specific time, or operations may be
queued until the local or target node becomes available.

The File Transfer Spooler Dequeuer (FTSDEQ) removes requests one by one, creates a
logical link to the FAL on the appropriate remote node, and executes the file transfer
requests. If you specify FTS, FTSDEQ is automatically included.
2.3.1.22 File Access Listener (FAL) - The File Access Listener (FAL), the network
file access server, performs file operations requested by users on other nodes. FAL operations include transferring and concatenating files, deleting files, submitting and executing batch or command files, listing directories, and accessing sequential records.

RSX-llS FAL supplies support for unit record equipment. FAL with RMS provides
RMS-11 record access capabilities as well as normal file access. This allows RMS on a
remote node to perform record access to sequential, relative, and indexed files on your
node.

Planning DECnet

2-17

2.3.1.23 Command File/Batch File Submission Task (MCM) - The command file I
batch file submission task (MCM) allows an NFT user on a remote node to submit an indirect command file or batch file for execution on the node you are generating. MCM
receives requests from the local FAL on behalf of the remote node. On RSX-11MI
-11M-PLUS nodes, MCM submits the file for execution by requesting the indirect command file task ( ... AT). For RSX-11M-PLUS only, MCM can optionally submit batch files
to the BATCH queue.
2.3.1.24 Remote Terminal Utility RMTACP (RMT/RMTACP) - The Remote Terminal utility RMTACP (RMT/RMTACP) resides on the user's local node and allows the
user's terminal to connect to any other RSX node in the network that supports the
remote network terminal driver and ACP, HT: and RMHACP.
2.3.1.25 Remote Terminal Host ACP and HTDRV (RMHACP/HTDRV) - The
remote terminal host ACP (RMHACP) and Host Terminal Device Driver (HTDRV) allow
users on remote RSX nodes to connect to the local node.
2.3.1.26 Network Command Terminal Server Process (NCT) - The
Network
Command Terminal Server Process (NCT) allows the user's terminal to be logically connected to any node in the network that supports network command terminals.
2.3.1.27 Remote Command Terminal Host (RTH) - The remote command terminal host (RTH) is the remote command terminal host support process. It resides on the
user's local node and allows users on other nodes to logically connect their terminals to
the local node through the network command terminal facility.
2.3.1.28 Terminal Communications Utility (TLK) - The Terminal Communications utility (TLK) allows you to engage in an interactive dialog with, or send single line
messages to, another user on a remote DECnet-RSX, DECnet-IAS, or DECnet-RSTS
node. The remote node must have the Listen utility (LSN) installed.
2.3.1.29 Listen Utility (LSN) - The Listen utility allows a remote user running TLK
to communicate with a terminal user on your node. LSN is automatically included for
all DECnet-RSX nodes on which TLK is specified.
2.3.1.30 Phone Communications Utility (PHO) - The Phone Communications utility (PHO) allows you to engage in an interactive dialog with another user on a remote
DECnet-RSX, DECnet-VAX, DECnet-Micro/RSX, or DECnet-PRO node. PRO provides many advanced features not included in the TLK utility (such as connect by
name). The advanced features of PHO require the use of accounting information not
available on RSX-11M or RSX-11S systems. PHO, therefore, is available only on RSX11M-PLUS and VAX/VMS systems.

2-18

DECnet-RSX Network Generation and Installation Guide

2.3.1.31 Remote Task Control Utility (TCl) - The Remote Task Control utility
(TCL) is the remote task control server task. TCL allows FORTRAN programs running
on other nodes to control task execution on the node where TeL is installed. TCL
accepts requests to run a task immediately or at some specified time, abort the current
running of a task, or cancel requests for future execution of a task. Routines to connect
to TCL and issue task requests are contained in the DECnet FORTRAN object library.

2.3.2

Optional PSI Components

The following paragraphs describe the optional PSI components that you can specify
for your DECnet-RSX/PSI node.

2.3.2.1 Data Link Mapping (DlM) - The data link mapping process (DLM) allows
DECnet nodes to use a Packet Switching Data Network (PSDN) to communicate with
each other. D LM performs the conversion between the protocols used by PSI and the
protocols used by DECnet to allow DECnet nodes to communicate over a PSDN.
2.3.2.2 Trace Capture Task (TRA) - The trace capture task (TRA) determines what
types of packets are being transmitted and received on a line, and monitors the line's
performance. TRA writes the information it has gathered to a disk file.
2.3.2.3 Trace Interpreter Task (TRI) - The trace interpreter task (TRI) interprets
the data in the file created by TRA and prints a file containing this information.
2.3.2.4 X.29 Terminal Access ACP with HTDRV (X29ACP) - The X.29 ancillary
control processor (X29ACP) task allows a user at a remote X. 29 terminal connected to a
packet assembly/disassembly (PAD) facility to communicate with the DECnet-RSX/PSI
software as if the terminal were locally connected. X29ACP is implemented as an RSX11M/-lIM-PLUS pseudodevice named HT: (which has a device driver HTDRV) and
the associated ACP (X29ACP). When a remote X.29 user issues a command to connect
to your node, a single unit of this HT: device is allocated to the user.
2.3.2.5 KMX Microcode Dumper (DUK) - The KMX microcode dumper (DUK),
required if you have a KMX device, dumps the KMX microcode to a disk.
2.3.2.6 KMX Microcode DUIJlP Analyzer (KDA) - The KMX microcode
analyzer (KDA) formats and p;ints the KMX microcode dumped by DUK.

dump

2.3.2.7 PSI High level language Library (PSIFOR.OlB) - The PSI high level language library (PSIFOR.OLB) allows you to write network tasks in FORTRAN. If you
want to code tasks using FORTRAN, include this library in your target node.

Planning DEenet

2-19

2.3.2.8 PSI Macro Library (PSI.MLB) - The PSI MACRO library (PSI.MLB) allows
you to write network tasks in MACRO-II. If you want to code tasks using MACRO-II,
include this library in your target node.

2.4

The Next Step

You should now have some idea of what type of information is required to do a
NETGEN and what decisions you are going to have to make along the way. The reading
path for the rest of the manual depends on the type of distribution kit you purchased. If
you received a pregenned RSX-IIM-PLUS end node kit, you can skip to Chapter 8. All
other readers should proceed to Chapter 3.

2-20

DECnet-RSX Network Generation and Installation Guide

3
Getting Ready for RSX-11 M/M-PLUS NETGEN

This chapter discusses the requirements that must be met before a RSX-IIM/M-PLUS
NETGEN can be performed. These requirements vary, depending on the target system.
Jfyour target system is RSX-IIM-PLUS, read Section 3.1 and skip Section 3.2. Jfyour
target system is RSX-IIM or RSX-llS, skip Section 3.1 and read Section 3.2. Section
3.3 is common to all RSX systems and describes the PREGEN procedure that may have
to be performed prior to NETGEN.

3.1

RSX-11 M-PLUS System Requirements

This section describes the pre-NETGEN requirements for RSX-IIM-PLUS systems.
RSX-IIM-PLUS systems require very little advance preparation for network generation. Section 3.1.1 describes the considerations for preSYSGENned RSX-IIM-PLUS
systems. The considerations for SYSGENned RSX-IIM-PLUS systems are discussed in
Section 3. 1. 2.

3.1.1

PreSYSGENned RSX-11 M-PLUS Systems

Pre-NETGEN considerations on preSYSGENned RSX-IIM-PLUS systems are listed
below and described in the sections that follow.
•

Hardware CSR selection

•

System selection

•

Customizing the start-up command procedure

•

Possible device reconfiguration

•

Possible PREG~N

3-1

3.1.1.1 Hardware CSR Selection - You must have the "floating" device control
and status registers (CSRs) at the proper addresses on the target processor. Tile proper
addresses for floating devices are determined according to the standard UNIBUS/QBUS
device rankings. A command file, [200,200]FLOAT.CMD, is provided on the distribution kit to aid you in determining the device CSR)address.
3.1.1.2 System Selection - There are two system images on the RSX-11M-PLUS
preSYSGENned distribution kit. One system (the Kernel I-space system) runs on a PDP11123-PLtJS or a PDP-11124 processor. The other system (the Kernel D-space system)
runs on the PDP-11144, PDP-11/70, PDP-11173, PDP-11183, and PDP-11184 processors. Before doing a NETGEN, you should use the procedure provided with the RSX11M-PLUS distribution kit to delete the system that you will not be using. If you do not
use the supplied procedure, some unwanted libraries may be left on the target system
disk and NETGEN may give you the option of using them (for example, supervisor
mode FCS or supervisor mode RMS libraries).
3.1.1.3 Customizing the Start-up Procedure - The
RSX-11M-PLUS
preSYSGEN ned systems come with a start-up procedure that will deallocate the Communications Executive (CEX) space needed by DECnet-RSX. In order to prevent this, the file
[l,2]SYSPARAM.DAT must be modified. You must modify the line 'DECNET = NO' to
read 'DECNET = YES'. The system must be rebooted after this change in order for the
change to take effect.
If you have properly modified the file, you will see the following lines during system

start-up:
Step 2 - Deal locating DECnet communications executive
System configuration includes DECnet

The presence of the second line indicates that the CEX space has not been deallocated.

3.1.1.4 Device Reconfiguration - The start-up file will automatically configure the
devices on the system so that they can be used by the Executive and not by the network. To release a device for DECnet use, observe the following procedures:
•

Check the CON_ONLINE~LL = parameter in your [l,2]SYSPARAM.DAT file to
be sure that it is set to YES.

•

Enter CON OFFLINE commands for those devices that you want DECnet to use.
You can enter commands individually or you can build a command file that will
execute these commands automatically after the startup procedure completes. If
you want to take the command file approach, you must specify the name of the
command file in the STARTFILE = parameter of [l,2]SYSPARAM.DAT. See the
RSX-IIM-PLUS System Management Guide for instructions on how to use the
CON OFFLINE command.

3-2

DECnet-RSX Network Generation and Installation Guide

3.1.1.5 PREGEN - Depending on the type of kit you received, you may have to perform a PREGEN. Proceed to Section 3.3 to see if your kit type requires a PREGEN.

3.1.2

SVSGENned RSX-11 M-PLUS Systems

The RSX-IIM-PLUS SYSGEN procedure displays three questions that must be
answered (as described in the following sections) to properly configure the RSX-IIMPLUS operating system for DECnet support. These questions specify the inclusion of:
•

Support for DECnet

•

DECnet communication devices

•

Queued file access

After reading the following subsections, proceed to Section 3.3 to determine if your
type of kit requires a PREGEN.
3.1.2.1 Including Support for DECnet, PSI, LAT - To create the proper support
for DECnet, PSI or LAT on the system you are generating, answer YES to the following
question in the Executive Options section of SYSGEN:
>* CE120 Do you want support for communications products

(such as DECnet, PSI and LAT)?[Y/N D:N]:

3.1.2.2 Including support for LAT Terminal Servers - If you want to support LA T
terminal servers, answer YES to the following question in the Executive Options section of SYSGEN:
>* CE122 Do you want host support for LAT terminal servers? [YIN D:N]

If you answer NO to this question you will generate a system that cannot be used with
DECnet.
3.1.2.3 RSX and DECnet Communications Devices - Because the RSX-IIMPLUS Executive does not need to know anything about hardware that will be used as
DECnet communications devices, and because no communications device can be
shared between DECnet and the Executive, do not specify any DECnet communications devices when answering questions in the Target Configuration section of
SYSGEN.

However, when you generate your RSX system, you must specify how much space is to
be allocated for interrupt vectors on your target system, and the space must be sufficient to include the interrupt vectors belonging to DECnet devices.
To specify properly, answer the following question with the highest vector address of
devices on your system, including the DECnet communications hardware.
>* CP9632

What is the highest interrupt vector address?

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-3

These vector addresses are available as a result of hardware installation.
For this question, do not specify the default, since the value that SYSGEN computes in
this case would not take into account DECnet communication devices. In answering
this question for your operating system, be sure to leave enough space for devices you
have or will soon have.
If your DECnet system includes any of the following character interrupt devices -

DUP, DU, DUV, DL, DLV, or DPV rupt vector address.

you must add 20 octal bytes to the highest inter-

The highest interrupt vector address you can specify is 774. This value, which includes
the 20 additional octal bytes required for DUP, DU, DUV, DL, DLV and DPV, allows
enough space for any devices you may choose.

3.1.2.4 Queued File Access Requests - FTS allows file access requests (file transfers, etc.) to be queued and processed in sequence. If you want FTS, the RSX-IIMPLUS queue manager (QMG) must be built during SYSGEN by answering YES to the following question:
>* CE180

Do you want to include the queue manager? [YIN D:N]:

For more information on FTS, see the DECnet-RSX Guide to User Utilities.

3.2

RSX-11 MIS System Requirements

This section describes the pre-NETGEN considerations necessary for RSX-IIM and
RSX-I1S systems. These considerations include:
•

Partition layout decisions (see Section 3.2.1)

•

Pool usage considerations (see Section 3.2.2)

•

Avoiding DECnet loading problems (see Section 3.2.3)

•

Special considerations for NMVACP (see Section 3.2.4)

•

SYSGEN considerations (see Section 3.2.5)

•

RSX-I1S considerations (see Section 3.2.6)

Before running NETG EN, you must decide how you want to layout the partitions in
your system. Partition layout determines what tasks are installed in what partitions.
You must then perform a system generation (SYSGEN) to configure your system for
DECnet and for your own requirements.
For the most part, there is considerable flexibility in the order in which you perform
these operations. The following steps, however, are fixed:
•

3-4

NETGEN follows SYSGEN. If you perform a new SYSGEN, you must perform a
newNETGEN.

DECnet-RSX Network Generation and Installation Guide

•

The size of the Executive address space (16K or 20K words) is determined during
SYSGEN (see Section 3.2.5.S), and cannot be changed without a new SYSGEN.

•

You must specify the starting address and the size of the partition for the Communications Executive (CEXPAR) during NETGEN. Changing the starting address
requires a new NETGEN. Changing the size of CEXPAR does not require a new
NETGEN, but if the size is reduced, the new size must be large enough to contain
the Communications Executive.

3.2.1

Determining Partition Layout

NETGEN builds the Communications Executive for the CEXPAR partition. You must
add this partition (see Section 3.2.5.11).
With the exception of the Communications Executive, NETGEN builds all tasks to run
in one partition, GEN. Depending on your system size and requirements for the network, you may need to add or modify partitions to obtain optimum performance. The
partition layout is defined by editing the SYSVMR command file (see Section 3.2.5.11)
and then modifying the NETINS.CMD command file so that the network tasks are
installed in the desired partitions (see Chapter 9). This section describes how you can
determine the proper partition layout for your system.
The way you configure your network depends on your system memory size. For
NETGEN purposes, your system size can be classified into small, medium, and large:
•

Small = Under SOK words of memory

•

Medium = SOK to 124K words of memory

•

Large = Over 124K words of memory

These definitions are based on light network loading, are somewhat arbitrary, and
serve only as a guide. What counts most is available memory, and this quantity is subject to the requirements imposed by the specific complement of network components
you select and by your other software and application programs.
To determine the available room in the GEN partition, you may need to perform a
NETGEN, inspect the size of components selected, and compute the amount of remaining space. Based on this computation, you may then need to reconfigure partitions.
It is possible to approximate the size of GEN by totaling the estimated size of non-GEN

components and subtracting that from the memory size. Table 3-1 gives the sizes of certain key components. These sizes are merely estimates and can vary due to individual
system requirements and to future patches. Final sizes should be determined by examining the map for each individual component.
In estimating the size of GEN, you also need to know the size of other partitions, as
shown in Figure 3-1. You can obtain this information after SYSGEN by executing a
VMR PAR command. If you need information on how to do this, see your system generation manual.

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-5

For large RSX-IIM systems, use of the single partition GEN produces satisfactory performance. For small to medium size systems, you may need to add partitions. There are
two areas where problems may occur - applications requirements and loading.
The applications problem occurs when user applications take ·so much of the system
resources that network performance is degraded. In special cases it may be necessary to
load network tasks in special partitions to ensure that they have sufficient memory in
which to run. Any such situation would need to be handled on an individual basis.
The loading problem may occur when using NCP to load the network in systems where
memory is at a premium. This problem is discussed in Section 3.2.3.

___

HIGH

GEN

FCPPAR ,OPTIONALI

SYSPAR

DRVPAR ,OPTIONAL,

TTPAR ,OPTIONALI

LDRPAR ,OPTIONAl!
___ 116 or 20K WORDS)
CEXPAR

DSR I POOl)

Executive
Address
Space

EXECUTIVE
~o

Figure 3-1 :

3-6

Partition Layout after SYSGEN (CEXPAR added)

DECnet-RSX Network Generation and Installation Guide

Table 3-1:

Approximate Sizes of Key Components

DECnet Components

Size*

RSX-IIM Executive

Varies (see Note 1)

DSRpool

Varies (see Note 1)

Communications Executive

lK-1.5K (see Note 1)

FllACP

2K-9K (see Note 2)

NETACP

4K-5K (see Note 3)

NMVACP

6K (see Note 4)

NTINIT

2.0K

NCP

12K

NTL

Minimum network

12K (see Note 5)

PSI Components

Size*

PLI

X25ACP

DLM

LAB

SDP/SDV

O.5K

KMX

X29ACP

HT:

O.75K

Minimum DECnet/PSI system

20K-22K (see Note 6)

* Approximate number of decimal words
Notes:
1.

RSX Executive, Dynamic Storage Region (DSR) pool, and Communications Executive
together total 16K or, if you choose the large Executive, 20K. You can change this figure
somewhat depending upon your needs and resources (see Section 3.2.5.10). The Communications Executive varies from 1 to 1.5K, depending upon what lines and options you have
selected.

Files 11 ACP (F lIACP) is available in four sizes ranging from approximately 2K to 9K. If you
put F llACP in a partition with other larger components (for example, NET ACP), use of the
smaller two FIIACPs are not recommended. The smallest FIIACP does not provide full

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-7

functionality, and the next larger is heavily overlaid. The small F llACPs are optimized for
minimum memory size, not performance, and since the memory is already allocated, a
higher performance FllACP is recommended. The third FllACP in order of size is 5K, provides full functionality, and is moderately overlaid. (See the RSX-11 M System Generation
Manual for more information on F llACP .)
3.

NETACP comes in two sizes, 4K and approximately 5K. The smaller version does not support network command terminals and layered products.

NMVACP is initially built for GEN as a 4K word task. In that size, it can handle only one user.
However, when additional requests come in, it extends itself to a larger size. You can also
install it with an increment to handle more users, thus eliminating the need for it to extend
itself. In this case, the formula for determining the decimal word size of NMVACP is:
4K(170 x V)
The variable, U, is the number of users that NMVACP can service (it starts with a base of 2).
See Section 3.2.4 for more information on NMVACP.

The minimum required network software (in addition to CEX) consists of AUX, ECL, XPT,
NETACP, network pool, at least one DDM, and RCP (for a full routing node). Excluding
NET ACP, which you may wish to load in a separate partition as shown here, these components total approximately 12K words (14K for DECnet-llM-PLUS).

The minimum DECnet/PSI system is 20K words (22K for DECnet-llM-PLUS). This minimum is for a node with a single DLM circuit. The minimum DECnet/PSI system consists of
AUX, ECL, XPT, DLM, PLI, one DDM (KMX, SDP, or SDV), LAB (only when the DDM is SDP
or SDV), and DLX (required on DECnet-IIM-PLUS).

3.2.1.1 Small RSX-11 M Systems - For small systems, the partition layout shown
in Figure 3-2 is recommended as a way of solving loading problems.

3-8

DECnet-RSX Network Generation and Installation Guide

, . . - - - - - - - - - - . . . , . . - HIGH
GEN
ALL OTHER TASKS
FCPPAR
F11ACP
NETACP
NMVACP
NTINIT
OTHER PARTITIONS
CEXPAR
CEX
DSR (POOL)

EXECUTIVE
" " - - - - - - - - -... . . - 0

Figure 3-2:

Partition Layout for Small RSX-11 M Systems

Here, NETACP, NMVACP, and NTINIT are installed in FCPPAR, where they run along
with FIIACP, the Files 11 ACP. NETACP must be generated as checkpointable. You
should set the size of FCPPAR to accommodate the largest task in the partition. This
will often be NETACP. You can arbitrarily choose a size for FCPPAR that is too large
(say 7K words) and then reduce the size appropriately after NETGEN, when you know
actual sizes. All other network and user tasks are installed in GEN.
This layout ensures that NMVACP and NTINIT do not tie up space in GEN that is
required for loading. Putting NETACP in FCPPAR also helps reduce fragmentation.
NOTE

This layout provides satisfactory performance if GEN is approximately 32K or
larger. For smaller systems, it is recommended that you avoid using NCP to
load the network. Instead, use VNP to prepare a suitable system image and then
boot the system directly. The smaller your GEN partition, the more important
this recommendation. For information on VNP, see the DECnet-RSX Guide to
Network Management Utilities.

3.2.1.2 Medium RSX-11 M Systems - For medium size RSX-l1M systems, the partition layout in Figure 3-3 is recommended.

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-9

GEN
ALL OTHER TASKS

4--

HIGH

ACPPAR
NETACP
FCPPAR
F11ACP
NMVACP
NTINIT

OTHER PARTITIONS
CEXPAR
CEX
DSR (POOL)

EXECUTIVE

......_ _ _ _ _ _ _ _ _ ...--0

Figure 3-3:

Partition Layout for Medium RSX-11 M Systems

Here NETACP has been removed from the FCPPAR partition and placed in its own partition, ACPPAR. (You can name this partition anything you like; in the examples in this
manual, it is ACPPAR.)
There is an advantage in splitting NETACP out from the FCPPAR partition. Since both
NETACP and FIIACP can be very active tasks, this layout avoids needlessly swapping
them in and out of memory. The result is improved performance. This layout has all the
other advantages recommended for small RSX-IIM systems.

3.2.1.3 Large RSX-11 M Systems - For large RSX-IIM systems, performance is satisfactory when all tasks are placed in the partition GEN. However, if the system performs a great deal of network I/O or if you wish to make NETACP non-checkpointable,
the partition layout shown for medium size RSX-IIM systems (Figure 3-3) may produce improved performance.

3-10

DECnet-RSX Network Generation and Installation Guide

3.2.2

Pool Usage Considerations

Along with the issue of partition layout, there is the additional consideration of pool
usage. In large systems this will probably not be a problem. In small and medium systems, you may want to reduce the amount of pool used by the network software. The
general breakdown of pool requirements for various DECnet features given below can
help in minimizing pool usage.
Installing a task on RSX-IIM or RSX-IIS systems takes about 24. words of pool
(depending on exact Executive configuration). From 6 to 30 DEC net tasks can be
installed depending on your NETGEN answers. If you have limited pool, you may want
to edit the NETINS.CMD file to limit the number of installed tasks (see Chapter 9).
Other network use of pool includes:
Control Buffers

18. words for each control buffer

Incoming remote
terminal
links

30. words for each allowed link
for HT: and RT: driver
data structures

Logical links

2. words for each allowed logical link

Circuits

21. words for each defined circuit

Loading and starting the network can use a wide range of values of system pool. During
network load and start-up, control buffers and common areas for DECnet processes are
allocated.
When running network tasks, the first logical link adds 18. words to pool usage and
each additional link adds 11. words. These do not include the pool usage associated
with QIOs on the logical links.
In general, minimizing the number of installed tasks and the number of active logical
links per task is the best way to cut down on pool usage.

3.2.3

Avoiding DECnet Loading Problems

In order to understand the loading of RSX-IIM DECnet, consider Figure 3-4. This figure is a version of Figure 3-1. The partitions SYSPAR, DRVPAR, TTPAR, and LDRPAR
are combined in OTHER PARTITIONS.

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-11

, . . - - - - - - - - - . . . , .....- HIGH
GEN
ALL OTHER TASKS
FCPPAR
F11ACP
OTHER PARTITIONS
SYSPAR
DRVPAR
TTPAR
LDRPAR

This represents
four partitions

CEXPAR
CEX
DSR (POOL)

EXECUTIVE

~--------.... .....-o

Figure 3-4:

Typical Partition Layout

This example assumes that you decided to put F lIACP, the Files 11 ACP, in its own partition, FCPPAR. This is the SYSGEN default. All DECnet tasks run in the system-controlled partition GEN.
When you load the network by using the NCP SET SYSTEM command, a series of
actions are initiated:
1. The Executive loads NCP in GEN.
2.

NCP makes an I/O request to NMDRV, the network management device driver.

3. The driver calls NMVACP, the network management volatile ACP.
4.

NMVACP in turn calls NTINIT, the Network Initialization task.

S. After performing initialization, NTINIT calls NTL, the Network Loader, to load the
network.
Thus, the entire sequence is as follows:

NCP--> NMDRV--> NMVACP--> NTINIT--> NTL--> NETWORK
Each task is loaded into GEN in the sequence indicated. Each task can checkpoint the
task that precedes it in the calling sequence, if it needs memory in which to run. However, because NCP has an I/O request outstanding, it cannot be checkpointed by other
tasks; it remains in memory until the entire cycle is completed. Likewise, when NTL
loads the network, it cannot checkpoint other tasks in order to make room for the network tasks. This is due to certain technical features of NTL.

3-12

DECnet-RSX Network Generation and Installation Guide

Generally, the loading procedure just described proceeds satisfactorily. However, consider the following example. Suppose the above loading procedure occurs and NCP,
NMDRV, NMVACP, NTINIT, and NTL are all in memory at the same time. No
checkpointing occurs because there is enough memory for all; there is even, say, 10K
of memory left over. NTL then tries to load the network. The load fails because the network requires a minimum of 12K to load and there is only 10K available. NTL tries to
load the network software in the existing space, but since it is incapable of
checkpointing other tasks, it fails and prints an error message.
This problem can occur only when attempting to load the network and, depending
upon the size of your system, can be prevented in a variety of ways, including:
•

Removing extraneous application tasks

•

Loading the network prior to booting the system, using the VNP SET SYSTEM command, to ensure maximum memory space.

•

Rearranging partition layout (see Sections 3.2.1.1 and 3.2.1.2).

3.2.4

Special Considerations for NMVACP

The network management volatile ACP task (NMVACP) may require special treatment
during NETGEN. A brief explanation of NMVACP may make this clearer.
NMVACP works in conjunction with the Network Control Program (NCP) task and the
network management device driver (NMDRV) to execute NCP commands. NCP scans
user commands and determines what action is required. It then passes this information
on to NMDRV, which performs certain preliminary processing. NMDRV, in turn, calls
NMVACP to execute the NCP command. These components are described in Section
2.3.
NMVACP can also execute Network Information and Control Exchange (NICE) commands that originate at a remote system.
NMVACP can handle a number of requests simultaneously by maintaining a record of
the status of each request in its own buffer space. If it receives more requests than it has
buffer space for, it makes a request to the RSX Executive to be checkpointed. Then it
may be brought back into memory as a larger task.
In large systems, this presents no problem. In small to medium RSX-IIM systems, however, the possibility of deadlock arises. (This is one reason why the partition layouts
shown in Sections 3.2.1.1 and 3.2.1.2 are recommended.) When NMVACP asks the
Executive to be checkpointed, if there is not sufficient space to come in as a larger task,
a deadlock may occur: the NCP commands SET SYS and SET EXE STA ON will never be
executed. The problem is compounded by the fact that NCP cannot be checkpointed
while it has an I/O request outstanding - that is, until the NCP command is completed.

Getting Ready for RSX-11 M/M-PLUS N ETGEN

3-13

To handle this problem, you can do either of the following:
•

Place NMVACP in the FCPPAR partition, along with FIIACP. This provides extra
memory in the partition GEN, as the memory in FCPPAR is already allocated.
Adding NMVACP so that it shares this partition does not require extra space. The
layouts suggested for small and medium RSX-IIM systems take this approach.

•

Modify the installation command file NETINS.CMD so that NMVACP is installed
with a memory increment to provide space for additional requests. NETINS.CMD
is described in Chapter 9. The line that installs NMVACP should be modified to read

INS NMVACP/INC=s~e/CKP=NO
Here size is the octal byte increase in size and can be computed by multiplying 524
octal bytes by the number of simultaneous users that you wish to support.
(NMVACP is originally built to handle 2 users.) The /CKP = NO switch eliminates
the possibility of deadlock, since NMVACP can never be checkpointed.

3.2.5

RSX-11 M/-11 S SYSGEN Considerations

Prior to running NETGEN, you must perform a SYSGEN that prepares the RSX operating system to accommodate the DECnet software you are creating. The following areas
affect DECnet-RSX software:
•

SYSGEN support for DECnet (Section 3.2.5.1)

•

RSX and DECnet communications devices (Section 3.2.5.2)

. ' Standard function system (RSX-IIM only) (Section 3.2.5.3)
•

Network command terminal support (Section 3.2.5.4)

•

System controlled partitions (Section 3.2.5.5)

•

Checkpoint support (RSX-I1M only) (Section 3.2.5.6)

•

Terminal driver support (Section 3.2.5.7)

•

Queued file access requests (RSX-IIM only) (Section 3'.2.5.8)

•

Large or small Executive (Section 3.2.5.9)

•

Editing SYSVMR.CMD (Section 3.2.5.10)

•

Adding CEXPAR (Section 3.2.5. 11 )

•

Adding other partitions (Section 3.2.5.12)

Some of the SYSGEN questions that follow may not be pisplayed because the support
required may be automatically included as a result of a previous question. For details
on running SYSGEN, see the system generation manual for your system.
RSX-I1S systems have additional SYSGEN considerations for SLD and NETPAN support. These are covered in Section 3.2.6.

3-14

DECnet-RSX Network Generation and Installation Guide

3.2.5.1 Including Support for DECnet - To create the proper support for DECnet
on the RSX-IIM/S system you are generating, answer YES to the following question in
the Executive Options section of SYSGEN:
>* 32.

Include support for communications products (such as OECnet)?
[YIN O:N]:

If you answer NO to this question, you will generate a system that will not support

DECnet.

3.2.5.2 RSX and DECnet Communications Devices - Because the RSX-IIM/S
Executive does not need to know anything about hardware that will be used as DECnet
communications devices, and because no communications device can be shared
between DEC net and the Executive, do not specify any DECnet communications
devices when answering questions in the Target Configuration section of SYSGEN.
However, when you generate your RSX system, you must specify how much space is to
be allocated for interrupt vectors on your target system, and the space must be sufficient to include the interrupt vectors belonging to DECnet devices.
To specify properly, answer the following question with the highest vector address of
devices on your system, including the DECnet communications hardware.
>* 14. Highest interrupt vector [0 R:0-774 0:0]::

DUP, DU, DL, DUV, DLV, or DPV - you must add 20 octal bytes to the highest interrupt vector address. The highest interrupt vector address you can specify is 774. This
value, which includes the 20 additional octal bytes required for DUP, DU, DUV, DL,
DLV and DPV, allows enough space for any devices you may choose.

3.2.5.3 Standard Function System (RSX-11 M Only) - For
Question 8. at the beginning of SYSGEN displays:

RSX-IIM

systems,

>* 8. Do you want a Standard Function System? [YIN]:

If you answer this question YES and select the standard function system, you will auto-

matically build the following options into your system (others may be included as
well):
•

Network command terminal support

•

System controlled partitions

•

Checkpointing (but without the system checkpointing file)

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-15

•

Full duplex terminal driver (with all options)

•

Large (20K) Executive

If you choose the standard function system, you will not be able to change your maxi-

mum node address dynamically because the standard function system does not support
the required system checkpointing file. The options in the list are discussed in the following sections. If you do not want all features of the standard function system, answer
the question NO and specify the features you do want.
3.2.5.4 Network Command Terminal Support - If you want to support network
command terminals, answer YES to Question 32A of the Executive Options section of
SYSGEN:
>* 32A. Include DECnet Network Command Terminal Support? [YIN]:
3.2.5.5 System-controlled Partitions - As a general rule, you should specify system-controlled partitions. You can do so by answering YES to Question 33 of the Executive Options section of SYSGEN:
>* 33. System control led partitions? [YIN]:
System-controlled partitions allow most network tasks to be loaded into one partition
(GEN), and automatically swapped in and out of memory. The only situation in which
you might not want to use system-controlled partitions is in a dedicated system in
which one task is loaded into each partition, thus ensuring that the task is always in
memory. Each such partition should be sized exactly for the particular task. Such a system would save the small amount of memory used by the Executive to handle systemcontrolled partitions.
3.2.5.6 Checkpointing Support (RSX-11 M Only) - The following parameters that
you specify in SYSGEN will affect the performance of specific DECnet components
only.
>* 49. Checkpointing [YIN]:
Answering YES enables checkpoint support. You should select checkpointing in some
form.
If you want to be able to change the number of nodes in your network without rebuild-

ing the routing control process (RCP), select dynamic checkpoint allocation support.
Question 50. in the Executive Options section of SYSGEN displays:
>* 50. Checkpointing with system checkpoint fi Ie [YIN]:

Answer YES if you want dynamic checkpoint support. If you select dynamic
checkpointing support and you execute the CFE DEFINE EXECUTOR MAXIMUM
NODE ADDRESS command, RCP will automatically extend itself to the required size
when it is loaded. See the DECnet-RSX Guide to Network Management Utilities for
more information on this CFE command.

3-16

DECnet-RSX Network Generation and Installation Guide

3.2.5.7 Terminal Driver Support - TLK, RMT, andX.29 will provide different capabilities depending on the options selected for the terminal driver. The following section discusses these options.
Support for TLK: The Terminal Communications utility (TLK) allows you to exchange
messages between terminals at different nodes in a network. There are two modes of
TLK - dialog mode and single message mode.
Question 1 of the Terminal Driver Options section of SYSGEN displays:
>* 1. Terminal driver desired (A/B/C/o, * prints table) [S]:
If you enter an asterisk (*), SYSGEN prints the following table:

Terminal drivers: A- Tai lorable half duplex
B- Tai lored half duplex (user oriented)
C- F u I I d u pie x
0- Basel ine half duplex

Answering A, B, or C produces satisfactory results. (If you are a PSI X.29 user, however, see Support for PSI X. 29, below.) Answering B automatically provides the necessary support for both versions of TLK. If you answer A or C, you must select the specific driver features desired. In either case (A or C), if you want either mode of TLK,
you must answer YES to the following question:
>* 8. Breakthrough write? [YIN]:

Additionally, if you want the dialog mode of TLK, you must answer YES to the following question:
>* 17. Read after prompt? [YIN]:

If you want the video dialog mode of TLK, and if you answered A or C to Question 1.

above, you must also answer YES to the following question in the same section of
SYSGEN:
>* 6. Unsol icited input character AST?

[YIN]:

Asynchronous system traps (ASTs) are software interrupts. Answering this question
YES allows your terminal to process characters entered on the keyboard at the same
time characters are being printed on the screen. Support for unsolicited input character
ASTs is automatically included if you answer B to Question 1. For more information on
TLK, see the DECnet-RSX Guide to User Utilities.

Support for RMT: The Remote Terminal utility (RMT) works with any of the terminal
drivers noted under TLK. However, it is recommended that you choose A, B, or C. Performance varies depending upon which terminal drivers are installed on the source and
the host nodes. For more information, see the discussion of RMT in the DECnet-RSX
Guide to User Utilities.

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-17

Support for PSI X.29: If you want support for PSI, you must answer C to the terminal
driver question (Question 1) noted under Support for TLK, above. PSI's X29ACP
requires the full duplex driver. You must also respond YES to Question 12:
>* 12. Set multiple characteristics? [YIN]:

in order to create MeR support for SET commands relating to terminals, such as:
SET IVT100=TI:
or
SET ITERM=TI:VT2XX

3.2.5.8 Queued File Access Requests (RSX-11 M only) - FTS allows file access
requests (file transfers, etc.) to be queued and processed in sequence. If you want FTS,
the RSX-11M Queue Manager (QMG) you must answer YES to the following question
in the Executive Options section of SYSGEN.
>* 35. Queue Manager and queued print spooler? [YIN]:

For more information on FTS, see the DECnet-RSX Guide to User Utilities.

3.2.5.9 Large or Small Executive - The RSX Executive can be built in either 16K or
20K of address space. In the Executive Option section of Phase 1, SYSGEN displays the
following question:
>* 22. Large (20K) Executive? [YIN]:

If you answer NO to this question, you will get the 16K Executive.

Your response to this question determines the amount of Executive address space allocated to the Executive plus DSR (pool). The size of the Executive itself is determined
solely by your answers to questions about Executive options. The remaining space is
allocated to pool. This allocation, however, can be altered by editing the
SYSVMR.CMD file (see Section 3.2.5.10).
Deciding which size Executive to choose must take into account your type of system,
its size, and its intended use. In general, for medium to large systems you want the 20K
Executive. The extra pool is useful, and you have enough memory so that the 4K allocated to pool does not significantly affect your ability to run tasks.
On small systems, you must decide between need for pool and need for extra task
space. You may decide to use the smaller Executive, since task space is often at a
greater premium than pool.
These tradeoff decisions depend upon network loading, types of programs being run,
and user program load. For example, if a node is being used only as a routing node, the
entire pool may not be used. In such a case performance would be improved by using
the 16K Executive, since there is less checkpointing of network tasks. You can also
reduce the Executive-plus-pool size to less than 16K (Section 3.2.5.11).

3-18

DECnet-RSX Network Generation and Installation Guide

3.2.5.10 Editing SYSVMR.CMD - It is necessary to add at least one partition,
CEXPAR, to the system in order to run DECnet. For medium size systems, you may also
want to add another partition as well (see Section 3.2.1.2). These partitions are added
by editing the SYSVMR.CMD command file.
In Phase 2 of SYSGEN, VMR Question I displays:

>* 1. Edit SYSVMR.CMO? [YIN]:
If you answer this YES, SYSGEN will pause and allow you to edit the file. If you answer

NO or press < RET> , you can edit the file later.
SYSVMR. CMD is an indirect command file used as input to the Virtual Monitor Console
Routine (VMR). VMR allows you to change the system image file used to boot the operating system. VMR commands are similar in format to those of MCR; the SYSVMR.CMD
file contains commands such as SET, LOA, and INS. When VMR executes these commands, it simulates their effect on the system image file. Thus, when the system is
booted, the commands have effectively been performed. (For more information on
VMR, see the SYSGEN manual for your system. Also, the RSX-llMIM-PLUS System
Management Guide contains reference information on VMR.)
The SYSVMR.CMD file is found on VIC [1,54] for RSX-I 1M. For RSX-IIS systems, you
must create a copy on VIC [1,64 ] (see Section 3.2.5. II). The first line of the file is the
name of the system image file that SYSVMR.CMD modifies. SET commands that define
partitions, follow that line.

3.2.5.11 Adding CEXPAR - The necessary partition (CEXPAR) is incorporated by
adding a SET IMAIN = CEXPAR line to the SYSVMR. CMD file.
The beginning of a typical SYSVMR.CMD file from an RSX-IIM system is shown
below. (A sample RSX-IIS SYSVMR.CMD file, with the appropriate lines to create
CEXPAR, is shown at the end of this section.)

RSX11M
SET IPOOL=*
SET IMAIN=EXCOM1:*:200:COM
INS EXCOM1
SET IMAIN=EXCOM2:*:200:COM
INS EXCOM2
SET IMAIN=LORPAR:*:26:TASK
INS LOR
FIX LOR ...
SET IMAIN=TTPAR:*:400:TASK
LOA TT:
SET IMAIN=ORVPAR:*:*:SYS

Getting Ready for RSX-11M/M-PLUS NETGEN

3-19

Add a line to the SYSVMR.CMD file between the SET /POOL = * command and the
SET /MAIN command for the first partition shown. To reflect the base address at
which CEXPAR loads, change the value of the SET /POOL = * line. These changes
cause CEXPAR to load immediately above the RSX Executive and pool. The format for
the two lines is:
SET IPOOL=base address
SET IMAIN=CEXPAR:*:s~e:COM

The asterisk (*) here means, load at the next available location. The word COM signifies
that this is a common block partition.
If VMR has previously been run, you must make a new copy of the system image before

running VMR to add CEXPAR. To do so, enter one of the following commands:
PIP RSXllS.SYS/CO/BL:n.=[1,64]RSXllS.TSK
PIP RSXIIM.SYS/CO/BL:n.=[1,54]RSXllM.TSK
If, in NETGEN, you specified the default values given in NET Section 4, Questions 1.00
and 1.01, use the base address values and size values given in Table 3-2. If you did not
specify the default values, make sure the values you use coincide with your response to
Questions 1.00 and 1.01.

Table 3-2:

CEXPAR Base Addresses and Sizes

Executive
(in words)

CEXSize*

Base Address * *

Par Size* *

20K

large

1130

20K

small

1135

16K

large

730

16K

small

735

* CEX size is determined as follows: CEX is automatically generated in one of two sizes by
NETGEN. For a DEC net only (that is, non-DECnet/PSI) system with only one line, NETGEN generates a small CEX. In all other cases, NETGEN generates a large CEX.

* * The base address and size are shown in octal bytes. Note that the final 2 digits of the address
and size are omitted in this table and in the discussion that follows, because that is the convention employed by VMR. Thus, an octal byte address of 112000 is' expressed as 1120. (In
NETGEN, however, you must express the last 2 digits.)

These sizes fill the Executive address space, using 20K or 16K words. The size of 43 or
50 (that is, 4300 or 5000 octal bytes) is an estimate ofCEX size, which varies somewhat
depending upon your configuration. Extra space is not wasted, but is used for Communications Executive pool.

3-20

DECnet-RSX Network Generation and Installation Guide

For RSX-11M systems, these values are usually adequate. For very small RSX-11S systems, or systems in which there is a strong need for DSR pool, you may want to make
the size of CEXPAR even smaller. To do so, specify a higher base address and a smaller
size for the CEXPAR partition (NET Section 4 Questions 1.00 and 1.01). You can run
NETGEN in restore mode to simplify this procedure.
In addition, for both RSX-11M and RSX-11 S systems, you can choose to load CEXPAR
at a lower base address. This causes other partitions to be loaded at correspondingly
lower base addresses. Loading other partitions below the defined end of the Executive
(16K or 20K) in this way is appropriate for single-user systems or for dedicated applications where a limited amount of pool is sufficient.
For an RSX-11S system, you must create a SYSVMR.CMD file on VIC [1,64]. For more
information about the RSX-11S SYSVMR.CMD file, see the RSX-I1S System Generation and Installation Guide. The following is a sample of an RSX-11 S SYSVMR. CMD
file that creates CEXPAR.

RSX11S
SET IPOOL=730
SET IMAIN=CEXPAR:*:50:COM
SET IMAIN=TTPAR:*:400:TASK
SET IMAIN=SYSPAR:*:40:TASK
SET IMAIN=MCRPAR:*:140:TASK
SET IMAIN=RSDV52:*:400:TASK
SET IMAIN=GEN:*:*:SYS
LOA TT:
INS TKN/FIX=YES
INS BASMCR/FIX=YES
PAR
TAS
After invoking VMR, you should still enter the time and date after running
SYSVMR. CMD as an indirect command file. The command PAR prints out a list of partitions, and the command TAS prints out a list of tasks.

3.2.5.12 Adding Other Partitions - You may have chosen, after reading the information provided in Section 3.2.1.2, to put NETACP in its own partition, here called
ACPPAR. To create the ACPPAR partition, edit SYSVMR.CMD and insert a SET IMAIN
command just before the SET IMAIN command that defines the partition GEN. The line
to insert takes the following form:

SET IMAIN:ACPPAR:*:s~e: [SYS]
[TASK]
Here the asterisk (*) indicates that ACPPAR is loaded in the next available space, and
size specifies the size of the largest task to occupy the partition. SYS indicates that this
is a system-controlled partition. If you did not select system-controlled partitions, use
the word TASK instead.
A size of 270 is generally adequate (again, the final 2 octal digits are implied).

Ge.tting Ready for RSX-11 M/M-PLUS NETGEN

3-21

3.2.6 RSX-11 S System Considerations
Because RSX-llS systems are often dedicated systems, it is difficult to establish specific rules for determining partition layout. The following are required procedures:
•

You must have the' 'floating" device CSRs at the proper addresses on the processor
to be loaded. The proper addresses are determined by the standard UNIBUS/QBUS
device rankings. A command file, [200,200]FLOAT.CMD, is provided on the kit to
aid in determining the CSR addresses.

•

You must add the partition CEXPAR (see Section 3.2.5.11).

•

It is recommended that you use system controlled partitions. The host system can
load all required tasks into GEN with system-controlled partitions and standard partition layout.

•

Determine if all desired tasks will fit in memory. Examine the maps to determine
sizes. Tasks that are not required all the time can be down-line loaded. This allows
several tasks to share the same partition. Refer to Section 3.2.1 to determine minimum network requirements.

3.2.6.1 Down-line Loading to a DMP/DMV Tributary - When down-line loading
an RSXl1-S system image from a multipoint master to a DMP/DMV tributary, the tributary address must be set in the switches on the DMP/DMV device on the target system.
Once the RSX-l1S system is down-line loaded, it will continue to use the tributary
address that is set in the switches on the device. This allows many tributaries to be
down-line loaded from the same system image, while each retains its unique tributary
address. If you want the down-line loaded system to override the tributary address on
the device, then - prior to NETGEN - you must edit the [137,10]DECPRM.CMD
file. The value $DOVER must be changed to TRUE.
3.2.6.2 Support for SLD - If you wish to use the satellite task loader (SLD) to perform down-line task loading on RSX-llS systems, you must edit the RSXMC.MAC file
to include two symbols: D$$ISK and C$$CKP. D$$ISK includes task loader support:
C$$CKP creates checkpointing support.
The procedure is as follows: In SYSGEN Phase I, Question 1 in the "Create Executive
Build Files" section displays:
>* 1. Do you wish to edit any of the Executive fi les? [YIN]:

Answer this question YES. SYSGEN then pauses to allow you to edit any file. You can
use the editor of your choice. Edit the RSXMC.MAC file on UIC [11,10]. First search for
the symbol A$$CPS. Then below that symbol, add the following symbols to the
RSXMC.MAC file:

C$$CKP=O
D$$ISK=O

;CHECKPOINTING SUPPORT
;TASK LOADER SUPPORT

3-22

DECnet-RSX Network Generation and Installation Guide

3.2.6.3 Including NETPAN (Up-line Panic Dump) - NETPAN allows a DECnet-11S
node to dump the contents of its memory to an adjacent DECnet-11M or -11M-PLUS
node. If you intend to include NET PAN in your node, and if your hardware configuration meets the specifications outlined in Appendix section B.1.2, you must edit the
RSXMC.MAC file to define one or more symbols. If your up-line dump device is an
Ethernet device, you must define N$$NI; if you want the system to restart from the system boot ROM, you must define R$$BOT. See Appendix B for details on the SYSGEN
requirements that must be satisfied in order to include NETPAN in your node.

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-23

3.3

PREGEN

PREGEN is a preliminary procedure you may have to perform before generating a network. It merges your required and optional distribution media so they can be used by
the NETGEN procedure. PREGEN can include the following distribution kits (disks or
magnetic tapes):
•

Network kit

•

DECnetkit

•

PSI kit (optional)

Each kit contains command files and object files. The command files become the corresponding parts of NETGEN: NET, DEC, and PSI. The object files are used to build the
appropriate NET, DEC, or PSI components.
PREGEN copies the command files from each distribution kit to the NETGEN disk. This
disk becomes the input to NETGEN, and allows NETGEN, once started, to display the
NETGEN questions. PREGEN copies the files listed in Table 3-3.
Table 3-3:

PREGEN Files

File

Contents

NETGEN command files

Command files from all available kits

Network object files

Object files for the NET procedure of NETGEN

DECnet object files

Object files for the DEC procedure of NETGEN

PSI object files

Object files for the PSI procedure of NETGEN

The type of media you have purchased and the type of system you are generating determines whether or not you have to perform the PREGEN procedure.

3.3.1

Who Must Perform PREGEN

If you have purchased an RL02, RK06, or RK07 distribution kit and you are generating

a DECnet-only system, PREGEN is not required. Your distribution kit contains all the
files you will need to run NETGEN (1 disk). In addition, if you have purchased an RL02,
RK06, or RK07 distribution kit for a combined DECnet/PSI system but want to generate
a DEC net only system, PREGEN is not required. If you do not have to perform
PREG EN, you should make a copy of your distribution media and run the NETGEN procedure from this copy. If you do not have to perform PREGEN, you can skip ahead to
Chapter 4.
PREGEN is required if any of the following conditions apply:
•

You have purchased a magnetic tape distribution kit.

•

You have purchased an RLO 1 distribution kit.

•

You are generating a combined DECnet and PSI system.

•

You are generating a PSI-only system.

3-24

DECnet-RSX Network Generation and Installation Guide

3.3.2 Two Types of PREGEN
PREGEN can be run for either a standard system or a small dual-disk system. A standard
PREGEN is performed for systems other than small dual-disk systems. A small dual-disk
PREGEN is performed for a system that has only two small disk drives (drives such as
RLOls, RL02s, RK06s, or RK07s), one of which serves as a system disk. A standard
PREGEN and a small dual-disk PREGEN are started the same way (see Section 3.3.6).

3.3.3

Overview of Standard PREGEN Operation

PREGEN first asks if you want to see the PREGEN notes, and if you do, displays them.
PREGEN then asks if you are running on a small dual-disk system. If you answer YES,
you get the small dual-disk PREGEN. If you answer NO, you get the standard PREGEN.
The remainder of PREGEN is performed once each for the network, DECnet, and
(optionally) PSI distribution kits. For each pass through the standard PREGEN procedure, the following occurs:
•

PREGEN asks if you want to copy the distribution kit. It does not ask you this question for the first kit, because you must copy the network distribution kit. Answer
the question YES for each distribution kit you have.

•

PREGEN then displays several questions about the distribution kit disk or tape.

•

PREGEN asks for the device on which the NETGEN disk is loaded and then asks several related questions.

•

PREGEN copies the command files from the distribution kit to the NETGEN disk.

•

PREGEN asks if you also want to place the object files on the NETGEN disk (Question 7.00). If you answer YES, PREGEN copies the object files and then skips ahead
to the next pass.

•

If you answer NO to Question 7.00, PREGEN asks you to specify the disk to which
the object files should be moved. PREGEN then asks several questions about the
object disk.

•

PREGEN copies the object files to the object disk, and then goes on to the next pass.

3.3.4 Overview of Small Dual-disk PREGEN Operation
There is not enough space on a small dual-disk system to copy all the object files from
the distribution kits to other disks. Each kit is used for the build phase of NETGEN.
Mount each distribution kit during PREGEN, then mount each kit again during
NETGEN.

Getting Ready for RSX-11 M/M-PLUS N ETG EN

3-25

For a small dual-disk system, PREGEN first asks if you want to see the PREGEN notes
and, if you do, displays them. PREGEN then asks if you are running on a small dual-disk
system. If you answer YES, you get the small dual-disk PREGEN. If you answer NO, you
get the standard PREGEN. The following is repeated once each for the network,
DECnet, and (optionally) PSI distribution kits. For each pass through the procedure,
the following occurs:
•

•

PREGEN then displays several questions about the distribution kit disk.

•

PREGEN copies the command files for each distribution kit to the system disk.

When the above process is completed, PREGEN asks for the device on which the
NETGEN disk is loaded and then asks several related questions. Finally, PREGEN copies
the command files from the system disk to the NETGEN disk, and then deletes the command files from the system disk.

3.3.5

Copying Distribution Media to Disks

If you have purchased distribution media in magtape form, you need the following:
Drives

Media

1 system disk drive

System disk

1 magnetic tape drive

NET kit
DEC kit
PSI kit (optional)

1 free disk drive

1 free disk (NETGEN disk)
or multiple disks

You cannot use tape distribution media for a small dual-disk PREGEN.
If you have purchased distribution media in disk form and you are on a system that is

not a small dual-disk system, you need the following for a standard PREGEN:
Drives

Disks

1 system disk drive

System disk

1 free disk drive

NET kit
DEC kit
PSI kit (optional)

1 free disk drive

1 free disk (NETGEN disk)
or multiple disks

3-26

DECnet-RSX Network Generation and Installation Guide

If you have purchased distribution media in disk form and you are on a system that is a

small dual-disk system, you need the following for a small dual-disk PREGEN:
Drives

Disks

1 system disk drive

System disk

1 free disk drive

NET kit

DEC kit

PSI kit (optional)
1 free disk (NETGEN disk)
With any output disk (except those with very large capacity such as RP05 or RP06), it is
recommended that PREGEN initialize the disk to ensure that the maximum capacity is
available for the distribution kit files. If you want PREGEN to initialize the disk, do not
mount it but make sure you run the Bad Block Locator utility (BAD) on any disks initialized by PREGEN before running PREGEN. If you do not want PREGEN to initialize the
disk, it does not matter whether you mount it or not.

3.3.6

Starting PREGEN

Depending upon the type of distribution kit you purchased, use one of the two procedures described here to start PREGEN. To run the PREGEN procedure under VAX/VMS
you need the same privileges and terminal characteristics required to perform a
NETGEN - see Section 4.2.2.
3.3.6.1 Disk Distribution Media - For distribution kits in the form of disks, use the
following procedure to start PREGEN:
1. Allocate the disk drive for the NETKIT (network kit) disk:
>ALL ddx:

Mount the NETKIT disk:

>MOU ddx: NETKIT

3. Using PIP, copy the PREGEN command file from the NETKIT disk (ddx:) to a disk
that will be available throughout PREGEN:
>PIP tdu:/NV=ddx: [137,lO]PREGEN.CMD
4. Allocate all other disk drives to be used:
>ALL ddu:

Execute PREG EN:
>@tdu : PREGEN

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-27

3.3.6.2 Magnetic Tape Distribution Media - For distribution kits in the form of
magnetic tapes, use the following procedure:
1.

Allocate the NETKIT (network kit) tape drive:
>ALL mmu:

Mount the NETKIT tape:
>MOU mmu: I FORE I GN

3. Using FLX, copy the PREGEN command file from the NETKIT tape to a disk that
will be available throughout PREG EN:
>FLX tdu: IUI=mmu: [137,10] PREGEN. CMO/RW[/ONS: n]

where n is either 800 or 1600, depending on which density tape you are using. The
default is 800, so if your tape density is 800, you can omit IDNS:n.
4.

Allocate all other disk drives to be used:
>ALL ddx

Execute PREGEN:
>@tdu : PREGEN

3-28

DECnet-RSX Network Generation and Installation Guide

3.3.7

Understanding the PREGEN Dialog

The following sections present the PREGEN dialog. Your response to Question 2.00
determines whether you will run a standard PREGEN or a small dual-disk PREGEN.
NOTE

The explanatory text after each question in the following, is a word-by-word
representation of what you will see on your screen when you perform a
PREG EN. It may not, however, be a line-by -line representation of what appears
on your screen.
>;

=============================================================

>; PREGEN - RSX-IIM/S/M-PLUS CEX System Pre-Generation Procedure
>;
Started at time on date
>; Copyright (C) 1981, 1982, 1983, 1985 by
>; Digital Equipment Corporation, Maynard, Mass.
>* 01.00 Do you wish to see the PREGEN notes? [YIN]:
Answer this question YES if you are not fami I iar with running
PREGEN.
Before starting PREGEN, you must have al located al I requi red
disk and/or tape drives which are not publ ic devices

If an output disk is al ready mounted, PREGEN wi I I use the disk
as is. If it is not mounted, PREGEN wi I I provide the option
of mounting the disk either with or without re-initiallzing
it.
The fol lowing tasks may be requi red:
PIP (if using disk distribution media).
FLX (if using magtape distribution media).
UFO (if any necessary UICs do not already exist).
If you are running on a system with only one free disk drive
(other than the drive used for your system diSk), and you are
using disk distribution media, your system disk must have
at least 5000 free blocks.
To obtain help for any question, type <ESC>. To exit or to
temporari Iy stop PREGEN execution, type <CTRL/Z>.
The disk in ddu: must remain mounted throughout the
entire PREGEN procedure.
For questions that can be answered YES or NO, the default
is NO unless otherwise specified. For questions with
defaults, the default is produced by pressing a carriage
return <RET> in response to the question.

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-29

In the notes, ddu: refers to the disk drive from which you are running this procedure. If
PIP, FLX, and UFD are not installed, PREGEN installs them and displays a message to
indicate this.
>* 02.00 Are you running on a smal I dual-disk system [YIN]:

A smal I dual-disk system is one which has only two sma I I disk
drives, such as RL01s, RL02s, RK06s or RK07s.
If you are running on a sma I I dual-dIsk system, answer this
question YES.
If your system has two free disk drives (other than the drive
used for your system disk), or if you are using magtape
distribution media, answer this question NO.
If you are running on a dual-disk system whIch has a large
system disk with at least 25000 free blocks, you may answer
either NO or YES. If you answer NO, all network generation
command fi les and object fi les wi II be placed on your system
dIsk. If you answer YES, the network generatIon command fi les
wi I I be placed on some disk other than your system disk (the
object fi les wi I I remain on the dIstrIbution kits, which must
be swapped with the NETGEN disk during the network generation).
If you answer NO and you are generating a DECnet-only system (a DECnet-RSX system

without PSI), 20,000 free blocks is sufficient.
If you answer YES, approximately 5000 blocks of free space are
required for temporary storage on your system disk. UIC
[137,10] wi I I be used for this purpose, and should not be
used for user f i I es. I f you do not have 5000 free blocks,
create a copy of your system disk and delete fi les such as maps,
I istings, command fi les, sources (other than RSXMC.MAC) , and
unnecessary task images.
If you answer YES and you are generating a DECnet-only system (a DECnet-RSX sys-

tem without PSI), 3000 free blocks is sufficient.
In addition, if you answer YES, your distribution kIt disks wIll
contain the object fi les used during the task bui Id phase of
NETGEN, and must be loaded as appropriate objects are needed to
bu i I d network tasks. For th i s reason, it is suggested that you
back up the distribution kits prior to performing the generation.

NOTE
If you answer NO to Question 2.00, the dialog shown in Section 3.3.8 appears.
If you answer YES to Question 2.00, the dialog shown in Section 3.3.9 appears.

3-30

DECnet-RSX Network Generation and Installation Guide

3.3.8

Standard PREGEN

PREGEN displays Question 3.00 for all kits except NETKIT because NET KIT is always
required.
>* 03.00 Copy the network distribution kit? [YIN]:

Answer this question YES if you have this distribution kit, and
you wi I I be generat i ng a network network.
>* 04.00 Where is the network distribution kit

loaded? [S]:

Specify the disk or tape drive where the distribution kit is or
wi I, I be loaded (e. g. DK1 0 r MMO).
>* 04.01 Is the disk (or tape) al ready loaded in ddu:?

[YIN]:

Answer this question YES when the distribution kit is loaded
and ready to be mount ed. PREGEN wi I I act ua I I y mount t he disk
(or tape).
>* 04.02 Is the tape 1600 BPI? [YIN]:

Answer this question YES if the distribution tape is encoded
with 1600 BPI (bits per Inch). Answer NO if the tape is
800 BPI.
>* 05.00 Where is the NETGEN disk loaded [S]:

The NETGEN disk, created by PREGEN, is the primary disk used
during a network generation. It wi II contain all cOlTlTland fi les,
from all distribut ion kits, which are necessary during the
questlonlanswer phase of the generation.
I f there is enough room on th i s disk, the data f i I es, obj ect
files, and task images which wi I I be used during the task
bUild phase of the generation may also be placed on this disk.
Specify the disk drive where the NETGEN disk is or wi I I be
loaded (e.g. DMO).
>* 05.01 Is the disk already loaded in ddu: (NETGEN disk drive)? [YIN]:

Answer this question YES when the NETGEN disk is loaded and
ready to be mounted. PREGEN wi I I actually mount the disk if
it IS not already mounted.
>* 06.00 Should the NETGEN disk be initial ized? [YIN]:

Answer this question YES If the disk should be Initial ized.
This wi I I place an initial fi Ie system on the disk, and
destroy any fi les that may have been on the disk previously.
If the disk has al ready been properly initial ized, answer NO.
If you answer YES, the disk wi II be given the label "NETGEN".

PREGEN displays Question 6.00 only if the NETGEN disk has not already been
mounted; and displays the following statement if you answer YES to Question 6.00.

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-31

The disk wi II be initial ized with the label "NETGEN".
>* 06.01 What

is the label used to mount the NETGEN disk [S]:

Specify the label PREGEN is to use when mounting the previously
initial ized NETGEN disk. This should be from 0 to 12
alphanumeric characters.

PREGEN displays Question 6.01 if you answer NO to Question 6.00. At this point, the
command files are copied to the NETGEN disk.
>* 07.00 Should the network object fi les be moved to the

NETGEN disk?[Y/N]
The object fi les are those fi les which are used by NETGEN only
during the bUild phase, when al I of the generation questions
have been answered, and the processes and tasks are going to
be bu i It.
If the object fi les are on the NETGEN disk, they are
immediately avai lable for use by NETGEN. If they are
on a separate Object disk, NETGEN wi I I have to wait unti I
that disk is ready to be mounted before it can bui Id the
tasks for that product. For systems with a I imited number of
disk drives, this may also involve dismounting and unloading
some other disk (such as the NETGEN disk).
If the NETGEN disk has enough free space, you may wish to
move al I of the object fi les to the NETGEN disk in order to
minimize the time needed to perform a network generation. If
so, answer this question YES. If there is not enough free
space or if you are not sure, answer NO.
The number of free blocks on the NETGEN disk is:
free block information

PREGEN will display information such as the following.
DMO: has 24080. blocks free, 3046. blocks used out of 27126.
La rgest cont i guous space = 11160. blocks
1652. f i Ie heade rs are free, 16. heade rs used out of 1668.

During this time, PREGEN executes a PIP ddu:IFR command, where ddu: is the
NETGEN disk drive.
The approximate number of blocks needed for the network objects
is:
number

If you answer Question 7.00 YES, PREGEN copies the files to the NETGEN disk and
skips the remaining questions in this pass. If you answer NO, PREGEN displays the

remaining questions.
For DECnet-only systems (RSX-DECnet without PSI), all initialized disks have sufficient capacity to hold all of the object files. For DECnet/PSI systems, all initialized disks
have sufficient capacity (except RL01).

3-32

DECnet-RSX Network Generation and Installation Guide

When performing an RLO 1 PREG EN for a D ECnet/PSI system, it is recommended that
you place the network object files on the NETGEN disk and place the DEC net and PSI
object files on a separate disk.
>* 08.00 Where is the network Object disk loaded? [S]:

This Object disk, created by PREGEN, wi I I contain al I object,
data and task fi les which are necessary during the phase of the
network generation which bui Ids the network product.
This Object disk may be a separate disk, or it may be a disk
which already holds other object files (or any other files).
Specify the disk drive where the Object disk is or wi I I be
loaded (e.g. DMO).
>* 08.01 Is the disk a I ready loaded in ddu:? [YIN] :

Answer this question YES when the Object disk is loaded and
ready to be mounted. PREGEN wi I I actually mount the disk if it
is not al ready mounted.

In Question 8.01,ddu: refers to the Object disk drive specified in Question 8.00.
>* 09.00 Should the Object disk be initialized? [YIN]:

Answer this question YES if the disk should be initial ized.
This wi I I place an initial fi Ie system on the disk, and destroy
any fi les that may have been on the disk previously. If the
disk has already been properly initialized, answer NO.
If you answer YES, the disk wi II be given the label "netobj".

PREGEN displays Question 9.00 only if the disk has not already been mounted. In successive passes, PREGEN replaces netobj with decobj and psiobj.
PREGEN displays the following statement if you answer Question 9.00 YES.
The dis k w i I I be i nit i a liz e d wit h the I abe I "netobj".
>* 09.01 What is the label used to mount the Object disk? [S]:

Specify the label PREGEN is to use when mounting the previously
initial ized Object disk. This should be from 0 to 12
alphanumeric characters.

PREGEN displays Question 9.01 if you answer Question 9.00 NO.
At this point, the object files are copied to the Object disk. This completes one pass of
PREGEN. If all distribution kits are complete, PREGEN stops here and displays the following statement. If not, PREGEN returns to Question 3.00 and displays questions for
the next distribution kit.
>; ========================================================
>; PREGEN - Stopped at time on date
>; ========================================================

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-33

3.3.9

PREGEN for a Small Dual-disk System

If you answer YES to Question 2.00, PREGEN displays the following dialog.
>* 03.00 Copy the DECnet distribution kit? [YIN]:

Answer this question YES if you have this distribution
kit, and you wi I I be gene rat i ng a DECnet netwo rk.
>* 04.00 Where

s the network d i st r i but ion kit loaded? [S]:

Specify the disk or tape drive where the distribution kit is or
wi I I be loaded (e.g. OK1 or MMO).
>* 04.01 Is the disk already loaded in ddu:? [YIN]:

Answer this question YES when the distribution kit IS loaded
and ready to be mounted. PREGEN wi I I actually mount the disk.

At this point, PREGEN copies the command files from the current distribution kit to the
system disk, which serves as a temporary storage device. If the kit just copied was the
network or the DECnet distribution kit, PREGEN returns to Question 3.00 to display
the same questions for the next distribution kit. When all required kits are copied,
PREGEN displays Question 5.00.
>* 05.00 Where is the NETGEN disk loaded? [S]:

The NETGEN disk, created by PREGEN, is the primary disk used
during a network generation. It Wi II contain all command fi les,
from al I distribution kits, which are necessary during the
questionlanswer phase of the generation.
Specify the disk drive where the NETGEN disk is or wi I I be
loaded (e.g. OMO).
>* 05.01 Is the disk a I ready loaded in ddu: (netgen disk drive)? [YIN]:

Answer this question YES when the NETGEN disk IS loaded and
ready to be mounted. PREGEN wi I I actually mount the disk if
it is not already mounted.
>* 06.00 Should the NETGEN disk be initial ized? [YIN]:

Answer this question YES if the disk should be initial ized.
This wi I I place an initial fi Ie system on the disk, and destroy
any f i I est hat rna y ha v e bee non the dis k pre v i 0 us I y. 1ft he
disk has al ready been properly initial ized, answer NO.
If you answer YES, the disk wi II be given the label "NETGEN".

PREGEN displays Question 6.00 only if the NETGEN disk has not already been
mounted; it displays the following statement if you answer YES to Question 6.00.

3-34

DECnet-RSX Network Generation and Installation Guide

The disk wi II be initial ized with the label "NETGEN".
>*

06.01 What is the label used to mount the NETGEN disk [S]:

Specify the label PREGEN is to use when mounting the previously
initial ized NETGEN disk. This should be from 0 to 12
alphanumeric characters.

PREGEN displays Question 6.01 if you answer NO to Question 6.00.
At this point, the command files stored on the system disk are copied to the NETGEN
disk and then deleted from the system disk. This completes PREGEN.
>;
>;
>;

============================================================
PREGEN - Stopped at time on date
============================================================

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-35

3.3.10

Sample Standard PREGEN

The following is a sample of a standard PREGEN. Not all of the COpy commands will
be executed for all systems. Where no response is indicated to PREGEN questions, the
default value has been selected by pressing < RET>
>SET IUIC=[137,10]
>@PREGEN
>;

>; =============================================================
>; PREGEN - RSX-IIM/S/M-PLUS CEX System Pre-Generation Procedure
>;
Sta rted at time on date
>; =============================================================
>;

>; Copyright (C) 1981, 1982, 1983, 1985 by
>; Digital Equipment Corporation, Maynard, Mass.
>;

>*
>*
>*
>*
>*
>*
>*

01.00 Do you wish to see the PREGEN notes? [YIN]:
02.00 Are you running on a smal I dual-disk system? [YIN]:
04.00 Where IS the Network distribution kit loaded? [S]: DL:
04.01 Is the disk al ready loaded in DL:? [YIN]: Y
05.00 Where is the NETGEN disk loaded? [S]: DMl:
05.01 Is the disk already loaded in DMl:? [YIN]: Y
06.00 Should the NETGEN disk be initial ized? [YIN]: Y

>; The disk wi II be initial ized with the label "NETGEN".
>;

>; Copying UIC "[137,10]".
>; Copying UIC "[1,2]".
>;

>* 07.00 Should the Networkobject files be moved tothe NETGEN disk? [Y/N]:Y

Copy commands and DICs for all Network Object Files are listed in the following format:
Co pyin g UI C "[ uic] "
>;
>;

DMO -- TTO: dismounted from DLO: *** Final dismount initiated ***
16:13:28 *** DLO: -- Dismount complete
>;

>* 03.00 Copy the DECnet distribution kit? [YIN]: Y
>* 04.00 Where is the DECnet distribution kit loaded? [S]: DL:
>* 04.01 Is the disk al ready loaded in DL:? [YIN]: Y
>;

>; Copying UIC "[137,10]".
>;

>* 07.00 Shou I d the DECnet ob j ect f i I es be moved to the NETGEN disk? [YIN] : Y

3-36

DECnet-RSX Network Generation and Installation Guide

Copy commands and VICs for all DECnet Object Files are listed in the following format:
Copy i ng UIC "[uic]"
>;
>;

DMO -- TTO: dismounted from DLO: *** Final dismount initiated ***
16:27:52 *** DLO: -- Dismount complete
>;
>;

>* 03.00 Copy the PSI distribution kit? [YIN]: Y
>* 04.00 Where is the PSI distribution kit loaded? [S]: OL:
>* 04.01 Is the disk already loaded in DL:? [YIN]: Y
>;

>; Copying UIC "[151,10]".
>;

>* 07.00 Shou I d the PSI obj ect f i I es be moved to the NETGEN disk?

[YIN]:

Copy commands and VICs for all PSI Object Files are listed in the following format:
>; Copy i ng UIC

"[uic]"

DMO -- TTO: dismounted from OLO: *** Final dismount initiated ***
time *** DLO: -- Dismount complete
>;
>;

DMO -- TTO: dismounted from DM1: *** Final dismount Initiated ***
time *** DM!: -- Dismount complete
>;
>;

=============================================================
>; =============================================================
>;

>; PREGEN - Stopped at time on date
>;

>@ <EOF>
>

Getting Ready for RSX-11 M/M-PLUS NETGEN

3-37

3.3.11

Sample Small Dual-disk PREGEN

The following is a sample of a small dual-disk PREGEN:
>SET IUIC=[137,10]
>@PREGEN
>;
>;

=============================================================

>; PREGEN - RSX-11M/S/M-PLUS CEX System Pre-Generation Procedure
>;
Sta rted at time on date
>;
>.,

============~================================================

>; Copyright (C) 1981, 1982, 1983, 1985 by
>; Oigital Equipment Corporation, Maynard, Mass.
>;

>* 01.00 00 you wish to see the PREGEN notes? [YIN]:
>* 02.00 Are you running on a sma I I dual-disk system? [YIN]: Y
>* 04.00 Where is the Network distribution kit loaded? [S]: OLO
>* 04.01 Is the disk already loaded in OLO:? [YIN]: Y

>.,

>; Copying UIC "[137,10]".
>;

OMO -- TT51: dismounted from OLO: *** Final dismount initiated ***
>;

>* 03.00 Copy the OECnet distribution kit? [YIN]: Y
>* 04.00 Where is the OECnet distribution kit loaded? [S]: OLO
>* 04.01 Is the disk already loaded in OLO:? [YIN]: Y
>;

>; Copying UIC "[137,10]".
>;

OMO -- TT51: dismounted from OLO: *** Final dismount
initiated ***
>;

>* 03.00 Copy the PSI distribution kit? [YIN]:
>* 05.00 Where is the NETGEN disk loaded? [S]: OLO
>* 05.01 Is the disk already loaded in OLO:? [YIN]: Y
>* 06.00 Should the NETGEN disk be initial ized? [YIN]: Y
>.,

>; The disk wi I I be initial ized with the label "NETGEN".
>;

OMO -- TT51: dismounted from OLO: *** Final dismount
initiated ***
>;
>; Copying Network f i I es
>.,
>; Copying UIC "[137,10]".
>.,
>., Copying OECnet f i I es
>.,
>; Copying UIC "[137,10]".
>;
>;

OMO -- TT51: dismounted from OLO: *** Final dismount initiated ***
>;
>;

============================================================
>; ============================================================

>; PREGEN - Stopped at time on date
>;

>@ <EOF>
>

3-38

DECnet-RSX Network Generation and Installation Guide

4
Using the RSX-11 M/M-PLUS NETGEN Procedure

This chapter describes how to use the RSX-IIMIM-PLUS NETGEN procedure.

4.1

NETGEN Parameter Defaults

During NETGEN, you can select default values for many component parameters. In
most cases you should select these default values. Later, you can modify most parameter values using NCP or CFE commands. If, however, the parameter value you wish to
change cannot be modified by the NETGEN process, by NCP or CFE commands, you
must edit the parameter file ([137,IO]DECPRM.CMD) found on the NETGEN disk prior to doing a NETGEN. This file contains documentation describing the parameters
included in the file and information about changing these parameters. Values changed
by editing this file will be applied appropriately during the NETGEN procedure.

4.2

Starting NETGEN

In order to run NETGEN, you must have access privilege to the target disk and you must
have your target disk mounted.
NOTE

NETGEN requires the large indirect command file processor. NETGEN will not
function properly with the small indirect command file processor.

4-1

4.2.1

Running NETGEN on an RSX-11 M/11 M-PLUS System

On an RSX-IIM/IIM-PLVS system, there are two ways to gain the required access
privilege:
•

•

Allocate the target disk for your exclusive use

In performing a NETGEN, you may prefer to use a hardcopy terminal to generate a
clear, permanent record.
To start NETGEN, make sure the NETGEN disk is mounted. To do this, enter the following if a PREGEN was performed.
> ALL dduu:
> MOU dduu: NETGEN

If PREGEN was not required, enter:
> ALL dduu:
> MOU dduu: DECK I T

NETGEN is performed by executing an indirect command file. Invoke NETGEN as follows:
>@dduu:[ 137,10] NETGEN

where dduu: is the NETGEN disk you just mounted. All sections of the NETGEN procedure are then automatically invoked in turn.

4.2.2

Running NETGEN on a VAX-11 RSX System

If you are doing a NETGEN on a VMS system with VAX-II RSX Version 2.0 system,

you should have access to the following required privileges:
•

Change mode to kernel privilege (CMKRNL). This allows you to change default
VICs and directories. You must have this privilege before you can set the remaining
privileges.

•

Logical I/O Privilege (LOG_IO). This privilege permits RSX-IIM VMR to do logical I/O on task image files.

•

System Protection Privilege (SYSPRV). This privilege allows you to create files in
directories not owned by the current VIC.

If you have the SETPRV privilege, you can establish the required privileges with the fol-

lowing MCR command:
> SET PROC/PR I V= (privilege [ , ... ])

To find out what privileges you have, use the following MCR command:
> SHO PROC/PRIV

For more information on using these privileges, see the VAXNMS System Manager's
Guide.

4-2

DECnet-RSX Network Generation and Installation Guide

In addition, you must use MCR as your command language interpreter (CLI) for
NETGEN. IfMCR is not your command language interpreter, then you must change to
MCR. To do this, log in using the ICLI = MCR qualifier. For example:
Username: SYSTEM/CLI=MCR
Password:
Finally, you must have your terminal set to the INOLINIL-EDITING characteristic. To
verify that your terminal is set properly, use the SHOW TERMINAL command. If the terminal has LINE_EDITING enabled, change the characteristic using the command
shown below:
SET TERMINAL/NOLINE-EDITING
If your terminal has LINE~DITING enabled, NETGEN will abort if you request a question explanation using the < ESC> key.
In performing a NETGEN, you may prefer to use a hardcopy terminal to generate a clear
permanent record.
To start NETGEN, make sure 'the NETGEN disk is mounted. To do this, enter the following if a PREG EN was performed.
> ALL dduu:
> MOU dduu: NETGEN

If PREGEN was not required, enter:
> ALL dduu:
> MOU dduu: DECK IT
NETGEN is performed by executing an indirect command file. Invoke NETGEN as follows:
>@dduu:[137,lO]NETGEN

where dduu: is the NETGEN disk you just mounted. All sections of the NETGEN procedure are then automatically invoked in turn.

4.3

NETGEN Organization

NETGEN is organized into three main parts - NET, DEC, and PSI.
•

NET displays questions relating to common components and the Communications
Executive (CEX). NET contains the general routines that support both DEC net and
PSI.

•

DEC displays DEC net-specific questions. When combined with NET, NETGEN
builds the routines that allow DECnet to run.

•

PSI displays questions specific to the Packetnet System Interface (for licensed
users). PSI allows you to access an X.25 Packet Switching Data Network (PSDN).

NET, DEC, and PSI are divided into sections with a logical grouping of related questions. Figure 4-1 illustrates the organization of the NETG EN process.
Using the RSX-11 M/M-PLUS NETGEN Procedure

4-3

TW0168

Figure 4-1 : Sections Within the NETGEN Process

4-4

DECnet-RSX Network Generation and Installation Guide

4.4

NETGEN Operating Modes and Saved Response Files

NETGEN can be run in any of the following operating modes:
•

Question and Answer Mode

•

Dry Run Mode

•

Restore Mode

•

Component Mode

The mode you select remains in effect throughout the entire generation procedure,
except question and answer mode.
In all modes except component mode, as you answer the questions, NETGEN saves
your responses in a saved response file. This file is described in Section 4.4.5.

4.4.1

Question and Answer Mode

NETGEN begins in question and answer mode. By answering the NETGEN questions,
you define and build your system. In question and answer mode, you can invoke dry
run mode, restore mode, component mode or you can stay in question and answer
mode.

4.4.2

Dry Run Mode

Dry run mode is a special case of question and answer mode. A dry run allows you to
answer all the NETGEN questions without defining or building your system; no processing is done to create a system that will run. Only the saved response file is created. This
file can later be used to perform a restore mode generation. If you want to proceed
through the NETGEN dialog without actually generating a network, invoke dry run
mode.

4.4.3

Restore Mode

If you have performed a NETGEN in question and answer mode or dry run mode, your

responses were saved in a saved response file. You can run NETGEN in restore mode to
use these responses to perform a new network generation, correct inappropriate
answers, or make minor changes. Restore mode builds or rebuilds all components. If
the saved response file does not contain a complete set of answers (due to a system
crash or early exit), restore mode will use the available answers and then switch to question and answer mode for the remainder of the NETGEN.

4.4.4

Component Mode

After you have defined and built your system, you can add or modify specific components by invoking component mode. For example, if you did not include NTDEMO in
your network generation, you can add it by running NETGEN in component mode.
Using the RSX-11 M/M-PLUS NETGEN Procedure

4-5

The questions displayed are restricted to those required for the particular component(s) you want to add or change. While restore mode rebuilds all components, component mode cannot build all components. For example, it cannot rebuild the Communications Executive or the device drivers. See Appendix C. No saved response file is
created for component mode NETGENs.

4.4.5

Saved Response Files

For all modes except component mode, a saved response file is created for each part
(NET, DEC, PSI) of NET GEN. NETGEN creates these various saved response files as you
answer questions. At the beginning of your second generation, NETGEN will give you
the option of using your saved responses from your first NETGEN.
At the beginning of your third or any succeeding NETGEN, you will be given the option
of using saved responses from your two previous generations. You can use either set of
saved response files as input files for your next generation, or you can answer all the
questions again. If you use one of these files, the set you choose will be saved and the
other set will be deleted. If you do not use either file, the newer version will be saved
and the older version will be deleted and replaced with a file containing your current
answers.
CAUTION

Do not attempt to edit the saved response files. They are coded in a form
unique to NETGEN, and any attempt to change them directly can result in a
defective network generation.

4.4.6

Duplicating Network Generations

You may want more than one network generation version, either exact versions or versions with slight differences between generations. The procedure for this type of generation differs from the usual restore mode generation.
In a restore mode generation, NETGEN overwrites any former generation and creates a
new version. If you want to keep separate versions, the necessary procedure is as follows: keep the old generation intact at the old disk location, then create a new VIC on
your target disk and move the saved response files from the old VIC to the new VIC,
and run the NETGEN procedure in restore mode. When restore mode asks for your VIC
group code, respond with your new VIC. The files you want to copy are all the files in
[xxx, 1], where xxx is the group code of your old generation.
If there is not sufficient space on your target disk, create a new target disk, copy the

saved answers from the old target disk to the new target disk, and run NETGEN in
restore mode.
In either case, NETGEN performs a new network generation based on your former
answers. As with any restore mode NETGEN, you can change your answers.

4-6

DECnet-RSX Network Generation and Installation Guide

4.5

NETGEN Dialog Format

The NETGEN dialog format consists of sections with questions, optional explanatory
text, formats for user responses, default values, and responses for end-of-section questions and break questions.
Each section is introduced with a header, as shown in the following example:
>; ============================================================
>; NET - Section 2 - Define the target system
>; ============================================================

The header identifies a specific part of NETGEN (NET) and the section number within
that part (Section 2) and then describes the purpose of the section (Define the target system).

4.5.1

Explanatory Text

During the NETGEN dialog, you can read expanded information for a question. You
elicit this text on a question by question basis by pressing the < ESCAPE> key after the
question is displayed. This causes the NETGEN command file to issue an explanation
and then reissue the question. For example, if you press < ESCAPE> in response to the
question:
> 08.00 Is this generation to be a dry run? [D=N] [YIN]:
NETGEN prints the following explanatory text and then repeats the question:
If you answer YES to this question, the system you specify wi I I
not actually be bui It. Your responses wi I I be saved, however,
and a real generation may be performed later, using these saved
responses.

NOTE

The explanatory text after the preceding question, and after the questions in
the entire NETGEN dialog as presented in Chapters 5, 6, and 7, is a word-byword representation of what you will see on your screen when you perform a
NETGEN. It may not, however, be a line-by-line representation of what
appears on your screen.

4.5.2

Response Formats

Your response to NETG EN questions will be one of three types:
•

YES/NO [YIN]

•

Numeric [N]

•

Character string [S]

NETGEN's default answer for a YES/NO question is NO unless otherwise stated.

Using the RSX-11 M/M-PLUS NETGEN Procedure

4-7

NOTE
If you respond to most NETGEN questions using default answers, you will gen-

erate a working system.
Numeric responses are octal or decimal. In the example [0 R: 1-377 D:5], 0 indicates
the octal number, R indicates the range, and D indicates the default. In the example [D
R: 1.-16. D: 1.], the first D indicates the decimal number, R indicates the range, and the
second D indicates the default. You do not need to include a decimal point after the
number you enter; it is assumed.
Character string responses can include letters, numbers, and special characters. In the
example [dduu,D = DBOI :][S], dd refers to two letters for the device, uu refers to two
numbers for the unit; D refers to the default; and S refers to a string that can include letters, numbers and special characters. See Table 4-1 for a summary of NETGEN
responses.
Table 4-1 :

Summary of NETGEN Responses

Response Type

NETGEN displays:

You enter:

YES/NO

[D = N][Y /N]:

Y to indicate YES, N to indicate NO, or press
<RET> for default NO.

Numeric
Octal

[0 R:m-n]:

An octal number in the range of m through

Numeric
Decimal

Character
String

4-8

n.
[0 R:m-n D:d]:

An octal number in the range of m through n
or press < RET> to indicate that the default
d should be used.

[D R:m.-n.]:

A decimal number in the range of m through

n.
[D R:m.-n. D:d.]:

A decimal number in the range of m through
n or press < RET> to indicate that the
default d should be used. You do not need to
include a decimal point after the number
you enter.

[8]:

A character string of any length.

[f, D:d][S]:

A character string in the form f (such as
ddu:), or press <RET> to indicate that the
default d should be used.

[8 R:m.-n]:

A character string from m through n characters in length.

DECnet-RSX Network Generation and Installation Guide

NOTE

Certain NETGEN questions may specify a string response in the range of O-n. In
such cases, the 0, specifying the low end of the range, represents the < RET>
response that is used to accept a value previously specified in a saved response
file.

4.5.3

Responses for End-of-section and Break Questions

Each NETGEN section ends with an end-of-section question, as shown in the following
example:
>; <EOS> Do you want to:
>*
<RET>-Continue, S-Skip, R-Repeat Section, P-Pause,
E-Exit

Respond by entering one of the following options:

< RET> to continue on to the next section. Responses are saved.
S

to skip a section or sections. The skip option is provided in restore mode
only. Use it when you have chosen to change your answers. Choosing Skip
causes NETGEN to resume using your saved responses and prevents you
from making any further changes to the remaining sections.

to repeat the section you just completed. Responses are discarded and the
entire section is repeated.

to pause. This response temporarily suspends the indirect command file and
gives you direct access to the operating system to perform a task. The system displays one of the following messages:
AT .--PAUSING. TO CONTINUE, TYPE "RES string"
or
AT.--PAUSING. TO CONTINUE, TYPE "UNS string"
After you have completed your task, you can continue processing by entering:
RES string
or
UNSstring
Enter RES or UNS, depending upon which one of the above messages was displayed, and enter the same string contained within that message. The endof-section question is then repeated.

to exit from NETGEN. Responses are saved, allowing you to terminate the
generation, and later return to the dialog using restore mode. Entering
< CTRLlZ > in response to an EOS question is identical to entering E.

Using the RSX-11 M/M-PLUS NETGEN Procedure

4-9

If you want to perform one of the above functions in the middle of a section, enter

< CTRLlZ > . This interrupts normal processing and displays the following break question:
>; <BRK> Do you want to:
>
<RET>-Continue, R-Repeat Section, P-Pause, E-Exit[S]:
If you enter E in response to a break question, your responses to the current section are

lost, but responses to previous sections are saved. Entering < CTRL/Z > in response to
a break question is identical to entering E.

4.6

Abnormal NETGEN Termination

If NETGEN terminates abnormally with an ... AT. error message (such as .. AT. - File
read error), or if it simply stops prematurely with an end-of-file (@< EOF > ) on the com-

mand file, the indirect command file processor (AT.) has most likely encountered a disk
error. You should restart NETGEN from the beginning using a different disk drive.

4-10

DECnet-RSX Network Generation and Installation Guide

5
NETGEN Dialog:

NET

This chapter presents the questions you will be asked in the first part of NET GEN. The
first part of the NETGEN dialog is NET. NET displays a series of questions concerning
the Communications Executive (CEX) and other common components.
The following NET dialog represents the text NETGEN displays when performing an
initial generation in question and answer mode. A component mode or a restore mode
generation sometimes displays a slight variation in the wording of specific questions.
For example, the defaults displayed sometimes vary from those displayed in this dialog. Your responses should always be guided by the question and text displayed on
your terminal.

5.1

Initialization Questions

In any mode, the following initialization questions are displayed. Figure 5-1 illustrates
the sequence of questions for general initialization.

5-1

TW0169

Figure 5-1:

5-2

NET - Section 1 - General Initialization

DECnet-RSX Network Generation and Installation Guide

============================================================

>; NET - RSX-IIM/S/M-PLUS CEX System Generation Procedure
>;
Sta rted at time on date
>; Copyright (C) 1985 by
>; Digital Equipment Corporation, Maynard, Mass.
Generate a CEX System, version version-num, for your
RSX-IIM/S/M-PLUS target system.
The fol lowing CEX Products may be generated
DECnet
PSI
The distribution kits must be up to the current patch
level. If necessary, you should stop the generation and
apply the necessary patches.

The logical device name LB: should already be assigned to
the device which contains the library fi les for the system you
are generating. These fi les are EXELIB.OLB, EXEMC.MLB,
RSXMAC.SML, SYSLIB.OLB, and, for RSX-llS, 11SLIB.OLB.
The logical device names IN:, OU:, LS: and MP: should not be
assigned to any devices before starting the generation. These
logical device names are used by NETGEN to refer to various
disks used during the generation.
>;<EOS> Do you want to:
>*
<RET>-Continue, E-Exit [S]:
<RET>- Continue with normal processing E

- Terminate the generation

============================================================
>; ============================================================

>; NET - Section 1 - General Initial ization

The notes/cautions describe the prerequisite conditions for
starting the NETGEN procedure.
01.00 Do you want to see the NETGEN notes/cautions? [D=N] [Y/N]:
NETGEN displays the following text if you answer Question 1.00 YES.
Note the fol lowing:
If you have not already performed a SYSGEN for your target
system, you must do so before continuing .
. The privi leged tasks generated during NETGEN use configuration
dependent parameters produced by SYSGEN. For
RSX-IIM/S/M-PLUS systems, the RSXMC.MAC fi Ie
and the RSXIIM.STB (or RSXllS.STB) fi Ie for the
target system are necessary. If they are not already
on the target device under the proper UIC, NETGEN wi I I move
them from the user's system device to the target device.

NETGEN Dialog:

NET

5-3

If your target device contains useful fi les, it is
recommended that you back it up before continuing, as
the device wi I I be write enabled during the NETGEN sequence.
The proper system I ibraries for the target system must be on
the logical device LB: under the UIC [1,1] whi Ie you are
performing your NETGEN. These I ibraries are:
EXELIB.OLB, EXEMC.MLB, RSXMAC.SML, and SYSLIB.OLB.
In addition, for an RSX-11S NETGEN, the 11SLIB.OLB fi Ie
is requi red.
If your target system supports the Queue Manager (QMG) ,
and you wish to generate network tasks which may use this
faci I ity, the Queue Manager object I ibrary must be on the
logical device LB: under UIC [1,24] whi Ie you are performing
your NETGEN. This fi Ie is:
QMG.OLB
If you wish to generate network tasks which requi re support
for RMS fi Ie access,You must have the RMS fi les on your
logical LB: device prior to performing your NETGEN. NETGEN
will look for these files under the UIC determined by SYSGEN.
The following are the fi les that NETGEN wi II look for:
RMSLIB.OLB, RMS11X.ODL and RMS12X.ODL.
RMSRES.TSK and RMSRES.STB (if resident libraries
are supported).
If your target system supports a memory resident or a
supervisor mode FCS I ibrary, and you wish the network
tasks to use this I ibrary, the FCS library fi les must be
on the logical device LB: under UIC [1,1] whi Ie you are
performing your NETGEN. These fi les are:
FCSRES.TSK and FCSRES.STB (memory resident FCS
library files).
FCSFSL.TSK and FCSFSL.STB (supervisor Mode FCS
library files).
The fol lowing Instal led tasks are required on your host
system whi Ie you are performing your NETGEN:
PIP, LBR, MAC, and TKB for RSX-11M/S/M-PLUS NETGENs.
If these tasks are not al ready instal led, and you are
logged in as a privi leged user, NETGEN wi I I instal I them
automatically. If TKB and MAC are instal led, you may want
to remove them and relnstal I them with an increment of 3000.
to 15000. bytes so the bui Id section of NETGEN wi I I run
faster.

5-4

DECnet-RSX Network Generation and Installation Guide

The fol lowing instal led tasks are required on your host
system whi Ie you are performing your NETGEN, under the
fol lowing conditions:
LOA and UNL if device drivers requi re loading.
MOU and OMO if disks requi re mounting.
UFO if the required UICs do not already exist.
If these tasks are not already installed, and you are
logged in as a pr i v i I eged user, NETGEN wi II i nsta II them
automatically.
NETGEN wi II not work wi th the sma II I nd i rect command
p rocesso r. I f you have rebu i Itt he i nd i rect command
processor and spec if i ed the OOL for the sma I lind i rect
command processor, you must instal I the indi rect command
processor that was provided as output from your SYSGEN.
If you are generating more than one network onto the same
target disk, the network help fi Ie ([1,2]NETHLP.ULB) and
the language libraries ([l,l]NETFOR.OLB and [l,l]NETLIB.MLB)
wi I I be replaced each time NETGEN is run. You may want to
save these fi les before doing the next NETGEN. This IS of
particular interest if one of the networks is for RSX-11S and
the other network IS for RSX-IIM or RSX-IIM-PLUS.
For questions that can be answered YES or NO, the default is
NO unless otherwise specified. For questions with defaults,
the default is produced by pressing a carriage return «RET»
in response to the question.
Additional explanatory text for each question may be obtained
by hitting the <ESCAPE> key.
Unless otherwise specified, it is possible to Interrupt a
series of questions by typing <CTRL/Z>. You wi I I then be
given the options of either restarting the section from the
top, temporari Iy stopping the generation, or terminating
the generation.
02.00 Target system device [dduu, O=SYOO:] [S]:
This device Will contain all files required to run the
network being generated. This includes al I task fi les,
databases, command fi les, and generation work fi les. This
must be a Fi les-ll disk device. The trai I ing colon (e.g. SY:)
IS optional. The default device is SY:.

NETGEN displays Questions 2.01 and 2.02 only if the device has not been mounted. If
you answer YES to Question 2.01, NETGEN will mount the device.
>* 02.01 Is dduu ready? [O=N] [Y IN] :
Load the disk into drive dduu: and start the drive
before answering YES to this question.

NETGEN Dialog:

NET

5-5

>* 02.02 Volume label

for dduu: [O=None] [5 R:O.-12.]

A volume label is requi red for al I mountable devices. This
volume label was set when the volume was initial ized.

The following error message and Question 2.03 appear only if the device could not be
properly mounted.
An error occurred when the volume was mounted. Possibly the
volume is not properly loaded, the volume label was incorrectly
suppl ied, the volume was not initial ized to Fi les-ll, or the
"MOU" task was not instal led.
>* 02.03 Do you wish to specify a different device? [O=N]

[YIN]:

If you answer YES, you wi I I be asked for a different device
name.
>* 03.00 Listing/map device? [dduu, O=None]

[5]:

This device wi I I contain al I list fi les and map fi les
generated by the assembler and task bui Ider. This must be
either a Fi les-il disk device, or the LP: or NL: devices.
The trai I ing colon (e.g. SY:) is optional. If no maps and
I istings are desi red at all, respond with a carriage return.
If you want maps and listings on a device other than a target device, use another avail-

able device. Approximately 3000 blocks are required for maps and listings, depending
on the optional components you specify. If you specify a device in Question 3.00 that
you have not yet defined to the system, you will be asked a series of questions similar to
Questions 2.01 through 2.03.
>* 04.00 UIC Group Code for NETGEN output? [0 R:I-377 0:5]

[5]:

Specify the UIC Group Code under which al I generated fi les
wi II reside. Work fi les, source fi les, I ibrary and object
f i I es, list i ng f i I es, map f i I es, and task and database f i I es
wi I I al I appear under various user codes within this group
(e.g. [5,*]). Files which already exist under these UICs may
be deleted or purged.
Note that the user must have privi leges to create, write, read,
and delete al I fi les under the chosen UIC Group Code.

When you do your first NETGEN, you can choose any DIC. However, if you do another
NETGEN and you want to make use of your previous answers, then you must specify
the same DIC you specified in your first NETGEN. Similarly, specify the same DIC if you
do a NETGEN in component mode.

5-6

DECnet-RSX Network Generation and Installation Guide

NETGEN displays Question 5.00 only if you have previously performed a full NETGEN.
>* 05.00 Do you want to perform a component mode generation? [D=N]

[YIN]:
If a CEX system has already been generated on [x,*], you
may wish to generate only components for that system.
Components are those tasks andlor databases which do not change
the previously defined network configuration. If you wish to
generate only components, answer YES, and only component
questions wi I I be asked.

Once you specify component mode, NETGEN runs in component mode throughout;
responses are not saved, and the NETINS.CMD file is not updated (see Chapter 9).
NETGEN displays the following text and Question 6.00 only if you have performed one
previous NETGEN.
Saved responses were detected. These were compi led on dare
at time wi th a User ID of "user-id".
>* 06.00 Do you wish to ~se these saved responses? [D=NJ

[YIN]:

Answer this question YES if you wish to use the responses
f rom date at time, User ID "user-id".
If you answer YES, these saved responses wi I I be used to
answer as many of the questions as possible, for the new
system you are generating. If these saved responses are
incomplete, NETGEN wi I I use the responses it finds and prompt
you for the remainder of the responses. If you answer NO, your
previous responses wi I I be ignored.

NETGEN displays the following text and Question 6.01 if you have previously performed two or more NETGENs.
Saved responses were detected. These were compi led on
date at time wi th a User ID of "user-id".

An earl ier set of saved responses was compi led on date at
time wi th a User ID of "user-id".
>* 06.01 Do you wish to use t he
N]:

I ate r set of saved responses?

[D=N] [Y I

Answer this question YES if you wish to use the responses
f rom date at time, User ID "user-id".
If you answer YES, these saved responses wi I I be used to
answer as many of the questions as possible, for the new system
you are generating, and the earl ier set of saved responses (from
date at time, User ID "user-id") wi I I be lost.
If these saved responses are incomplete, NETGEN wi I I use
the responses it finds and prompt you for the remainder of
the responses.
If you answer NO, you wi I I be asked about the earl ier set of
saved responses.

NETGEN Dialog:

NET

5-7

>* 06.02 Do you wish to use the earl ier set of saved responses? [D=N]
[YIN] :
Answer this question YES if you wish to use the responses from
date at time, User 10 "user-id".

If you answer YES, these saved responses wi I I be used to
answer as many of the questions as possible, for the new
system you are generating, and the later set of saved responses
(from date at time, User 10 "user-id") wi II be lost.
If these saved responses are incomplete, NETGEN wi I I use the
responses it finds and prompt you for the remainder of the
responses.
If you answe r NO, these saved responses (f rom date at time,
User 10 "user-id") wi II be lost.

>* 06.03 Do you want to display these saved responses? [D=N] [YIN]:
If you answer YES to this question, when the saved responses
are used, the data restored via these saved responses wi I I be
displayed on the terminal.
It is recommended that you answer Question 6.03 YES.

>* 06.04 Do you want to change these saved responses? [D=N] [YIN]:
If you answer YES to this question, when each section is
finished, you wi I I be given the option of repeating the
section. If you choose to repeat the section, al I questions
wi I I be asked again, and you answer the section anew. This
al lows you to change your previous responses.
When possible, prevIous responses are displayed as defaults, so
that typing <RET> will cause the original response to be used
again. Note that for some questions this is not possible.
Responses to these questions must always be fully specified,
each time. For example, al I of the questions in NET section 3
are of the type which must always be completely respecified.
If you answer NO to this question, your previous responses
are used as they stand, and you are given no option for
changing them.
>* 06.05 Have you performed a SYSGEN since your last NETGEN?
[D=N] [Y IN] :
Answer this question YES if you have performed a new SYSGEN
since the NETGEN being restored was done. Answer NO if a new
SYSGEN has not been done. This wi I I determine whether or not
new RSXMC.MAC and RSXIIM.STB fi les are needed.
If you answer YES, several sections may be automatically
repeated, in order to reset answers or values which were based
on the old system fi les. If you answer NO, sections wi I I be
repeated only at your request.

5-8

DECnet-RSX Network Generation and Installation Guide

NETGEN does not display Question 7.00 in component mode.
>* 07.00 User 10 for saving new responses? [D=None]

[S R:0.-30.]:

The responses given during this generation wi I I be saved for
later reference or re-use. You should provide some 10 (up to
30 characters) by which you wi I I be able to tel I, at some
future time, whether or not these saved responses are usable.
Note that this string must not contain any single-quote (') or
double-quote (") characters.
It is advisable that you do not use a default here. You do not need to code in the date

and time; they are stored. Any characters, including spaces and punctuation marks, can
be used such as LARRY DAY: DRY RUN 2.
NETGEN does not display Questions 8.00,9.00, or 10.00 in component mode.
>* 08.00 Is this generation to be a dry run? [D=N]

[YIN]:

If you answer YES to this question, the system you specify wi I I
not actually be bui It. Your responses wi I I be saved, however,
and a real generation may be performed later, using these saved
responses.
>* 09.00 Do you want a standard function network? [D=N] [YIN]:

Answering YES to this question wi I I result in NETGEN
asking a minimum number of questions. A ful I complement
of components wi I I be generated and most optional parameters
wi II be filled in with predefined values. If a value has not
been defined for a parameter, NETGEN wi I I ask the required
questions.
If you answer YES to Question 9, NETGEN does not display Question 10.
>* 10.00 Should al I components be generated? [D=N]

[YIN]:

If you answer YES to this question, your system WI I I be
generated with the ful I complement of components but you
wi I I sti I I have to specify the values for al I optional
parameters. If you answer NO, you WI I I be asked on an
individual basis whether or not each component is to be
generated.

NOTE

Certain components (for example, the network management components, NTL, and NTINIT) are always generated regardless of your
response to Question 10.00.

NETGEN Dialog:

NET

5-9

Question 11.00 is not displayed in Dry Run Mode.
>* 11.00 Should old fi les be deleted? [D=N] [YIN]:
If this question is answered YES, al I old fi les on the
target, I isting and map devices under the target UICs
(except those fi les which wi II be needed during this
generation) wi I I be deleted.
This is recommended if the target UICs contain fi les from
a previous generation, or if saved responses are being used
to recover from a system crash.
>;<EOS> Do you want to:
>*
<RET>-Continue, R-Repeat section, P-Pause, E-Exit

5-10

DECnet-RSX Network Generation and Installation Guide

5.2

Defining the Target System

This section defines a number of characteristics for your target system. These characteristics define a DECnet-only system, a DECnet/PSI system, and operating system characteristics. Figure 5-2 illustrates the sequence of questions for defining the target system.

TW0170

Figure 5-2:

NET - Section 2 - Define the Target System

NETGEN Dialog:

NET

5-11

>
>
>

============================================================
NET - Section 2 - DefIne the Target System
============================================================

>* 01.00 Use the RSXMC. MAC on dduu: [uic]? [D=N]

[Y IN] :

RSXMC.MAC is the fi Ie which defines the features bui It into the
target operating system. A copy of this fi Ie has been found on
the Target Devlce/UIC. If this is a correct copy, answer the
question YES.

NETGEN displays Question 1.00 if the RSXMC.MAC file is found on the target disk and
under the network DIC.
NETGEN displays Question 2.00 only if NETGEN cannot locate this file on the target
disk and under the network DIC.
>* 02.00 RSXMC.MAC location? (dduu:[g,m], D=target
[ S] :

device: [uicj)

RSXMC.MAC is the fi Ie whIch defines the features bui It
into the target system. This fi Ie must be moved from its
present locat ion to dduu: [uic]. The present locat ion
must be specified as a device and UIC, in any of the following
formats.
dduu: [ggg,mmm] Specify both device and UIC
dduu: Specify device, with UIC of default uic
[ggg,mmm] Specify UIC, on device target device:

After you specify its location, NETGEN moves RSXMC.MAC to the correct location.
This file is customarily stored on LB:[ 11,10]. If you do not know the location of the file,
you may need to press < CTRLlZ > to get a break question and then pause to gain
access to the operating system in order to search for it. Enter its location in one of the
indicated formats.
NOTE
If you specify a device in response to this question, and if the device

you specify is unknown to NETGEN, NETGEN will ask a series of questions similar to Questions 2.01 and 2.02, in Section 1.
dduu: [UIC] RSXMC. MAC

is be i ng scanned to def i ne you r
target system. This may take up to several minutes.

The target is an RSX-I1M-PLUS system with
A 20K-Executive
Kernel data space enabled
Extended memory support (more than 124K words of memory)
Multi-user protection
Extended instruction set (EIS)
Powerfa i I recovery support
Dynamic task checkpoint al location
PLAS support
No Queue Manager (QMG) support

5-12

DECnet-RSX Network Generation and Installation Guide

This list represents the information NETGEN obtained from the RSXMC.MAC file. Your
system may display a different list.
NETGEN displays Question 3.00, if the RSXI1M.STB file is found on the target disk and
under the network UIC. For RSX-llS systems, NETGEN displays the file name,
RSXllS.STB.
>* 03.00 Use the RSXllM. STB on dduu: [uic]

[D=N]? [YIN] :

RSXllM.STB is the fi Ie which defines the global symbols for
the target operating system. A copy of this fi Ie has been found
on the Target Device/UIC. If this is a correct copy, answer
the question YES.

NETGEN displays Question 4.00 only if NETGEN cannot locate this file on the target
disk and under the network UIC. For RSX-llS systems, NETGEN displays the file name
RSXllS.STB.
>* 04.00 RSXllM.STB location (dduu: [g,m], D=dduuluiQ)

[S]:

RSXllM.STB is the fi Ie which defines the global symbols
for the target system. This fi Ie must be moved from its
p resent I ocat i on to dduu: [uic]. The p resent I ocat ion
must be spec if i ed as a dev i ce and UIC, in any of the
fol lowing formats.
dduu: [ggg,mmm] Specify both device and UIC
dduu: Specify device, with UIC of default UIC
[ggg,mmm] Specify UIC, on device dduu:

After you specify its location, NETGEN moves RSXIIM.STB to the correct location.
This file is customarily stored on LB:[1,54] for RSX-IIM/I1M-PLUS systems and on
LB:[1,64] for RSX-llS systems. If you do not know the location of the file, you may
need to press < CTRL/Z > to get a break question, and then pause to gain access to the
operating system in order to search for it. Enter its location in one of the indicated formats.
NOTE
If you specify a device in response to this question, and if the device

you specify is unknown to NETGEN, NETGEN displays a series of questions similar to Questions 2.01 and 2.02 in Section 1.
NETGEN displays Question 5.00 only forRSX-11M-PLUS systems, if you specified the
supervisor mode FCS library during SYSGEN.
>* 05.00 Should tasks

I Ink to the supervisor mode FCS library?

[D=N] [Y IN] :
If you have a supervisor mode FCS I ibrary and you would I ike
the network tasks to I ink to it, answer this question YES.
Note that this I ibrary must be located as fol lows:
LBOO: [l,l]FCSFSL.TSK
LBOO: [l,l]FCSFSL.STB

NETGEN Dialog:

NET

5-13

Answering this question YES, causes tasks that link to this library to be smaller.
NETGEN displays Question 6.00 only for RSX-IIM/-IIM-PLUS systems, if you specified the Memory Resident FCS library during SYSGEN.
>* 06.00 Should tasks I ink to the memory resident FCS library?
[D=N] [YIN]:
If you have a memory resident FCS I ibrary and you would
I ike the network tasks to I ink to it, answer this question
YES.
Note that this 1 ibrary must be located as fol lows:
LBOO:[1,1]FCSRES.TSK
LBOO: [1,1]FCSRES.STB

Answering this question YES, causes tasks that link to this library to be smaller.
The following statement appears in question and answer mode or restore mode for a
DEC net-only system. It is not displayed in component mode.
The DECnet CEX Product wi I I be generated.

Depending on the type of kit you purchased, NETGEN displays one of the following
statements.
•

Routing or nonrouting nodes can be generated from your distribution kit.

•

Nonrouting nodes can be generated from your distribution kit.

NETGEN displays Question 7.00 in component mode only.
>*07.00 Generate CEX system components? [D=N] [YIN]
If you want to include the indicated CEX Product in your
network system, answer the question YES.

NETGEN displays Questions 7.01 and 7.02 in question and answer mode only if your
NETGEN disk contains the software required to generate the combined DECnet/PSI
product. NETGEN displays similar questions in component mode if you are modifying
a DECnet-only system or a DECnet/PSI system.
>* 07.01 Generate the DECnet product? [D=N] [YIN]:
If you want to include the indicated CEX Product in your
network system, answer the question YES.
If you are adding or changing specific PSI components to an existing DECnet/PSI system, and you do not need to supply DECnet components, answer this question NO. If

you are adding or changing specific DECnet components to an existing DEC net system,
answer this question YES.
If you want to include the DECnet components in your original DEC net/PSI genera-

tion, answer this question YES.

5-14

DECnet-RSX Network Generation and Installation Guide

>* 07.02 Generate the PSI product? [D=N] [YIN]:
If you want to include the indicated CEX Product in your
network system, answer the question YES.
>;<EOS> Do you want to:
>*
<RET>-Continue, S-Skip, R-Repeat section, P-Pause,
E-Exit

NETGEN Dialog:

NET

5-15

5.3

Defining the System Lines

To define your system lines, NETGEN asks you to specify a device driver module
(DDM) process name. The DDM identifies a process that controls an 1/0 device.
DECnet includes DDCMP devices, EPM (the Ethernet Protocol Manager) devices, and
DLM (data link mapping) pseudodevices. PSI includes X.25 devices.
•

DECnet DDMs support devices that allow DECnet users to communicate with
other DECnet users using the DECnet DDCMP protocol.

•

EPM DDMs support Ethernet devices that allow DECnet users to communicate
with other DECnet users using the Ethernet protocol.

•

DLM DDMs are pseudodevices that allow DECnet users to communicate with other
DEC net users using X.25 circuits over a Packet Switching Data Network (PSDN).
DLM circuits are referenced as if they are circuits on a multipoint tributary.

•

PSI DDMs support devices that are connected to a PSDN.

DLM and PSI circuits are available only if you have purchased a license for the
Packetnet System Interface.
In Questions 2.00 through 5.00, NETGEN displays the variable dev. Device names (de v)
can have different forms. In each case, NETGEN will display the correct form for the
specific communications device you specify in Question 1.00. For example, KDP-21.3 would be displayed on-line using the following format. This is the longest form a
device name can take.
dev-x-y.z

where:
dev

is the current device driver module.

is the number of this controller.

is the number of this line on the controller.

is the number of this multipoint tributary on the current line.

NOTE

Some devices (DMC, DMR, DUP) do not support multiple lines or tributaries. For example, DMC supports one line having the format, dev-x.

5-16

DECnet-RSX Network Generation and Installation Guide

Figure 5-3 illustrates the sequence of questions for defining the system lines.

TW0171

Figure 5-3:

NET - Section 3 - Define the System Lines

NETGEN Dialog:

NET

5-17

>; ============================================================
>; NET - Section 3 - Define the System Lines
>; ============================================================
>* 01.00 Device Driver Process name? [<RET>=Done]

[S R:0-3]:

The Device Driver Module (DDM) Process name identifies a process
which controls an liD device. Enter a carriage return when
there are no more devices to support.
The legal DECnet device driver process names for UNIBUS
processors:
Name
DMC
DMP
DU
DUP
DV
KDP
UNA
DLM

Physical device
Name
DMC11 or DMR11
DL
DMP11
DZ
DUll
DHU
DUP11
KDZ
DV11
PCL
KMC11/DUP11
DEUNA
PSI Data Link Mapping

Physical device
DL11
DZ11
DHU11
KMC11/DZ11
PCL11

The legal DECnet device d rive r process names for Q-BUS processors:
Name
DMV
DPV
DUV
DLM
QNA

Physical device
Name
DMV11
DLV
DPV11
DZV
DUV11
DHV
PSI Data Link Mapping
DEQNA

Physical device
DLV11
DZV11 or DZQ11
DHV11

Not all of these devices may be available to you. The following devices are available for
DECnet/PSI systems other than those that use the Q-BUS:
The legal PSI device driver process names are:
Name
SOP
KMX
KMY

Physical device
DUP11
KMS11-BD
KMS11-PX

If you have a Q-BUS (LSI) processor, NETGEN displays the following list for DECnetl

PSI systems:
The legal PSI device driver process names are:
Name
SDV

5-18

Physical device
DPV11

DECnet-RSX Network Generation and Installation Guide

NETGEN displays Question 1.01 only if NETGEN does not recognize the device you
entered.
>* 01.01 Do you have a user written dev Device Driver? [D=N] [YIN]:

NETGEN does not have a Device Driver (DDM) Process for the dev.
If a user written Device Driver Process exists, NETGEN wi I I
create the necessary command fi les to bui Id a device driver
for your device.
If you answer YES to this question, NETGEN displays a series of questions (not shown

in this manual) to help you specify the parameters for the particular device.
>* 02.00 How many dev controllers are there? [D R:l.-16 0:1.]:

Enter the number of dev controllers that are instal led on the
target system. Do not include any of the controllers which were
specified during the target system's SYSGEN, since devices
cannot be shared between the CEX system and the operating
system.
If you have 3 DUP controllers, enter 3. For data link mapping (DLM), considered to be a

multipoint controller, you can specify more than one "controller" in this question.
This allows you to group lines logically.
NETGEN displays Question 2.01 through 2.06, if you specified more than one device
controller in Question 2.00.
>* 02.01 Are al I devs instal led at the same device priority?

[D=N] [YIN]:
All peripheral interfaces on PDP-l1 processors are installed
with a hardware device priority in the range of 4 to 6. This,
together with its position on the UNIBUS (or Q-BUS), determines
its priority in arbitrating for control of the BUS.
Answer this question YES if al I DEVs have been instal led at the
same device priority.
If you answer Question 2.01 YES, NETGEN displays Question 2.02.
>* 02.02 What is the dev device priority? [0 R:4-6 D:5]:

Enter the hardware device priority level for all Devs
on the target system.
Note that if the hardware device priority is later changed, the
network value can be changed with NCP, VNP, or CFE on a per line
basis.
>* 02.03 Wi I I al I the dev I ines run at the same I ine speed? [D=N] [YIN]:

If al I of the dev I ines wi I I run at the same I ine speed,
then answer this question YES. If you answer this question NO,
you wi I I be asked for the I ine speed on a per I ine basis.

NETGEN Dialog:

NET

5-19

If you answer Question 2.03 YES, NETGEN displays Question 2.04.
>* 02.04 Line speed fo r dev [D R: min-max D: default] :

Specify the I ine speed in bits per second. If the I ine speed is
greater than 56000 bps, enter 56000. If the I ine speed is
externally determined, enter the speed at which the I ine is
expected to operate.
Note that this value can be changed with CFE.
Legal I ine speeds for synchronous and asynchronous devices
50
75

110
134

150
300

600
1200

1800
2000

2400
3600

4800
7200

9600

Add i tiona I I ega I line speeds fo r synch ronous dev ices
19200 For al I synchronous devices, except the DUP11
56000 For the DMC, DMP, and DMV devices only

Not all of these speeds are available for each device. Consult instructions for the specific device. See additional text under Question 4.01 for information about the importance of setting the correct line speed.
>* 02.05 Wi I I al I dev

ines run as FULL duplex lines [D=N] [YIN]:

If al I the dev I ines are to run as FULL duplex, then answer
this question YES. If you answer this question NO, you wi I I be
asked if al I dev I ines wi I I run as HALF duplex. If you answer
both of these questions NO, you wi I I be able to choose on a per
I ine basis whether the I ine is HALF or FULL duplex.
Note that the duplex mode can later be changed with NCP, VNP,
or CFE.
If you answer Question 2.05 NO, NETGEN displays Question 2.06.
>* 02.06 Wi I I al I dev

ines run as HALF duplex lines [D=N] [YIN]:

If al I the dev I ines are to run as HALF duplex, then answer this
question YES. If you answer this question NO, you wi I I be able
to choose on a per line basis whether the I ine is HALF or FULL
duplex.
Note that the duplex mode can later be changed with NCP, VNP, or CFE.

NOTE

In the dialog that follows, if no specific reference is made to the ability
to change a parameter using NCP, VNP, or CFE, that parameter will
require a new NETGEN.

5-20

DECnet-RSX Network Generation and Installation Guide

NETGEN displays Question 3.00 only for multiplexer devices.
>* 03.00 Number of lines used on dev? [0 R:l.-max-allowed. 0:1.]:

Specify the number of I ines to be used on the multiplexer device
controller.
>* 03.01 CSR add ress fo r dev [0 R: 160000-177777 0: default CSR] :

the Control/Status Register (CSR) address for the
controller. Use 177777 if the address is unknown.

S~ecify

Note that the CSR address can later be changed with NCP, VNP,
or CFE.

The CSR address and the vector address are set up in the hardware when the device is
first installed.
>* 03.02 Vector address for dev [0 R:0-774 O:default Vector Address]:

Specify the interrupt vector address for the controller. Use 0
if the address is not known.
Note that the vector address can later be changed with NCP,VNP,
or CFE.

If you specify more than one device and these devices have been installed with different device priorities, NETGEN displays Question 3.03.
> 03.03 Device priority for dev [0 R:4-6 0:5]:

Specify the hardware device priority for the controller.
Note that if the hardware device priority is later changed, the
network value can be changed with NCP, VNP, or CFE.
>* 04.00 Is dev a FULL or HALF duplex line [O=FULL] [S]:

Specify the duplex mode the I ine is to run in. This should
either be FULL or HALF (these may be abbreviated to F or H).
Note that the duplex mode can later be changed with NCP, VNP,
or CFE.
>* 04.01 Line speed fo r dev [0 R: min-max 0: default] :

Specify the I ine speed in bits per second. If the I ine speed
is greater than 56000 bps, enter 56000. If the I ine speed is
externally determined, enter the speed at which the I ine is
expected to operate.
Note that this value can be changed with CFE.
Legal I ine speeds for synchronous and asynchronous devices
50
110
150
600
1800
2400
4800
9600
75
134
300
1200
2000
3600
7200

NETGEN Dialog:

NET

5-21

Additional legal I ine speeds for synchronous devices
19200 For al I synchronous devices, except the DUP11
56000 For the DMC, DMP, and DMV devices only

Not all of these speeds are available for each device. Consult instructions for the specific device.
The speed of a synchronous device is determined by the speed of the modem. The
speed of an asynchronous device is determined by switch settings on the line device.
The line speed specified is used to calculate timing criteria for the line and buffer
requirements. If the specified speed is incorrect, the timing will be incorrect and the
large buffer allocation (see the DECnet-RSX Network Management Concepts and Procedures manual) may be incorrect. If the specified line speed is higher than the actual
line speed, there may be CRC errors on large data buffer or large data buffer allocation
failures. If the specified line speed is lower than the actual line speed, there is a long
delay on line timeouts. Be sure to use CFE to change the line speed on synchronous
lines if the modem speed is changed.
NETGEN displays Question 4.02 only for DLII or DLVII devices. For Q-BUS systems
DLII-E will be replaced with DLV-IIE.
>* 04.02 Is dev a DL11-E or equivalent device? [D=N]

[YIN]:

Answer the question YES only if the device is a DL11-E or
equivalent device. These devices al low the Clear To Send
signal to pass through the interface.

NETGEN must establish special handling for the clear-to-send signal for devices that do
have this signal available. You should answer NO if the device is any of the following:
DL-IIA,B,C,D
DL-IIWA,WB
DLV-IIF,J
NETGEN displays Question 4.03 for each secondary device controlled by a primary
device. For example, the primary device might be a KDP that controls a number of secondary DUP devices.
>* 04.03 Unit CSR address for dev [0 R:160000-177777 D:default add.]:

Specify the Control/Status Register (CSR) address for the
device which the dev is control I ing. Use 177777 if the
address is unknown.
Note that the unit CSR address can later be changed with NCP,
VNP, or CFE.
If you specify a device that requires a modem control device, such as KMX, NETGEN

displays Question 4.04.
>* 04.04 Modem control CSR address for dev [0 R:160000-177777

5-22

D:de~:

DECnet-RSX Network Generation and Installation Guide

Specify the Control/Status Register (CSR) address for the
modem control unit associated with the DEV. Use 177777 if
the address is unknown.
Note that the modem control CSR address can later be changed
with NCP, VNP, or CFE.
NETGEN displays Questions 4.05 and 4.06 only for asynchronous devices that support
a modem.
>* 04.05 Is modem signal monitoring required for dev? [D=N] [YIN]:

If the I ine is connected to a modem for a dial-up switched
ci rcuit service, you should select modem monitoring. Modem
signal monitoring checks the Carrier Detect signal (CD or CAR)
for asynchronous lines.
If this signal is missing for two consecutive samples, taken
at one second intervals, the I ine wi I I be declared "down"
without waiting for a DECnet protocol to detect the fai lure.
If your modem supports this signal, and you want modem
monitoring, answer this question YES.
>* 04.06 Does the modem switch carrier with Request To Send?

[D=N] [YIN]:
I~ is necessary to distinguish asynchronous I ine modems
which present continuous carrier once the connection is
establ ished, such as the BELL 103 or BELL 212, from those
which switch carrier along with Request To Send, such as the
BELL 202.

Answer the question YES if you are using a BELL 202 modem
or the equivalent.
If the device is a DLII-E or DLV-IIE and the modem (a Bell 202 or equivalent)
switches carrier with request to send (RTS), you should answer NO to question 4.05.
Answering YES would require the monitoring of data set ready (DSR), a signal that is
not present on DL or DLV devices. If the modem does not switch carrier with RTS
(such as Bell103, Bell 113, or Be11212 modems), answer YES to Question 4.05 and NO

to Question 4.06, since carrier (CAR) will be monitored (a signal that is present on the
DL and DLV devices). If you answer both Question 4.05 and 4.06 YES for DL or DLV
devices, NETGEN automatically resets both answers to NO.
NETG EN displays Question 4.07 for every line/circuit having enough information available (for example, CSR and vector) to enable the line/circuit.
>* 04.07 Set the state for dev ON when loading the network? [D=N] [YIN]

If you answer this question YES, the DEV device driver (DDM)
and al I necessary databases wi I I automatically be loaded into
memory and the state set to ON at network initial ization time.

NETGEN Dialog:

NET

5-23

If you want the state on during normal network operation,
answering this YES wi I I save the NCP or VNP command required
to set the state. If dev is only to be used occasionally,
answer NO.
Note that the state can later be changed with CFE so that it
wi I I be either ON or OFF at network initial ization time.

Answer this question YES if you want your lines/circuits enabled at network initialization time.
Uyou specify devices that support multipoint capabilities, NETGEN displays Question
5.00.
>* 05.00 Is dev multipoint? [O=N]

[YIN]:

Answer this question YES only if the I ine is a master or
slave port on a multipoint line.
>* 05.01 Is dev the multipoint master? [O=N]

[YIN]:

Answer this question YES if this I ine is the master, and
NO if it is a slave. The multipoint master is the station that
controls the I ine by pol I ing the slave ports for activity.

NETGEN displays Question 5.02 if the device is a multipoint master.
>* 05.02 How many tributaries should dev pol I? [0 R:l.-32. 0:8.]:

Specify the number of tributaries to be connected to the
multipoint I ine. This is the number of tributaries which wi I I
be pol led by the master.
Each multipoint tributary on a I ine should be identified by
a unique tributary address. The addresses used for the
tributaries on this I ine wi II be 1 through "number of
tributaries" specified.
Note that the tributary address can later be changed with NCP,
VNP, or CFE.

NETGEN displays Question 5.03 if the device is a multipoint tributary.
>* 05.03 What

is the tributary address of dev [0 R:l.-255. 0: den :

Specify the tributary address which is to identify this
multipoint slave to the multipoint master. Each multipoint
slave should be identified by a unique tributary address.
Note that the tributary address can later be changed with
NCP, VNP, or CFE.

5-24

DECnet-RSX Network Generation and Installation Guide

Questions 5.04, 5.05, and 5.06 are displayed only if you specified a PCL device in Question 1.00.
>* 05.04 How many PCls wi I I communicate on the bus? [0 R:1.-16. 0:8.]:

The PCll18 provides ful I duplex communication between a
number of systems on a paral lei Time Division Multiplexed
(TOM) bus.
If this system is to be a ful I OECnet routing node, specify
the maximum number of PCls which wi I I communicate over this
bus (no more than 16). Include this PCl (PCl-x.z) in this
number.
If this system is to be a OECnet non-routing node, this number
must be "1".
>* 05.05 What is the TOM bus address for the local PCl?

[ 0 R: 1 . -31. 0: deJ1 :
Specify the address used by other PCls to communicate with
the local PCl. This is i·ts address on the Time Division
Multiplexed (TOM) bus. This address must be the same address
that was set in the PCl hardware during installation.
PCl-x.z is always used to address the local PCl receiver.
It can be used to provide a bUI It-in loopback capabi I ity.
Note that the TOM bus address can later be changed with NCP,
VNP, or CFE.
>* 05.06 What is the TOM bus address for PCl-x.z [0 R:1.-31. O:dej]:

Specify the address used by the local PCl to communicate with
the PCl on the other system. This is the PCl's address on the
Time Division Multiplexed (TOM) bus. This address must be the
same address that was set in the PCl's hardware during
installation.
Note that the TOM bus address can later be changed with NCP,
VNP, or CFE.

Based on the value you specified in Question 5.04, Question 5.06 will be repeated
appropriately.
NETGEN displays Questions 5.07 through 5.12 for DECnet/PSI systems, if you specified D LM in Question 1.00.
>* 05.07 How many tributaries should

OlM~y

support?

[0 R:1.-32. 0:1.]:
The OlM al lows your OECnet system to access remote OECnet
systems using a PSI ci rcuit as the communications channel.
It is implemented as a multipoint I ine, where each tributary
connects through a distinct PSI circuit.

NETGEN Dialog:

NET

5-25

For the number of tributaries, specify the number of DECnet
systems to which this system wi I I connect using PSI circuits.
For each DLM controller, you may specify up to 32 tributaries.
As it wi I I require a new NETGEN to change the number you specify
you may wish to define extra tributaries to use for future
expansion.

You need one tributary for each switched or permanent circuit. If you specified more
than one D LM controller in Question 2.00, NETGEN displays this question for each controller, thus affecting the total number of tributaries you specify.
>* 05.08 Does DLM-x define only Switched Virtual Circuits (SVC)

[D=N]? [YIN]:
A Switched Virtual Circuit (SVC) is an X.25 circuit which
al lows this node to dynamically define the circuit destination
to the Packet Switching Data Network (PSDN). A Permanent
Virtual Circuit (PVC) is a circuit that has the destination
permanently defined upon subscribing to the PSDN.
Answer this question YES, if you have subscribed to only SVC's.
If you answer this question NO, you wi I I be asked on a per
ci rcuit basis whether it defines an SVC.

Each X.25 circuit must be either a SVC or a PVC.
If you answer Question 5.08 NO, you will be asked 5.09.
>* 05.09 Does DLM-x.y define a Switched Virtual Circuits (SVC)

[D=N]? [Y IN] :

A Switched Vi rtual Ci rcuit (SVC) is an X.25 ci rcuit which
al lows this node to dynamically define the circuit destination
to the Packet Switching Data Network (PSDN). An answer of NO
wi II indicate that the ci rcuit is to be a Permanent Vi rtual
Circuit (PVC), and that the destination was permanently defined
upon subscribing to the PSDN.

NETGEN displays Questions 5.10 through 5.12 only if you specified a Switched Virtual
Circuit in Question 5.09 or 5.08.
>* 05.10 Wi I I DLM-x.y be used for incoming cal Is? [D=N] [YIN]:

An answer of YES indicates that this circuit wi I I accept
only incoming cal Is. An answer of NO indicates that this
ci rcuit wi I I be used only to place outgoing cal Is.
Note that this attribute can later be changed with CFE.

All SVC circuits in X.25 must be classified as incoming only or outgoing only to establish who places the call. Once established, circuits are full duplex.

5-26

DECnet-RSX Network Generation and Installation Guide

NETGEN displays Question 5.11 only if you defined this circuit as an outgoing circuit
in Question 5.10.
>* 05.11 OTE Address which OLM-x.y cal Is? [O=none]
[S R:O.-16.]:
Specify the ful I address of the remote OTE to be cal led
over th i sci rcu it. The fu I I address cons i sts of from 1 to 16
hexadecimal digits, and is comprised of the address which
the Packet Switching Data Network (PSON) assigned to the
remote OTE when it subscribed to the PSDN, fol lowed by the
subaddress as defined by the remote DTE system manager. The
two are concatenated to form one string of characters. The
number of characters for each field wi I I vary depending upon
the PSON and the requirements of the remote OECnet node.
This SVC has been set up to use the default packet size and
default window size that was set up upon subscribing to the
PSON. This packet size is independent of the segment size used
by the OECnet Routing process. If a packet size or window size
negotiation faci I ity is avai lable for this PSDN and if you have
subscribed to It, you may make use of the negotiation faci I ity
by changing the default values with CFE.
Note that the OTE address can later be changed with CFE.
>* 05.12 Address of the DTE that may cal I DLM-x.y [S]:
Specify the address of the remote OTE that this ci rcuit wi I I
accept ca I Is from. I faze ro I s spec If i ed, t his c i rcu i t wi I I
accept cal Is from any remote OTE. An incoming cal I must sti I I
have a val id destination subaddress. The remote OTE address
consists of from 1 to 16 hexadecimal digits, and is the address
which the Packet Switching Data Network (PSON) assigned to the
remote DTE when it subscribed to the PSON. The number of digits
in the DTE address wi I I vary depending on the PSDN.
This SVC has been set up to use the default packet size and
default window size that was set up upon subscribing to the
PSON. This packet size is independent of the segment size
used by the DECnet Routing process. If a packet size or
window size negotiation faci I ity is avai lable for this PSDN
and if you have subscribed to it, you may make use of the
negotiation faci I ity by changing the default values with CFE.
Note that the OTE address can later be changed with CFE.
>; <EOS> Do you want to:
>*
_ <RET>-Continue, S-Skip, R-Repeat section, P-Pause,
E-Exit [S]:

NETGEN Dialog:

NET

5-27

5.4

Defining the CEX System

This section defines the Communications Executive (CEX) and related components. Figure 5-4 illustrates the sequence of questions for defining the CEX system on RSX-IIM
and RSX-llS systems. For RSX-IIM-PLUS systems, this section does not ask any questions; it just creates the build files for the device driver modules (DDMs).

TW0172

Figure 5-4:

5-28

NET - Section 4 - Define the CEX System

DECnet-RSX Network Generation and Installation Guide

>
>
>

============================================================
NET - Section 4 - Define the CEX System
============================================================

Creating bui Id fi les for CEX, the Communications Executive.

NETGEN displays Questions 1.00 and 1.01 for RSX-11M and RSX-11S systems only.
CEX is built for all systems, except in component mode. CEXPAR is the name of the system partition for CEX. This partition is created by using VMR commands after SYSGEN
but before you load the system. For more information see Section 3.2.5.11.
>* 01.00 Base add ress fo r pa rt it ion CEXPAR [0 R: 14000-range-top D: default] :

Specify the base address for the CEX partition. This partition
must reside enti rely within the target system's RSX Executive
address space (i .e. below 120000).
Speci fying range-top provides the minimum amount of space
for the CEX that supports the options that you have selected.
Spec i fy i ng default prov i des enough space for the CEX wh i ch
wi I I be generated for your system plus additional space to be
used by CEX as a buffer pool.
The space that is left over after CEX is loaded into the
partition is not wasted. This space wi I I be used by CEX as
a buffer pool. If CEX needs more buffer pool space than can
be accommodated by the CEX partition, it wi II dynamically
al locate space out of DSR.

Depending on the number of lines selected and the types of devices required for your
system, the range-top displayed varies. This value will always reflect the base address
required for the minimum partition size for your system.
The default displayed varies with the type of system. All systems are considered large
unless they are DECnet only and have only one line, in which case they are considered
small. The default values are:
Executive

CEXSize

Values

16K
16K
20K
20K

Small
Large
Small
Large

73500
73000
113500
113000

If you selected the range-top in Question 1.00, you will have just enough space for the

smallest CEX that is legal for your system, and NETGEN will not display Question 1.01.
>* 01.01 Length of part it ion CEXPAR [0 R:4300-max-Iength D:default]:

Spec i fy t he I engt h of the CEX pa rt it ion. Spec i fy max-length
if the CEX partition extends out to the end of the RSX
Executive's address space. This is generally advisable, since
any extra space at the end of the partition wi I I be used by CEX
as a buffer pool, and wi I I not be wasted.

NETGEN Dialog:

NET

5-29

NETGEN displays the following question only if you answered NO to Questions
9.00 and 10.00 (standard function system option and all components option) in NET
Section 1.
>* 02.00 Do you want network event logging? [D=N] [YIN]
Answer the question YES if events detected by the network
software are to be logged. These events wi I I be logged on the
console terminal unless logging is changed using NCP, VNP,
or CFE.
It is recommended that you answer this question YES in order
to faci I itate network fault detection/correction and network
management.

The following statement and Question 3.00 are displayed in question and answer mode
or restore mode, if you specified devices that require software DDCMP.
Creating bui Id fi les for STCRC, the CRC-16 calculation routine.
>* 03.00 Does the target system have a KG-11? [D=N] [YIN]
Answer the question YES only if the target system has the KG-11
communications arithmetic option. The KG-11 option computes the
cycl ic redundancy check characters which are used by some CEX
products for error detection in serially transmitted and
received data. If you answer NO, a software algorithm wi I I
be used.
Cycl ic redundancy check (CRC-16) characters are used to verify
the correctness of information transmitted across the network.
Creating bui Id fi les for CETAB, the CEX configuration tables.
NOTE
If you are performing a NETGEN for an 11144 or 11170 processor, you

may not want to use the KG-II. On these processors, selecting the
KG-II will actually reduce your throughput below that of the software CRC routine. To use the software CRC routine, answer NO to
Question 3.00.
The actual CETAB.MAC file is created in Section 8. The build files created here are used
to assemble and build the CETAB.MAC file.
Creating bui Id fi les for the DDMs, the device driver modules.
The device driver modules were defined in NET Section 3.
>; <EOS>
>*

5-30

Do you want to:
<RET>-Continue, S-Skip, R-Repeat section, P-Pause,
E-Exit [S]:

DECnet-RSX Network Generation and Installation Guide

5.5

Defining the Communications Executive Support Components

This section defines the support components - NTINIT, NTL, MLD and EVL/EVC. Figure 5-5 illustrates the sequence of questions for defining the Communications Executive support components.

NET BUILDS SUPPORT
FOR CEX COMPONENTS
NETWORK INITIALIZER
NETWORK LOADER
GENERAL MICROCODE LOADER
NETWORK EVENT LOGGER

TW0173

Figure 5-5:

NET - Section 5 - Define the Comm Exec Support Components

NETGEN Dialog:

NET

5-31

>;
>;
>;
>*

============================================================
NET - Section 5 - Define the Comm Exec Support Components
============================================================
01.00 Do you want any of the Support Components? [D=N] [YIN]:

The Communications Executive Support Components are:
NTINIT - The network initial izer.
NTL - The network loader.
MLD - The general microcode loader.
EVL/EVC - The event logger and col lector.
If you answer YES to this question, you wi I I be asked about
each component individually. In general, you should only
answer YES when there is a clear reason to bui Id or rebui Id
one or more of these components.

NETGEN displays Question 1.00 only in component mode. If you answer NO, you
immediately jump to the End of Section (EOS) question.
You should answer YES only when there is a definite reason to build or rebuild one of
the components, for example, to add components not originally selected or to build a
patched version of a component.
The following statement is displayed in question and answer mode.
Creating bui Id fi les for NTINIT, the network initial izer.

Question 2.00 is displayed in component mode.
>* 02.00 Do you want NTINIT? [D=N]

[YIN]:

NTINIT is the network initial ization task. It is responsible
for performing all initialization which is required before
the network is started.
If you wish to rebui Id NTINIT, answer this question YES,
otherwise answer NO.

In question and answer mode or restore mode, NTINIT is always built. In component
mode, rebuilding NTINIT is optional.
The following statement is displayed in question and answer mode.
Creating bui Id fi les for NTL, the network loader.

Question 3.00 is displayed in component mode.
>* 03.00 Do you want NTL? [D=N] [YIN]:
NTL is the network loader task. It is responsible for loading
and unloading the Comm Exec and al I of the processes.
If you wish to rebui Id NTL, answer this question YES,
otherwise answer NO.

5-32

DECnet-RSX Network Generation and Installation Guide

In question and answer mode or restore mode, NTL is built for RSX-IIMIM-PLUS systems. In component mode, rebuilding NTL is optional. NTL is never built for an RSXlIS system.
The following statement is displayed in question and answer mode.
Creating bui Id fi les for MLD, the general microcode loader.

NETGEN always displays Question 4.00 in component mode because it does not know
if you have a device requiring MLD.
>* 04.00 Do you want MLD? [D=N]

[YIN]:

MLD is the general microcode loader. It is responsible for
loading UNA, KMC, and KMS devices with the proper microcode.
If your target system has these devices in Its configuration,
and you wish to generate a new version of the general microcode
loader, then answer this question YES.
If your target system has none of these devices in its
configuration, answer this question NO.
If you specified a device requiring microcode to be loaded, NETGEN displays Question

4.00.
The following statement is displayed in question and answer mode and restore mode if
you specified event logging in Section 4 Question 2.00 or if the target system is RSX11M-PLUS.
Creating bui Id files for EVL/EVC, the event logger and
collector.

Question 5.00 is displayed in component mode.
>* 05.00 Do you want EVL/EVC? [D=N]

[YIN]:

EVL IS the event logging process. It is responsible for
detecting those events which are to be logged, and recording
them.
EVC is the event collecting task. It is responsible for
collecting the events logged by EVL, and for sending them to
the appropriate sinks.
If your target system was generated with event logging, and
you wish to generate new versions of EVL and EVC, then answer
this question YES.
If your target system was not generated with event logging,
answer this question NO.

In question and answer mode or restore mode, EVL and Eve are always built if you
specified event logging in Section 4, Question 2.00 or if the target system is RSX -11 MPLUS. For RSX-IIM/S systems, you cannot add or rebuild EVL and Eve in component
mode if you did not specify event logging in the original NETGEN.

NETGEN Dialog:

NET

5-33

>* 05.01 How many Event Buffers should EVL use? [0 R:4.-50. 0:6.]:
The Event Logger Process (EVL) must be bui It with some number
of internal event buffers which wi I I be used to hold events
detected by the Network. If events occur often enough so that
there are sometimes more concurrent events than there are event
buffers, some of the event information wi I I be lost.
EVL may be rebui It at a later time with more event buffers by
performing a Component Generation.

Questions 5.02 and 5.03 are asked only when the generation is being performed for an
RSX-l1S system and EVL/EVC support has been chosen by answering YES to Question
6 (in component mode) or requesting event logging in Section 4 Question 2.00 (in question and answer mode). For RSX-IIMIM-PLUS systems, monitor and remote event logging are always included if you chose EVLlEVC support.
>* 05.02 Should EVC support monitor event logging? [O=N] [YIN]:
I

If you answer this question YES, EVC wi I I support the abl I ity to
send event records to a user written task.
If you answer NO, events wi I I not be logged to a user task.
This support may be changed at a later time by performing a
Component Generation.
>* 05.03 Should EVC support remote event logging? [O=N] [YIN]:
If you answer this question YES, EVC wi I I support the abi I ity
to send event records to an Event Receiver task on a remote
node.
If you answer NO, events wi I I not be logged to remote nodes.
This support may be changed at a later time by performing a
Component Generation.
>; <EOS> 00 you want to:
>*
<RET>-Continue, S-Skip, R-Repeat section, P-Pause,
E-Exit [S]:

5-34

DECnet-RSX Network Generation and Installation Guide

5.6

Defining the System Management Utilities

This section defines the system management utilities - NCP, NMDRV, NMVACP, CFE,
NDA, VNP, LOO, and EVF. Figure 5-6 illustrates the sequence of questions for defining
the system management utilities.

NET BUILDS SUPPORT FOR
SYSTEM MANAGEMENT UTILITIES
NETWORK CONTROL PROGRAM
NETWORK MANAGEMENT DEVICE DRIVER
NETWORK MANAGEMENT VOLATILE ACP
CONFIGURATION FILE EDITOR
NETWORK CRASH DUMP ANALYZER
VIRTUAL NETWORK PROCESSOR
LOOP TESTER PROGRAM
EVENT FILE INTERPRETER

TW0174

Figure 5-6:

NET - Section 6 - Define the System Management Utilities

NETGEN Dialog:

NET

5-35

>
>
>

============================================================
NET - Section 6 - Define the System Management Uti I ities
============================================================

>* 01.00 Do you want any of the System Management Uti I ities?

[D=N] [YIN]:

The System Management Uti I ities are:
NCP - The network control program.
NMDRV - The network management driver.
NMVACP - The network management ACP.
CFE - The configuration fi Ie editor.
NDA - The network dump analyzer.
VNP - The vi rtual network control program.
LOO - The loop test sender task.
EVF - The event fi Ie interpreter.
If you answer YES to this question, you wi I I be asked about
each component individually. In general, you should only
answer YES when there is a clear reason to bui Id or rebui Id
one or more of these components.

NETGEN displays Question 1.00 only in component mode. If you answer NO, you
immediately jump to the end-of-section (EOS) question.
The following statement is displayed in question and answer mode.
Creating bui Id fi les for NCP, the network control program.

Question 2.00 is displayed in component mode.
>* 02.00 Do you want NCP? [D=N]

[YIN]:

NCP is the network control program. It al lows the system
manager to perform a I lon-I i ne network management funct ions.
If you wish to rebui Id NCP, answer this question YES, otherwise
answer NO.

In question and answer mode or restore mode, NCP is required for RSX-IIM/M-PLUS
systems. For RSX-llS systems, it is optional.
In component mode, rebuilding NCP is optional.
The following statement is displayed in question and answer mode.
Creating bui Id fi les for NMVACP and NMDRV, the Network
Management Volati Ie ACP and driver.

Question 3.00 is displayed in component mode.
>* 03.00 Do you want NMVACP and NMDRV? [D=N]

[YIN]:

NMVACP and NMDRV are the Network Management Volati Ie ACP
and its associated driver. They are responsible for
performing many of the operations required by the various
NCP commands.
If you wish to generate a new version of NMVACP and NMDRV,
answer the question YES.

5-36

DECnet-RSX Network Generation and Installation Guide

In question and answer mode or restore mode, NMVACP and NMDRV are required for
RSX-IIMIM-PLUS systems. In component mode, NMVACP and NMDRV are optional.
For RSX-llS systems, NMDRV is built only if you have specified event logging;
NMVACP is never built.
The following statement is displayed in question and answer mode.
Creating bui Id fi les for CFE, the configuration fi Ie editor.

Question 4.00 is displayed in component mode.
>* 04.00 Do you want CFE? [D=N]

[YIN]:

CFE is the configuration fi Ie editor. It al lows the system
manager to make changes to the network's permanent database
(CETAB.MAC) .
If you wish to rebui Id CFE, answer this question YES,
otherwise answer NO.

In question and answer mode or restore mode, CFE is required for all systems. In component mode, rebuilding CFE is optional.
The following statement is displayed in question and answer mode.
Creating bui Id fi les for NDA, the network dump analyzer.

Question 5.00 is displayed in component mode.
>* 05.00 Do you want NDA? [D=N]

[YIN]:

NDA is the network dump analysis program. It provides the
system manager with a tool which can analyze a system crash
dump and print the network databases in a user readable
format in much the same way the RSX crash dump analyzer
(CDA) does for the system databases.
If you wish to rebui Id NDA, answer this question YES,
otherwise answer NO.

In question and answer mode or restore mode, NDA is required for all systems. In component mode, rebuilding NDA is optional.
The following statement is displayed in question and answer mode.
Creating bui Id fi les for VNP, the virtual network control program.

Question 6.00 is displayed in component mode.
>* 06.00 Do you want VNP? [D=N]

[YIN]:

VNP is the virtual network control program. It al lows the
system manager to perform network management functions on
a system image fi Ie, rather than on the running system.
If you wish to rebui Id VNP, answer this question YES,
otherwise answer NO.

NETGEN Dialog:

NET

5-37

In question and answer mode or restore mode, VNP is required for RSX-l1S systems.
For RSX-IIMIM-PLUS systems, VNP is optional.
NETGEN displays the following statement for RSX-I1M and RSX-I1M-PLUS systems
in question and answer mode.
Creating bui Id fi les for LaO, the loop test sender task.

Question 7.00 is displayed in component mode.
>* 07.00 Do you want LOa? [D=N] [YIN]:
LOa is the loop test sender task. It is responsible for
performing both node and circuit level loopback tests, in
response to an NCP LOOP command.
For loop node tests, LOa sends to the MIR server task. For
loop circuit tests, LOa sends to the LIN server task.
If you wish to bui Id LOa, answer this question YES,
otherwise answer NO.

LOO is never built for RSX-llS systems.
If you are in question and answer mode and you answer Question 7.00 YES, NETGEN

displays the following statement for RSX-IIM systems. However, if you have selected
other options that automatically include DLX, then the following statement is not displayed.
Note that to enable LOa to perform loop circuit tests, the
DLX process must be selected in this generation.

The following statement is displayed in question and answer mode if you answered
YES to Questions 9.00 or 10.00 (the standard function system option or all components option) in NET Section 01.
Creating bui Id fi les for EVF, the event fi Ie interpreter.

Question 8.00 is displayed in component mode or if you answered NO to Question
9.00 or 10.00 in NET Section 01. EVF is never built for RSX-11 S systems.
>* 8.00 Do you want EVF? [D=N] [YIN]:
EVR and EVC can write events out to an event fi Ie. This
uti I ity interprets the event fi Ie, and displays the
requested information.
If you wish to bui Id EVF, answer this question YES, otherwise
answer NO.
>; <EOS> Do you want to:
>*
<RET>-Continue, S-Skip, R-Repeat section, P-Pause,
E-Exit [S]:

5-38

DECnet-RSX Network Generation and Installation Guide

5.7

Defining the CEX Products

This section displays questions that are specific to DECnet and PSI if you are including
PSI components in your DECnet system.
>; ============================================================
>; NET - Section 7 - Define the CEX Products
>; ============================================================

Questions concerning the fol lowing CEX Products wi I I now be
asked.
DECnet
PSI
At this point, network generation continues with the text for the DEC portion of
NETGEN (Chapter 6), followed by the PSI portion of NETGEN (Chapter 7) if you have
purchased a PSI license. After completion of these two procedures, NETGEN continues
with NET Section 8. Complete the CEX System Definitions.

NETGEN Dialog:

NET

5-39

5.8

Completing the CEX System Definitions

This section defines the Direct Line Access Controller (DLX), and creates the
CETAB.MAC file. Figure 5-7 illustrates the sequence of questions for completing the
CEX system definitions.

TW0175

Figure 5-7:

5-40

NET - Section 8 - Complete the CEX System Definitions

DECnet-RSX Network Generation and Installation Guide

>; =============================================================
>; NET - Section 8 - Complete the CEX System Definitions
>; =============================================================

Creating bui Id fi les for DLX, the Direct Line Access process.

If you have selected components that require DLX, the above statement is displayed in
question and answer mode or restore mode.
If none of the components you selected require DLX, NETGEN displays Question 1.00
in question and answer mode. It is always displayed in component mode.
>* 01.00 Do you want DLX? [D=N]

[YIN]:

DLX, the Direct Line Access controller, provides a user
interface to I ines at the data I ink level. This is requi red
for any services which must bypass the normal network data
channels.
For DECnet, such services include down-line system loading,
up-I ine system dumping, and NCP I ine loop tests. There
is also a user QIO interface which al lows user tasks to access
I ines via the data I ink protocol (usually DDCMP).
If no such services have been selected during this generation,
but you plan on adding these services at a later date by
performing a Component mode generation, you must select DLX for
your system now.
>* 02.00 What

is the Large Data Buffer (LDB) size?
[D R:192.-1484. D:292.]:

Large Data Buffers (LDBs) are used to transmit and receive
messages across the network. The LOB size determines the
largest segment which can be transmitted or received. This
does not affect the size of the largest user data transfer
request, but it can affect performance.
Specify the size of the Large Data Buffers (LDBs) as a number of
bytes. The size chosen for LDBs should be the same throughout
the network.
A size of 292 is large enough for al I normal network traffic.

The default for this question is usually 292. The default provides satisfactory performance for normal network traffic. A value of 292 accommodates 1/2 disk block (256
bytes) of user data. This allows space for data and DEenet protocols. For systems that
have a high volume of file transfers, a size of 548 allows a full disk block (512 bytes) to
be transferred at a time.
The following factors can affect optimum LDB size:
•

Large LDBs are good for bulk data transfers, providing you have sufficient available
memory.

NETGEN Dialog:

NET

5-41

•

Large LDBs have a higher probability of an error due to line noise for a given
packet, and longer retransmission times. For high speed lines, this may not be significant.

•

If you choose to have large LDBs for efficient file transfers, you may only have a
small number of buffers, depending on the amount of memory available. Consequently, processes may have to wait for buffers.

>;<EOS> Do you want to:
>*
<RET>-Continue, S-Skip, R-Repeat section, P-Pause,
E-Exit [S]:

5-42

DECnet-RSX Network Generation and Installation Guide

5.9

Building the CEX System

This section actually builds the system you have specified. It is repeated three times,
once each for NET, DEC, and PSI. If all object kits were placed on the NETGEN disk during PREGEN, this section proceeds to completion without user intervention.
>;
>;
>.,
>;

=============================================================
NET - Sect ion 9 - Bu i I d the CEX System at time on
date

=============================================================
AI I questions have now been asked and the selected components
wi I I now be bui It. This may take from one to three hours,
depending on the selection of components and the system you
are running on.
If an operator's presence is required to switch disks during the build process, NETGEN

displays the following statement.
Note that the NETGEN disk does not contain al I of the requi red
object fi les. An operator must be present to mount each object
disk as it is required.
>* 01.00 network object disk device? [dduu:, D=dej] [S]:

Specify the disk drive where the object disk is loaded.
This may be the same drive as is used for the NETGEN diSk,
or it may be a different drive.
If the disk has not yet been mounted, NETGEN wi I I mount It. If
another disk is currently mounted in the drive, that disk wi I I
be dismounted.
Th is disk must conta i n the network obj ects.
If you plan to use a disk drive on which another disk is mounted, do not replace the
disk until you have answered Question 1.00. If the second disk is spun up before you

answer Question 1.00, problems will result. After you answer Question 1.00, NETGEN
displays the following statement.
Dismount i ng device. P I ease remove the disk and rep I ace it wit h
the p rope r network ob j ect disk.
>* 01.01 Is device ready? [D=N] [YIN]:

Loa d the dis kin t 0 d r i ve device and s tar t the d r i ve be for e
answering YES to this question.
If the device has not been mounted, NETGEN displays Question 01.02.
>* 01.02 Label for device? [D=default label] [S R:0-12J:

Spec i fy the I abe I to be used when mount i ng device. I f the disk
was an object disk created by PREGEN to hold only the network
o b j e c t s, the I abe I w i I I be default. If not, the I abe I w i I I ha ve
been user defined.
Mount i ng device.

NETGEN now mounts the device specified. Where network is shown, NETGEN displays either Network, DECnet, or PSI.
NETGEN Dialog:

NET

5-43

5.10

Generation Cleanup

This section performs final NETGEN processing.
>;
>;
>;

============================================================
NET - Section 10 - Generation Clean-up
============================================================

Creating NETCFG.TXT,
Creating NETCFE.CMD,
Creating NETINS.CMD,
Creating NETREM.CMD,

the network configuration description fi Ie.
the network configuration command fi Ie.
the network installation command fi Ie.
the network removal command fi Ie.

The network HELP fl Ie I ibrary has been placed on your target
diSk, as:
DB01: [1,2]NETHLP.ULB
T his f i I e s h0 u I d be mo v edt 0 you r tar get s y stem's lib r a r y
device (LB:).
>; ============================================================
>; NET - RSX-11M/S/M-PLUS CEX System Generation Procedure
>;
Stopped at time on date
>; ============================================================

At this point you have completed the NETGEN procedure. You should now proceed to
install your network using the NETINS.CMD file generated by NETGEN. This procedure is described in Chapter 9.

5-44

DECnet-RSX Network Generation and Installation Guide

6
NETGEN Dialog: DEC

DEC is the second part of the NETGEN dialog. NETGEN displays a series of questions
relating to DECnet components immediately following NET Section 7.
The DEC dialog represents the text NETGEN displays when performing an initial generation in question and answer mode. A component mode or a restore mode generation
sometimes displays a slight variation in the wording of specific questions. For example,
the defaults sometimes vary from those displayed in this dialog. Your responses should
always be guided by the question and text displayed on your terminal.

6.1

Defining the Target and Remote Nodes

In this section of NETG EN, you are expected to specify sets of information that identify
the node you are generating and the nodes with which it will communicate. In addition, you will supply information about certain node and routing characteristics. Figure 6-1 illustrates the sequence of questions for defining the target and remote nodes.

6-1

TW0176

Figure 6-1:

6-2

DEC - Section 1 - Define the Target and Remote Nodes

DECnet-RSX Network Generation and Installation Guide

>; ============================================================
>; DEC - DECnet CEX Product Generation Procedure
>;
St art i ng quest ions at time on date
>; ============================================================
>; Copyright (C) 1985 by
>; Digital Equipment Corporation, Maynard, Mass.

Generate the DECnet CEX Product for your RSX-11M/S/M-PLUS target
system.
>; =============================================================
>; DEC - Section 1 - Define the Target and Remote Nodes
>; =============================================================

If you are in component mode, NETGEN displays the following statement and then dis~

plays Question 6.00.
dduu: [x,54]CETAB.MAC is being scanned to define your target node.

NOTE

To change most parameters defined in this section, you can use NCP,
VNP, or CFE. You must do a component mode generation to change
the parameters defined in Questions 6.00 and 8.00. To change parameters defined in Question 4.00, you must do a complete new generation
or a restore mode generation.
>* 01.00 What

is the target node name? [S R:0-6]:

Enter the node name for the DECnet node being generated. This
must be 1 to 6 alphanumeric characters, at least one of which
must be an alphabetic (A-Z).
AI tho ugh i tis not r e qui red, i tis h i g h I Y r e c onme nd edt hat t his
name be used by al I other nodes in the network to identify this
node.
Note that the node name can later be changed with CFE.
>* 02.00 What

is the target node address? [S R:0.-8.]:

Enter the node address that has been assigned for the node being
generated. The format is:
AREA.NODE~UMBER

(Example: 3.17)

NETGEN Dialog:

DEC

6-3

The node address, consisting of the AREA and NODE~UMBER fields,
wi I I be the address of this node within the network. If you are
entering values in both the AREA and NODE~UMBER fields, you
must also enter the dot between the fields.
Note that the node address can later be changed with CFE.
>* 03.00 Target node ID? [D=None]

[S R:0.-32.]:

Enter the 10 string for the node being generated. This string
may be from 0 to 32 characters. Note that this string must not
contain any single-quote ('), double-quote (") or comma (,)
characters.
The node 10 may be any arbitrary string which is useful in
identifying the node.
Note that the node 10 can later be changed with CFE.
The node 10 is not passed on to other nodes via routing
messages.
If you have purchased a full function DECnet kit, NETGEN displays Question 4.00 in

question and answer mode. This question is not displayed in component mode.
>* 04.00 Do you want to generate a routing node? [D=N]:
A routing node maintains routing databases and sends and
receives routing messages. It can support multiple DECnet
I ines and circuits.
A non-routing node can provide access to a network through an
adjacent routing node whi Ie maintaining no routing databases of
its own. Because it does not maintain any routing databases and
does not send or rece i ve rout i ng messages, it requ i res fewer
system resources.
When generating a non-routing node, the fol lowing should be
noted:
It is recommended that only one DECnet I ine be generated. If
more than one I ine is generated, only one I ine may be
enabled at a time.
If the I ine is a multipoint master, it is recommended that it
has only one drop. If more than one drop is generated, only
one drop may be enabled at a time.
Answer this question YES if you wish to generate a routing
node, and NO if you wish to generate a non-routing node.

NETGEN displays Question 4.01 if you specified a routing node.
>* 04.01 Do you want to generate a level 2 routing node?
[D=N] [Y IN]

A level 2 routing node maintains a routing database for al I
areas in the network as wei I as a routing database for al I
nodes in its local area. It al lows nodes in its local area to

6-4

DECnet-RSX Network Generation and Installation Guide

access nodes in the network that are outside of the local area.
A level 1 routing node maintains a routing database for al I
nodes in its local area but does not have the abi I Ity to
di rectly access nodes outside of Its local area. To access
nodes outside of its local area, it must route messages through
a level 2 routing node.
Because a level 1 routing node does not al low di rect access to
ot he r a reas of the netwo r k, It requ i res I ess code and fewe r
databases.
Answer this question YES if you wish to generate a level 2
routing node, and NO if you wish to generate a level 1 routing
node.

NETG EN displays Question 5.00 if you are generating a routing node. In Question 5.00,
n is the node number that you specified in response to Question 2.00.
> 05.00 Highest node number in this area [0 R:n -1023. O:n.J:

Enter the highest node number used by any node in this area.
This number affects the size of the Internal routing database
used by this node. It also controls the size of the routing
messages exchanged between nodes.
More efficient operation can be obtained by setting the node
numbers of the nodes in the area in the range 1 to the number
of nodes in the area. You may wish to make the highest node
number sl ightly greater than the number of nodes in the area,
to al low for expansion of the network.
Note that al I nodes in the area should use the same value for
the highest node number.
The value for the highest node number can later be changed
with CFE.

NETGEN asks Question 5.01 only if you are generating a Level 2 routing node. The n in
Question 5.01 is replaced by the area number used in Question 2.00.
>*

05.01 Highest area number in network? [0 R:n.-63. D:nJ:

Enter the highest area number used by any node In the network.
This number affects the size of the Internal databases used by
this node.
More efficient operation can be obtained by setting the area
numbers of the areas in the network In the range 1 to the number
of areas. The highest area number would then be equal to the
number of areas in the network.
You may wish to make the highest area number slightly greater
than the number of areas in the network, to allow for expansion
of the network.
Note that al I nodes in the network should use the same value for
the highest area number.

NETGEN Dialog:

DEC

6-5

The value for the highest area number can later be changed with
CFE.

NETGEN displays the following statement, and displays Question 6.00 in question and
answer mode if you answered NO to Question 9.00 and 10.00 (the standard system
option and all components option) in Section 1. If you included extended network support, NETGEN will not display Question 6.00 in component mode and will automatically include this support.
To include support for network command terminals and products
layered on DECnet, you must fi rst include the required extended
network support.
>* 06.00 Do you want to include this extended network support? [D=Y]:
Answer this question YES, to include the extended network
support for the fol lowing options:
- Network Command Terminal server (NCT)
- Remote Network Command Terminal host support (RTH)
- Products layered on DECnet:
- X.25 Gateway Access
- SNA Gateway Access
If you do not include the extended network support, you wi I I
not be able to include these features in your network.
Including this support wi I I cause ECl and NETACP to increase
in size by a total of 1300. words.
You may add the extended network support later by performing a
Component Mode generation.
>* 07.00 Remote node name? [<RET>=Done] [S R:0-6]:
You may, at this time, define a series of node names which
correspond to nodes in your network. When the network IS
loaded, these node names wi I I automatically be set. Enter a
carriage return when you have no more remote node names to
define.
Each node name wi I I be assigned to a node address. Standard
procedure is to start with the node name for the lowest node
address used, and to progress from there to the highest node
address.
It is recommended that each node in the network assign the
same node name to each individual node, reducing the confusion
of moving from node to node.
Remote node names can be added or deleted using NCP, VNP, or
CFE.

Questions 7.00 and 7.01 are repeated until you indicate you are done by pressing
<RET>.
>* 07.01 What is the remote node address? [S R:O.-B.]:

6-6

DECnet-RSX Network Generation and Installation Guide

Enter the node address that node name has been assigned
in the network. The format is:
AREA.NODE~UMBER
(Example: 3.18)
Where AREA is the area number within the network and
NODE~UMBER is the node number within that area. The area
must be a decimal number in the range of 1 to 1023. If only
NODE~UMBER is specified, the area that was specified for the
node being generated, wi I I be used as the default area number.
The node number within the area must be a decimal number in the
range 1 to 1023. If you are entering values in both the AREA
and NODE~UMBER fields, you must also enter the dot between
the fields.
Each node address defines a unique node in the network, and must
be assigned when the network is first set up.
The remote node address can be changed using NCP, VNP, or CFE.

The variable node-name refers to the name you specified in Question 7.00.
>* 08.00 Do you want the language interface libraries? [D=N] [YIN]:
The MACRO user interface al lows user written MACRO-II tasks
to interface to the DECnet software. The MACRO user library
contains the macros which a programmer may use to code these
tasks.
The FORTRAN/COBOL/BASIC+2 object library al lows user tasks
written in these high level languages to interface to the DECnet
software. The object I ibrary contains the user interface
routines and the Network Fi Ie Access Routines (NFARS).
If you wi I I be writing DECnet appl ication programs that require
either of these I ibraries, answer this question YES.
These I ibraries may be added at a later time by doing a
Component Mode generation.
If you specified the language libraries in Question 8. 00 and did not choose dry run
mode, NETGEN displays the following statements. The variable, dduu:, is your target
disk.

The DECnet MACRO user I ibrary wi I I be placed on your target disk
as "dduu:[I, I]NETLIB.MLB".
The DECnet FORTRAN/COBOL/BASIC+2 object I ibrary wi I I be placed
on your target disk as "dduu:[I,I]NETFOR.OLB".
If necessary, please move them to the I ibrary disk (LB:) on
your target system, if this is different from your target
disk (dduu:).
>; <EOS> Do you want to:
>*
<RET>-Continue, S-Skip, R-Repeat section, P-Pause,
E-Exit [S]:

NETGEN Dialog:

DEC

6-7

6.2

Defining the OECnet Communications Components

This section defines the DECnet communications components. These components
includeXPT, RCP, ECL, NETACP, DCP, and EPM. Figure 6-2 illustrates the sequence of
questions for defining the DECnet Communications Components.

DEC BUILDS DECnet COMMUNICATIONS
SUPPORT COMPONENTS
TRANSPORT PROCESS
ROUT~G CONTROL PROCESSOR
END COMMUNICATION LAYER AND DRIVER
NETWORK ACP
DIGIT AL COMMUNICATIONS PROCESS
ETHERNET PROTOCOL MANAGER PROCESS

TW0177

Figure 6-2:

6-8

DEC - Section 2 - Define the DECnet Communications Components

DECnet-RSX Network Generation and Installation Guide

>; =============================================================
>; DEC - Section 2 - Define the DECnet Communications Components
>; =============================================================

>* 01.00 Do you want any of the Communications Components? [D=N]

[YIN]:

The DECnet Communications Components are:
XPT - The routing process
RCP - The routing event processing task
ECl - The end communication layer and driver
NETACP - The network services ACP.
DCP - The DDCMP I ine protocol process.
EPM
- The Ethernet protocol manager process.
If you answer YES to this question, you wi I I be asked about each
component individually. In general, you should only answer YES
when there is a clear reason to bui Id or rebui Id one or more of
these components.

NETGEN displays Question 1.00 only in component mode. If you answer NO, you will
immediately jump to the end-of-section (EOS) question with one exception. NETGEN
displays the following statement if you did not select extended network support in a
previous generation and you are now selecting it in component mode.
ECl and NETACP wi I I be rebui It to add the extended network
support for network command terminals and products layered
on DECnet.

The following statement is displayed in question and answer mode. Question 2.00 is
displayed in component mode.
Creating bui Id fi les for XPT, the routing process.
>* 02.00 Do you want XPT? [D=N] [YIN]:

XPT is the rout i ng process. It is respons i b I e for t ransmi tt i ng
and receiving packets fromlto the local node and forwarding
route-through packets.
If you wish to rebui Id XPT, answer this question YES, otherwise
answer NO.
If the node being generated is a routing node, the following statement is displayed
whenever XPT is built or rebuilt. If you have generated a routing node and you are not

rebuilding XPT, Question 3.00 is displayed in component mode.
Creating bui Id fi les for RCP, the routing event processing task.
>* 03.00 Do you want RCP? [D=N]

[YIN]:

RCP IS the routing event processing task. It is responsible for
managing the routing database.
If you wish to bui Id RCP, answer this question YES, otherwise
answer NO.

NETGEN Dialog:

DEC

6-9

The following statement is displayed in question and answer mode. Question 4.00 is
displayed in component mode.
Creating bui Id fi les for ECl, the network services process and
d rive r .
>* 04.00 Do you want ECl? [D=N]

[YIN]:

ECl is the end communication layer and driver. It is
responsible for al I of the time critical network activities,
and for handl ing all user task QIOs. When loaded, it appears
to the CEX system as a process, and to the RSX system as a
device (NS:) which can accept QIO requests.
If you wish to rebui Id ECl, answer this question YES, otherwise
answer NO.

The following statement is displayed in question and answer mode, and in component
mode if you have chosen to rebuild ECl or XPT. Question 5.00 is displayed in component mode only if NETACP is not being automatically rebuilt for you.
Creating bui Id fi les for NETACP, the network services ACP.
>* 05.00 Do you want NETACP?

[D=N] [YIN]:

NETACP is the network services anci I lary control process. It is
responsible for performing al I of the less time critical network
activities. NETACP works in conjunction with the ECl
p rocessld rive r .
If you wish to rebui Id NETACP, answer this question YES,
otherwise answer NO.
>* 05.01 Should NETACP be checkpointable? [D=N]

[YIN]:

Making NETACP checkpointable wi I I al low it to share partition
space with other tasks. Checkpointing wi II increase the amount
of time required for connect and disconnect processing.
If your target system is sma I I (below 80K words) then it is
advisable to Instal I NETACP in the same partition as F11ACP.
In this case, NETACP must be made checkpointable.
In medium to large target systems (above 80K words), NETACP may
be Instal led either alone in its own partition or in a system
control led partition such as GEN. In either case, NETACP should
be made non-checkpointable.

NETGEN displays Question 5.01 only for RSX-IIM/-IIM-PlUS syste~s.
If you are in question and answer mode and you have specified devices that require soft-

ware DDCMP support, NETGEN displays the following statement. Question 6.00 is displayed in component mode.
Creating bui Id fi les for DCP, the DDCMP' I ine protocol process.
>* 06.00 Do you want DCP? [D=N]

6-10

[YIN]:

DECnet-RSX Network Generation and Installation Guide

DCP is the DDCMP I ine protocol process. It is responsible for
providing an error-free data path over a physical device
channe I.
If you wish to generate a new version of the DCP process, then
answer this question YES, otherwise answer NO.
If you are in question and answer mode and you have specified devices that require
EPM, NETG EN displays the following statement. If you have specified devices that

require EPM, Question 7.00 is displayed in component mode.
Creating bui Id fi les for EPM, the Ethernet protocol manager process.
>* 07.00 Do you want EPM? [D-N] [YIN]:
EPM is the Ethernet protocol manager process. It is
responsible for control I ing access to the Ethernet.
If you wish to generate a new version of the EPM process, then
answer this question YES, otherwise answer NO.
>; <EOS> Do you want to:
>*
<RET>-Continue, S-Skip, R-Repeat section, P-Pause,
E-Exit [S]:

NETGEN Dialog:

DEC

6-11

6.3

Defining the DECnet Network Management Components

This section defines components that perform network management functions. These
components include NICE, EVR, NTD, NTDEMO, LIN, MIR, and NVP. Figure 6-3 illustrates the sequence of questions for defining the DECnet Network Management Components.

6-12

DECnet-RSX Network Generation and Installation Guide

TW0178

Figure 6-3:

DEC - Section 3 - Define the DECnet Network Management
Components

NETGEN Dialog:

DEC

6-13

>
>
>

=============================================================
DEC Section 3-Define the DECnet Network Management Components
=============================================================

>* 01.00 Do you want any of the Network Management Components?

[D=N] [YIN]:
The DECnet Network Management Components are:
NICE
EVR
NTD
NTDEMO
LIN
MIR
NVP

- The network information and control task
- The network event receiver task
- The node state display task
- The node state display server task
- The I ink watcher task
- The loop test mirror task
- The network connect verification task

If you answer YES to this question, you wi I I be asked about each
component individually. In general, you should only answer YES
when there is a clear reason to bui Id or rebui Id one or more of
these components.
If you are in component mode, NETGEN displays Question 1.00. If you answer NO,
you immediately jump to the end-of-section (EOS) question. If you are in question and
answer mode and you answered NO to Questions 9.00 and 10.00 (the standard system
option and all components option) in NET Section 1, NETGEN conditionally displays
the remaining questions in Section 3.
>* 02.00 Do you want NICE? [D=N] [YIN]:

NICE is the network information and control server task. It
al lows a system manager to perform network management functions
from any other node on the network.
If you wish to bui Id NICE, answer this question YES, otherwise
answer NO.
If you answer NO, you may add NICE later by performing a
component mode generation.

Question 3.00 is always displayed for RSX-11S systems. It is only displayed for RSX11M/11M-PLUS systems if you answered NO to Questions 9.00 and 10.00 (the standard function option and all components option) in NET Section 1.
>* 03.00 Do you want EVR? [D=N] [YIN]:
EVR is the network event receiver task. It is responsible for
receiving event notification messages from other nodes, and for
logging those messages appropriately.
If you wish to bui Id EVR, answer this question YES, otherwise
answer NO.
If you answer NO, you may add EVR later ,by performing a
Component mode generation.

6-14

DECnet-RSX Network Generation and Installation Guide

Question 3.01 is only displayed for RSX-llS systems and only if Question 3.00 was
answered YES. For RSX-IIM/M-PLUS systems, monitor and file event logging are
always included.
>* 03.01 Include monitor event logging? [D=N] [YIN]:

Monitor logging al lows EVR to hand event data to a user
written task for later processing. Answer this question YES
if you wish to include support for this option, and
NO otherwise.
>* 04.00 Do you want NTD? [D=N] [YIN]:

NTD is the node state display task. It al lows the user to
monitor the current state of a node on a VT100 series, VT200
series, or VT52 video terminal. It uses the NTDEMO server task
to obtain information about the node.
If you wish to bui Id NTD, answer this question YES, otherwise
answer NO.
If you answer NO, you may add NTD later by performing a
Component mode generation.
If you specified NTD in Question 4.00, NETGEN displays the following statement in
question and answer mode. If you did not specify NTD in Question 4.00, Question

5.00 is displayed in question and answer mode. Question 5.00 is always displayed in
component mode.
Creating bui Id fi les for NTDEMO, the node state display server
task.
>* 05.00 Do you want NTDEMO? [D=N] [YIN]:

NTDEMO is the node state display server task. It provides
information to NTD, the node display task, about the node on
which it resides.
If you wish to bui Id NTDEMO, answer this question YES, otherwise
answer NO.
If you answer NO, NTDEMO may be added later by performing a
Component mode generation.

NETGEN displays the following statement for RSX-IIM-PLUS systems in question and
answer mode. Question 6.00 is displayed in component mode for RSX-IIM/-IIMPLUS systems. It is also displayed in question and answer mode for RSX-IIM systems.
Creating bui Id fi les for LIN, the I ink watcher task.
>* 06.00 Do you want LIN? [D=N] [YIN]:

LIN is the link watcher task. It is respons I b I e for determi n i ng
what is to be done with an automatic service request. The
fol lowing types of request are recognized when received from
another node on the network.

NETGEN Dialog:

DEC

6-15

Down-line system load requests from adjacent nodes.
Up-I ine system dump requests from adjacent nodes.
Loop CI rcuit test requests from adjacent nodes.
If you wish to bui Id LIN, answer this question YES, otherwise
answer NO.
If you answer NO, you may add LIN later by performing a
Component Mode generation (to do this, you must select DLX
during this generation).
Note that if you answer YES to this question, the DLX process
wi I I automatically be included In your target system.
If you are in question and answer mode and you answer Question 6.00 NO, NETGEN

displays the following statement.
Note that to add LIN later, during a Component mode generation,
the DLX process must be selected in this generation.

NETGEN displays the following statement in question and answer mode. Question
7.00 is displayed in component mode.
Creating bUild files for MIR, the loop test mirror task.
>* 07.00 Do you want MIR? [D=N] [YIN]:
MIR IS the loop test mi rror task. It accepts loop node messages
from the LOO task (or the equivalent), and reflects the messages
back, providing a logical link loopback test.
If you wish to bui Id MIR, answer this question YES, otherwise
answer NO.
If you are in question and answer mode and your system is a multiuser system, the fol-

lowing statement is displayed. Question 8.00 is displayed in component mode.
Creating bui Id fi les for NVP, the network connect verification
task.
>* 08.00 Do you want NVP? [D=N] [YIN]:
NVP is the network connect verification task. It is responsible
for verifying accounting data provided with a connect request
against the system accounting fl Ie. Verification may be enabled
or disabled on a per object basis.
If you wish to bui Id NVP, answer this question YES, otherwise
answer NO.
>; <EOS> Do you want to:
>*
<RET>-Contlnue, S-SkIP, R-Repeat section, P-Pause,
E-Exlt [S]:

6-16

DECnet-RSX Network Generation and Installation Guide

6.4

Defining the DECnet Satellite Support Components

This section defines the DECnet satellite support components DLL, DUM, CCR, HLD,
and SLD for RSX-llS systems. Figure 6-4 illustrates the sequence of questions for
defining the D ECnet satellite support components.
Before answering the questions in this section of the NETGEN procedure it may be useful to read the chapter in the DECnet-RSX Network Management Concepts and Procedures manual, which describes the support available for remote RSX-llS systems.

TWQ179

Figure 6-4:

DEC - Section 4 - Define the DECnet Satellite Support Components

NETGEN Dialog:

DEC

6-17

>; =============================================================
>; DEC-Section 4-Define the DECnet Satel I ite Support Components
>; =============================================================
>* 01.00 Do you want any of the Satel I ite Support Components?

[D=N] [YIN]:
The DECnet Satel I ite Support Components are:
DLL - The down-I ine system loader, host component
DUM - The up-I ine system dumper, host component
CCR - The Console Carrier Requester, host component
HLD - The down-I ine task loader, host component
SLD - The down-I ine task loader, satel I ite component
If you answer YES to this question, you wi I I be asked about each
component individually. In general, you should only answer YES
when there is a clear reason to bui Id or rebui Id one or more of
these components.
If you are in component mode, NETGEN displays Question 1.00. If you answer NO to

this question, you immediately jump to the end-of-section (EOS) question. The escape
text will display the components your system will support. It will not display all the
components listed in Question 1.00.
If you are in question and answer mode, and you answered YES to Questions 9.00 or

10.00 (the standard function option and all components option) in NET Section 1,
NETGEN displays Question 2.00. The escape text will display the components your system will support. It will not display all the components listed in Question 2.00. If you
are in question and answer mode and you answered NO to Questions 9.00 and 10.00 in
NET Section 1, NETGEN conditionally displays the remaining questions in Section 4.
>* 02.00 Do you want the Satel I ite Support Components? [D=N]

[YIN]:

The DECnet Satel I ite Support Components are:
DLL - The down-I ine system loader, host component
DUM - The up-I ine system dumper, host component
CCR - The console carrier requester, host component
HLD - The down-I ine task loader, host component
SLD - The down-I ine task loader, satel I ite component
If you answer YES to this question, al I of the Satel I ite Support
Components wi I I be generated. If you answer this question NO,
you can add these components by doing a Component Mode
Generation.
If you are in question and answer mode or component mode and LIN was specified,

NETGEN displays Question 3.00 for RSX-IIM and RSX-IIM-PLUS systems. This question is never displayed for RSX-I1S systems.
>* 03.00 Do you want DLL? [D=N]

6-18

[YIN]:

DECnet-RSX Network Generation and Installation Guide

DLL is the down-I ine system loader task. It resides on a host
system, and is responsible for loading an adjacent node with the
proper operating system. The host system must be adjacent to
the system to be loaded, and must be able to access the target
system's image fi Ie.
DLL determines the proper system image fi Ie to load from
information contained in the load request, from the circuit over
which to load the target system, and from a database set up by
the system manager.
DLL may be invoked via the NCP load command, or it may be
invoked automatically by a load request which has been
received over a network circuit. To be invoked automatically,
the LIN task must have been generated. DLL uses the DLX
interface.
If you wish to bui Id DLL, answer this question YES, otherwise
answer NO.
If you answer NO, you may add DLL later by performing a
Component mode generation (to do this, you must select DLX
during this generation).
Note that if you answer YES to this question, the DLX process
wi I I automatically be included in your target system.
If you are in question and answer mode or component mode and LIN was specified,

NETGEN displays Question 4.00 forRSX-llMandRSX-llM-PLUSsystems. Thisquestion is never displayed for RSX-llS systems.
>* 04.00 Do you want DUM? [D=N]

[YIN]:

DUM is the up-I ine system dumper task. It resides on a host
system, and is responsible for dumping memory on an adjacent
node to a fi Ie. The host system must be adjacent to the system
to be dumped, and must be able to access the fi Ie into which the
target system memory is to be dumped.
DUM determines the proper fi Ie to dump to from information
contained in the dump request, from the circuit over which to
dump the target system, and from a database set up by the system
manager.
DUM can only be invoked automatically by a dump request which
has been received over a network ci rcuit. For this reason,
the LIN task must have been generated. To make sure an RSX-llS
system may be up-I ine dumped, the user must have replaced the
RSX-llS PANIC module with the DECnet NETPAN module when the
RSX-llS system was initially SYSGENed. DUM uses the DLX
interface.
If you wish to bui Id DUM, answer this question YES, otherwise
answer NO.
If you answer NO, you may add DUM later by performing a
Component mode generation (to do this, you must select DLX
during this generation).

NETGEN Dialog:

DEC

6-19

Note that if you answer YES to this question, the DLX process
wi I I automatically be included in your target system.

NETGEN displays Question 5.00 only if you have selected Ethernet devices.
>* 05.00 Do you want CCR? [D=N] [YIN]:

CCR is the console carrier requester task. It resides on a host
system, and communicates with the console carrier server (CCS)
which resides on a remote server node. The console carrier
requester in conjunction with the console carrier server provide
access to console functions on a server system that does not
have a console of its own. The console carrier requester
provides the user interface to the console functions that wi I I
be avai lable for the remote system.
If you wish to bui Id CCR, answer this question YES, otherwise
answer NO.
If you answer NO, you may add CCR later by performing a
Component mode generation (to do this, you must select DLX
during this generation).
Note that if you answer YES to this question, the DLX process
wi I I automatically be included in your target system.
If you are in question and answer mode or component mode, NETGEN displays Ques-

tion 6.00 for RSX-IIM and RSX-IIM-PLUS systems. This question is never asked for
RSX-l1S systems.
>* 06.00 Do you want HLD? [D=N] [YIN]:

HLD is tne host down-I ine task loader. It resides on a host
system, and is responsible for sending task images over a
connection to be loaded into the target system's memory. The
host system must have access to the target system's task image
f i I es.
HLD determines the task image fi Ie to load from information
contained in the connect request, from the node which requested
the connection, and from a database set up by the system
manager. HLD may only be invoked automatically by a connect
request.
To down-I ine load task images into an RSX-llS system,
SLD must have been selected for the RSX-llS system during its
NETGEN.
If you wish to bui Id HLD, answer this question YES,
otherwise answer NO.
If you answer NO, you may add HLD later by performing a
Component mode generation.
If you answered Question 6.00 YES, NETGEN displays the following statement.

6-20

DECnet-RSX Network Generation and Installation Guide

To generate the HLD database which describes the tasks to
down-I ine load, you may execute the HLDDAT.CMD command procedure
at any time after your network generation IS complete. This
command fi Ie wi I I be located as fol lows.

dduu:[x,1]HLDDAT.CMD

The HLDDAT. CMD command procedure for generating the HLD database is described
in the DECnet-RSX Network Management Concepts and Procedures manual.
>* 06.01 Number of incoming connections to support? [D R:1.-32. D:4.]:
Enter the number of simultaneous incoming connections that
HLD should accept. This will limit the number of down-line
task loads which can be performed at the same time.
If you are in question and answer mode or component mode, NETGEN displays Question 7.00 for RSX-llS systems built with remote task loading support. Question 7.0 is
never asked for RSX-IIM/M-PLUS systems.

>* 07.00 Do you want SLD? [D=N] [YIN]:
SLD is the satell ite down-line task loader. It resides on
a satel I ite system, and is responsible for connecting to HLD
on a host system and for requesting the task Images for it to
load into memory.
SLD determines the task to load. SLD may only be invoked
automatically by a run, overlay or checkpoint request
Note that certain SYSGEN requi rements must be met before
SLD can be properly added to an RSX-11S system.
If you wish to bui Id SLD, answer this question YES,
otherwise answer NO.
If you answer NO, you may add SLD later by performing a
Component Mode generation.
If your RSX-llS system was generated without remote task loading support, NETGEN

displays the following statement:
SLD cannot be bui It for the target RSX-11S system. The
target system was generated without remote task loading
support.
>; <EOS> Do you want to:
>*
<RET>-Continue, S-Sklp R-Repeat section, P-Pause,
E-Ex i t [S]:

NETGEN Dialog:

DEC

6-21

6.5

Defining the DECnet File Utilities

This section defines components that perform file transfers and remote file operations.
These components include NFT, FTS, FAL and MeM. Figure 6-5 illustrates the
sequence of questions for defining the DEC net File Utilities.

TW0180

Figure 6-5:

6-22

DEC - Section 5 - Define the DECnet File Utilities

DECnet-RSX Network Generation and Installation Guide

>; ============================================================
>; DEC - Section 5 - Define the DECnet Fi Ie Uti I ities
>; ============================================================

>* 01.00 Do you want any of the Fi Ie Uti I ities? [D=N]

[YIN]:

The DECnet Fi Ie Uti I ities are:
NFT - The network fi Ie access user task.
FTS - The network fi Ie access spooler user task.
FAL - The network fi Ie access server.
MCM - The network command fi Ie submission server.
If you answer YES to this question, you wi I I be asked about
each component individually. In general, you should only answer
YES when there is a clear reason to bui Id or rebui Id one or more
of these components.

NETGEN displays Question 1.00 in component mode. If you answer NO, you immediately jump to the end-of-section (EOS) question.
If you are in question and answer mode or component mode, NETGEN displays Ques-

tion 2.00 for RSX-IIM and RSX-IIM-PLUS systems. Question 2.00 is never displayed
for RSX-11 S systems.
>* 02.00 Do you want NFT? [D=N]

[YIN]:

NFT is the network fi Ie access user task. It al lows users to
access fi les on other nodes in the network. The types of fl Ie
operations al lowed include fi Ie transfer, fi Ie deletion,
directory I isting, and commandlbatch fi Ie execution.
If you wish to bui Id NFT, answer this question YES, otherwise
answer NO.
If you answer NO, you may add NFT later by performing a
Component Mode generation.

Question 03.00 is asked only if the Queue Manager was included when the SYSGEN
was performed and the file, LB:[1,24]QMG.OLB, has not been deleted.
>* 03.00 Do you want FTS? [D=N]

[YIN]:

FTS is the network fi Ie access spooler user task. It al lows
users to access fi les on other nodes in the network in a
spooled fashion. That is, fi Ie access operations may execute
immediately, may wait for a specific time, or may wait for the
source or target node to become avai lable on the network. The
types of fi Ie operation al lowed include fi Ie transfer, fi Ie
deletion, and commandlbatch fi Ie execution.
When FTS is selected, the queue management tasks, FTQ and
FTSDEQ, are automatically included.
If you wish to bui Id FTS, answer this question YES,
otherwise answer NO.

NETGEN Dialog:

DEC

6-23

If you answer NO, you may add FTS later by performing a
Component mode generation.
>* 04.00 Do you want FAL? [D=N] [YIN]:

FAL is the network fi Ie access server. It is responsible for
performing al I of the fi Ie operations requested by users
residing on other nodes.
If you wish to bui Id FAL, answer this question YES, otherwise
answer NO.
If you answer NO, you may add FAL later by performing a
Component mode generation.

NETGEN displays Question 4.01 for RSX-11M systems only. For RSX-11M-PLUS systems, amulticopy version ofRMS FAL will be generated. ForRSX-11S systems, the lIS
version of FCS FAL is used.
>* 04.01 Should FAL support RMS fi Ie access? [D=Y] [YIN]:

Answer this question YES if you wish to al low network access
to RMS fi Ie services on your node. Answer this question NO if
you wish to al low only FCS fl Ie access.
In order to include RMS support, you must have previously
placed the proper RMS fi les on your LB: device under the
proper UIC's. If you wish to use the RMS routines (not memory
resident RMS library), these fi les and UIC's are:
LB: [l,l]RMSLIB.OLB
LB: [1,1]RMS11X.ODL
LB: [1,1]RMS12X.ODL
LB: [l,l]VMLIB.OLB
If you wish to use the memory resident RMS library, these
fi les and UIC's are:
LB: [l,l]RMSRLX.ODL
LB: [l,l]RMSRES.TSK
LB: [l,l]RMSRES.STB
LB: [l,l]VMLIB.OLB
If the proper fi les are not present on LB: under the proper
UIC, you may pause and move them now.

NETGEN displays Question 4.02 only for RSX-11M systems (for FCS and RMS versions
ofFAL). RSX-11M-PLUS and RSX-11S versions are never overlaid.
>* 04.02 Should FAL be overlaid? [D=N] [YIN]:

Answer this question YES if FAL should be overlaid. This wi I I
reduce the size of FAL by approximately 4000 words but wi I I
also reduce the performance.
Note that if FAL is going to support RMS fi Ie access, and if
the RMS routines are to be included directly in the FAL task
image (i .e. a memory resident RMS library is not supported),
then the RMS routines wi I I always be overlaid. This question
only controls the overlaying of the FAL mainl ine code.

6-24

DECnet-RSX Network Generation and Installation Guide

NETGEN displays Question 4.03 only for RSX-IIM systems if the RMS memory-resident library is found. For RSX-IIM-PLUS systems, the RMS memory-resident library
will always be used if present. If RMSRES is not found, you will be given the option of
building the FCS FAL.
>* 4.03 00 you want to use the memory resident RMS library? [O=N] [YIN]:

A memory resident RMS I ibrary has been found on LB: [1,1].
Answer this question YES if you wish FAL to I ink to this
library.
Using a memory resident RMS I ibrary wi I I result in better
performance and wi I I reduce task memory requirements. Using
this option wi I I, however, require that the RMSRES I ibrary be
permanently instal led and loaded on your system (this I ibrary is
shareable between al I tasks which use RMS, so the extra memory
requi red by the I ibrary matters only if there is only one RMS
task) .

After you do a component mode generation, you may need to update the NETINS.CMD
file. For example, if you specified memory-resident RMS in a previous generation,
NETINS.CMD will automatically check to see if the memory-resident RMS library was
installed. If you are generating a version of FAL that does not require the memory-resident RMS library, you should remove this automatic checking process by editing the
NETINS.CMD file.
NETGEN displays Question 4.04 for RMS FAL on RSX-IIM systems only if you
answered NO to Question 4.03.
>* 04.04 00 you want the minimum size RMS? [O=N]

[YIN]:

Answer this question YES if you wish the RMS portion FAL
to be as smal I as possible, and NO otherwise.
Requesting the minimum size RMS wi I I reduce FAL'S memory
requi rements, but wi I I also result in reduced performance.
Use of the minimum size RMS wi I I save 3K bytes of FAL task
space.
If you answer this question YES, the RMS portion of FAL wi I I
be bui It using the RMS11X.OOL overlay description fi Ie. If
you answer this question NO, the RMS portion of FAL wi I I be
bui It using the RMS12X.OOL overlay description fi Ie. These
fi les are distributed with your RSX kit.

NETGEN displays Question 4.05 for RSX-IIM and RSX-llS systems.
>* 04.05 Number of incoming connections to support?

[0 R: 1.-10. 0:4.]:
Enter the number of simultaneous incoming connections that
FAL shou I d accept. Th is wi I I I imi t the number of network f i Ie
access requests which can be serviced at the same time.

NETGEN Dialog:

DEC

6-25

Question 4.06 is displayed for all systems.
>* 04.06 User data buffer size? [0 260. :2048. 0:1024.]:

Specify the size of the buffer to be used for user data
transfers. The buffer must be large enough to hold the
largest record that FAL wi I I ever have to transfer. For
task image fi les, a buffer of 512 bytes is sufficient.
Note that the data buffer size affects the efficiency of
the data transfers and the size of FAL's dynamic memory
pool.
If your system does not have multiuser protection, NETGEN displays Questions 4.07

through 4.09 for RSX-IIM and RSX-IIM-PLUS systems. For RSX-llS systems, only
Question 4.07 is displayed, and only the first line of the explanatory text fs displayed.
>* 04.07 What is the privi leged password?
[O=PRIV] [S R:0.-8.]:
Enter the password to use to check for privi leged access
to f i I es .
This password is requi red since the target system does not
have multi-user protection but OECnet must provide access
protection.
Use of this password with any UIC wi I I al low network fi Ie
access to any fi les which could be accessed by a local user
logged onto that UIC.
>* 04.08 What is the non-privi leged password?

[O=GUEST] [S R:0.-8.]:
Enter the password to use to check for non-privi leged
access to fi les. The UIC associated with this password is
defined in the next question.
This password is requi red since the target system does not
have multi-user protection but OECnet must provide access
protection.
Use of this UIC and password wi I I al low network fi Ie access
to any fi les which could be accessed by a local user logged
onto the UIC.
>* 04.09 What is the guest UIC? [0=200,200] [S R:0.-9.]:

Enter the guest UIC for network fi Ie access. The format for
entering a UIC is either "[grp,mem]" or "grp,mem".
This UIC is associated with the password defined in the
previous question.
Use of this UIC and password wi I I al low network fi Ie
access to any fi les which could be accessed by a local
user logged onto the UIC.

6-26

DECnet-RSX Network Generation and Installation Guide

If you are in question and answer mode or component mode and you have specified

NFT or FAL, NETGEN displays Question 5.00 for RSX-IIM and RSX-IIM-PLUS systems. Question 5.00 is always asked in component mode and is never asked for RSX11 S systems.
>* 05.00 Do you want MCM? [D=N] [YIN]:
MCM is the network command or batch fi Ie submission server.
It receives fi Ie submission requests from NFT or FAL, and
submits the indicated command or batch fi les for execution
on the system on which it resides.
Note that MCM uses the RSX-ll send data mechanism, rather
than DECnet connections, to receive requests: For this
reason either NFT or FAL or both must have been selected
for the target system.
If you wish to bui Id MCM, answer this question YES, otherwise
answer NO.
If you answer NO, you may add MCM later by performing a
Component Mode generation.

Question 5.01 is displayed forRSX-IIM-PLUS systems only.
>* 05.01 Should requests be queued to Batch? [D=N] [YIN]:
If you answer this question YES, NFT Submit (/SB) and Execute
(lEX) requests wi I I be entered into the Batch queue for
execution.
If you answer NO, the requests wi I I be sent to the Indirect
Command Fi Ie processor (AT.).
>; <EOS> Do you want to:
>*
<RET>-Continue, S-Skip R-Repeat section, P-Pause,
E-Exit [S]:

NETGEN Dialog:

DEC

6-27

6.6

Defining the DECnet Terminal and Control Utilities

This section defines the DECnet terminal and control utilities - RMT (with RMT ACP),
HT: (with RMHACP), NCT, RTH, TLK, LSN, PHO, LAT, and TCL. These tasks allow
you to establish a remote terminal, establish a network command terminal, carryon a
dialog between nodes, and control a task on a remote node. Figure 6-6 illustrates the
sequence of questions for defining the DECnet terminal and control utilities.

TW0181

Figure 6-6:

6-28

DEC - Section 6 - Define the DECnet Terminal and Control Utilities

DECnet-RSX Network Generation and Installation Guide

>; ============================================================
>; DEC - Section 6 - Define the DECnet Terminal

and Control Uti I ities
>; ============================================================

>* 01.00 Do you want any of the Terminal and Control Util ities?

[D=N] [Y IN] :
The DECnet Terminal and Control Uti I ities are:
RMT/RMTACP
HT:/RMHACP
NCT
RTH
TLK
LSN
PHO
LAT
TCL

- The remote network terminal task and
ACP.
- The remote network terminal driver and
ACP.
- The network command terminal server.
- The remote command terminal host
support task.
- The remote talk user task.
- The remote talk server task.
- The inter-terminal conversational
task.
-. The Local Area Transport terminal
services process.
- The remote task control server task.

NETGEN displays Question 1.00 in component mode. Uyou answer NO, you immediately jump to the end-of-section (EOS) question. The escape text associated with question 01.00 will display the components your system will support. It may not display all
the components shown in Question 1.00.
>* 02.00 Do you want RMT/RMTACP? [D=N] [YIN]:

RMT and RMTACP are the remote network terminal task and ACP.
They reside on the user's local system, and al low the user's
terminal to be logically connected to any other RSX node in
the network which supports the HT: device and its ACP, RMHACP.
If you wish to bui Id RMT and RMTACP, answer this question YES,
otherwise answer NO.
If you answer NO, you may add RMT and RMTACP later by performing
a Component Mode generation.
>* 02.01 Maximum number of simultaneous RMT users? [0 R:l.-15. 0:4.]:

Enter the maximum number of simultaneous RMT users that
you wish your system to support.
>* 03.00 Do you want HT:/RMHACP? [D=N] [YIN]:

HT: and RMHACP are the remote network terminal driver
and ACP. They reside on a host system, and al low users
on other nodes in the network to logically connect their
terminals to the host system, through the host's
HT: terminal devices.

NETGEN Dialog:

DEC

6-29

If you wish to bui Id HT: and RMHACP, answer this question
YES, otherwise answer NO.
If you answer NO, you may add HT: and RMHACP later by
performing a Component Mode generation.
>* 03.01 Number of incoming connections to support? [D R:l.-16. D:4.]:

Enter the number of simultaneous incoming connections that
RMHACP should accept. This wi II I imit the number of network
terminal users that can log onto your system at the same time.

NETGEN displays Question 3.02 for DECnet systems with PSI. The variable n is the
value you specified in Question 3.0 I.
>* 03.02 Number of HT: units to generate? [D R:n.-64. D:n.]:

Enter the number of HT: device units to generate. Each
incoming RMHACP connection requires one HT: unit.
Note that if you plan to include the X29ACP remote terminal PSI
component, you must specify enough HT: units for both RMHACP
and X29ACP. These units wi I I be dynamically al located on an
as-needed basis by these two components, so you need only
al locate enough HT: units to satisfy their concurrent
requirements.

NETGEN displays Questions 4.00 and 5.00 if you selected extended network support
by answering YES to Question 6.00 in DEC Section I.
>* 04.00 Do you want NCT? [D=N]

[YIN]:

NCT is the network command terminal server. It resides on
the user's local node, and al lows the user's terminal to
be logically connected to any other node in the network with
the network command terminal faci I ity.
If you wish to bui Id NCT answer this question YES, otherwise
answer NO.
If you answer NO, you may add NCT later by performing a
Component Mode generation.

Question 05.00 is displayed for RSX-IIM systems only if Network Command Terminal
support was specified for inclusion during SYSGEN. (See Section 3.2.5.4.). RSX-IIMPLUS always includes support for Network Command Terminals. Question 05.00 is,
therefore, always displayed if you are generating an RSX-IIM-PLUS node.
>* 05.00 Do you want RTH? [D=N]

[YIN]:

RTH is the remote command terminal host support process. It
resides on the user's local node and al lows users on other nodes
to logically connect their terminals to the local system through
the network command terminal faci I ity.
If vou wish to bui!d RTH answer th!s question YES, otherwise
answer it NO.

6-30

DECnet-RSX Network Generation and Installation Guide

If you answer NO, you may add RTH later by performing a
Component Mode generation.
>* 06.00 Do you want TLK? [O=N]

[YIN]:

TLK is the remote talk user task. It resides on the user's
a user on an RSX node which supports the LSN task.
If you wish to bui Id TLK, answer this question YES, otherwise
answer NO.
If you answer NO, you may add TLK later by performing a
Component Mode generation.
If you are in question and answer mode and you specified TLK, NETGEN displays the
following statement. If you did not specify TLK, NETGEN displays Question 7.00.

Question 7.00 is always displayed in component mode.
Creating bui Id fi les for LSN, the remote talk server task.
>* 07.00 Do you want LSN? [O=N]

[YIN]:

LSN is the remote talk server task. It resides on a target
system, and al lows a user on an RSX node which supports the
TLK task to hold a "conversation" with a user on the system
which supports the LSN task.
If you wish to bui Id LSN, answer this question YES, otherwise
answer NO.
If you answer NO, you may add LSN later by performing a
Component Mode generation.
>* 07.01 Number of

incoming connections to support? [0 R:1.-20. 0:4.]:

Enter the number of simultaneous incoming connections that
LSN should accept.

Question 8.00 is asked only for RSX-IIM-PLUS systems because PHONE uses accounting information that is not available on RSX-IIM and RSX-llS systems.
>* 08.00 Do you want PHO? [O=N]

[YIN]:

PHO is the video terminal PHONE task. It resides on the
user's local system, and al lows a user to hold a "conversation"
with a user on any node which supports the PHO task.
If you wish to bui Id PHO, answer this question YES, otherwise
answer NO.

Question 9.00 is asked only for RSX-IIM-PLUS systems for which LAT terminal support was specified during SYSGEN.
>* 09.00 Do you want LAT and LCP [D=N]

NETGEN Dialog:

DEC

[YIN]:

6-31

The local Area Transport (lAT) terminal service process
resides on the user's local system and al lows terminals
on a lAT terminal server to connect to the local system as
if they were directly connected to the system. lAT terminal
s e r vic e d i ff e r s from 0 the r r emo t e term ina I f a c iii tie s be c au s e
it only communicates with terminal servers on the Ethernet
and uses protocols that are optimized for Ethernet.
lCP is the lAT Control Program. It al lows the system manager
to perform all on-I ine lAT management functions. (eg. starting
or stopping the lAT process).
If you wish to bui Id the lAT termInal service process and lCP,
answer this question YES, otherwise answer NO. If you answer
this question NO, you may add lAT/lCP later by performing a
Component Mode generatIon.
>* 10.00 Do you want TCl? [D=N] [YIN]:
TCl is the remote task control server task. It allows users
on other nodes to control task execution on the node on which
TCl is instal led. TCl wi I I accept requests over DECnet
connections to run, abort and cancel tasks on its local node.
Routines to connect to TCl and issue these requests are
contained in the DECnet FORTRAN/COBOl/BASIC2 object library.
If you wish to bui Id TCl, answer this question YES, otherwise
answer NO.
If you answer NO, you may add TCl later by performing a
Component Mode generatIon.

If you selected TeL and your system does not have multiuser protection, NETGEN displays Questions 10.01, 10.02, and 10.03.
>* 010.01 What is the p r i v i I eged passwo rd? [<RET>=None]
[ S R: 0 . -8 . ] :

Enter the password to use to check for privi leged access.
Privi leged access provides the abll ity to run tasks under
any UIC, and to cancel or abort any task.
This password is requl red since the target system does not
have multi-user protection but DECnet must provide access
protection.
>* 010.02 What is the non-privIleged password? [<RET>=None]
[S R:0.-8.]:
Enter the password to use to check for non-privi leged
access. Non-privi leged access provides the abi I ity to
run tasks under the non-privi leged UIC.
>* 010.03 What is the non-privi leged UIC?
[S R:0.-9.]:

6-32

[<RET>=No~e]

DECnet-RSX Network Generation and Installation Guide

Enter the UIC under which non-privi leged users may run
tasks. If no UIC is specified, then nonprivileged users
can run tasks under any UIC. Non-privi leged users can
abort or cancel only those tasks which they have
initiated.
The format for entering a UIC is either "[grp,mem]" or
"grp ,mem".
>; <EOS> Do you want to:
>*
<RET>-Continue, S-Skip R-Repeat section, P-Pause,
E-Exit [S]:
>;

============================================================

>; DEC - DECnet CEX Product Generation Procedure
>;
DECnet question/answer section completed at
>;
time on date

If you are doing a component mode generation and you are adding extended network

support, NCT, RTH, or LAT ILCP, NETG EN displays the following statement. This statement may vary based on what you added.
CETAB.MAC is being rebui It to reflect the fol lowing new options:
- the addition of the extended network support for
terminals and layered products.
- the addition of NCT (Network Command Terminal server)
- the addition of RTH (Remote Command Terminal host
support)
- the addition of LAT (Local Area Transport terminal
service process and LAT Control Program).
Adding NCT, RTH, or LAT/LCP wi I I change the data structure
used by your network. If you are currently running the network
you are modifying, it is recommended that you reload the
network as soon as this generation is done.
NOTE
If you are generating the PSI product, NETGEN proceeds with the PSI
generation (Chapter 7). If you are not generating the PSI product,

NETGEN proceeds with the remainder of the NET generation procedure, NET Section 8 - Complete the CEX System (Section 5.9).

NETGEN Dialog:

DEC

6-33

7
NETGEN Dialog: PSI

Chapter 7 presents the third part of the NETGEN dialog: PSI (the Packetnet System
Interface). NETGEN displays a series of questions relating to PSI components if you
have purchased a license for the RSX-ll PSI product and you answered YES to Question 07.01 in Section 1 of the NET portion of NETGEN. These questions immediately
follow DEC Section 6.
If you are not familiar with the Packet Switching Data Network (PSDN) terminology,
refer to the Introduction to RSX-11 PSI, the RSX-11 PSI User's Guide, and the PSI Reference Card.

In PSDN terminology, data terminal equipment (DTE) refers to your computer equipment that connects to the PSDN, while data communications equipment (DCE) is the
PSDN's equipment. Your DTE (computer or X.29 terminal) connects to a DCE on the
PSDN.

7.1

Specifying Target Network Characteristics

This section defines the characteristics for the Packet Switching Data Network (PSDN)
to which you will be connecting. Figure 7-1 illustrates the sequence of questions for
specifying the target network characteristics.

7-1

SPECIFY
TRANSPORT
SUBADDRESS
RANGE

TW0182

Figure 7-1:

7-2

PSI- Section 1 - Specify Target Network Characteristics

DECnet-RSX Network Generation and Installation Guide

>; ============================================================
>; PSI - PSI CEX Product Generation Procedure
>.,
S tar tin g que s t ion sat time 0 n date
>; ============================================================
>;

>; ============================================================
>; >; PSI - Section 1 - Specify Target Network Characteristics
>; ============================================================
>* 01.00 Packet Switching Data Network name? [S R:1. -8.]:
Enter the name of the Packet Switching Data Network (PSDN) to
which you want to connect your DTEs. Refer to the PSI Reference
Card for detai Is of currently supported network names. Only
specify alphanumeric characters in the network name, ignoring
any other characters such as hyphens. The network name wi I I be
truncated to the first six characters.
If your PSDN netnam is not yet supported by RSX-ll PSI, NETGEN displays the follow-

ing statement.
RSX-11 PSI does not yet support the network nemam. NETGEN
wi I I configure RSX-11 PSI for the default network. When NETGEN
is complete, please contact your Software Special ist who wi I I
reconfigure RSX-11 PSI for netnam. This wi II not requi re
another NETGEN.

Support for X.29 is always included in a standard function network. If, therefore, you
answered YES to Question 09.00 in Section 1 of the NET portion of NETG EN, Question
02.00, following, is not asked.
>* 02.00 Do you want X.29 support? [D=N] [YIN]:
If you need support for remote X.29 terminals, answer YES to
this question.
>* 03.00 What is the transport subaddress range? [S R:O -9.]:
The transport subaddress range is used to identify which cal Is
are di rected to DECnet and which cal Is are directed to PSI. Any
cal Is within the range you specify wi I I be passed to DECnet via

DLM.

Specify the subaddress range using one of the fol lowing formats:
LLLL - HHHH where LLLL is the low end of the range and
HHHH the high end. LLLL and HHHH are in
the range 0 - 9999.
LLLL

where HHHH defaults to LLLL.

<RET>

indicates there is no subaddress range.

Your response to Question 3.00 determines which incoming calls are routed to D LM. If
you specify 0 in this range, all incoming calls are sent to DLM.

NETGEN Dialog:

PSI

7-3

However, X.29 uses subaddress o. If you have X.29 support make sure you do not specify 0 for DLM. Doing so will cause incoming X.29 calls to go to DLM.
>; <EOS> Do you want to:
>*
<RET> - Continue, R - Repeat section, P - Pause,
E - Exit.

7-4

DECnet-RSX Network Generation and Installation Guide

7.2

Specifying Local DTEs

This section defines the characteristics of your local DTEs. Figure 7-2 illustrates the
sequence of questions for specifying local DTEs.

SPECIFY
CHANNEL NUMBERS
TO BE USED AS
OUTGOING CIRCUITS

SPECIFY
PERMANENT
VIRTUAL CIRCUIT
(PVC)

TW0183

Figure 7-2:

PSI - Section 2 - Specify Local DTEs

NETGEN Dialog:

PSI

7-5

============================================================
>; ============================================================
>;

>; PSI - Section 2 - Specify Local OTEs

>* 01.00 What is the local OTE address? [S R:O -15.]:
The local OTE is connected to the PSON, and gives you access to
the network. The OTE address is the number assigned to the
local DTE by the network vendor. See the PSI Reference Card
for the format of the OTE address for this network.
Enter the address of the local OTE in the required format, (up
to 15 digits). If you have already specified al I OTEs, answer
<RET>.
>* 02.00 Which I ine is associated with this OrE? [S R:l. -9.]:
Identify the I ine associated with this OTE. Use a I ine-id in
the form ddd-cc[-uu], where:
ddd identifies the device.
cc identifies the controller.
uu optionally identifies the unit.
of this line.
The fol lowing I ines have been generated:
(list of PSI lines)

When NETGEN displays the list of PSI lines, specify one of these lines. This list was created as a result of the devices you specified in NET Section 3 Question 1.
>* 03.00 How many vi rtual ci rcuits can be active at once? [0
R:l. -255.]:
Specify the maximum number of vi rtual ci rcuits that can be
active simultaneously on this OTE. The number you specify is
equivalent to the total number of cal Is, (incoming and
outgoing), plus PVCs, that can be handled at anyone time.
Thus the number you specify should include X.29 circuits.
If you are defining more than one OTE, then the total number of
active ci rcuits in the system must not exceed 255.
>* 04.00 What channels can be used for outgoing cal Is? [S]:
The logical channel numbers you specify here wi I I be used for
outgoing and maybe incoming cal Is for this OTE. Specify a list
of channel numbers, separating the numbers by hyphens to
indicate ranges, and commas to indicate individual numbers.
For example:
20-10,8,3

defines channel numbers 20 through 10, 8 and 3. Channel numbers
are in the range 1 to 4095, and are assigned by the network
vendor.

7-6

DECnet-RSX Network Generation and Installation Guide

Note that channels wi I I be used in the same order as they are
spec i f i ed. You must spec i fy the high end of a channe I range
first, fol lowed by the low end, otherwise the PSI software wi I I
not operate correctly.
If there are no channel numbers to be specified, answer with
<RET>.

NETGEN displays the following statement before Question 5.00.
Specify permanent virtual circuits (PVCs) for this OTE.
Note that any OLM PVCs must also be included.
The fol lowing OLM circuits have been generated:
OLM-x.y S
OLM-x.y P

In the above list of DLM circuits, the circuit ID will be followed by an S or a P. An S
refers to switched virtual circuits (SVCs) and a P refers to permanent virtual circuits
(PVCs).
>* 05.00 Enter a PVC used by this OTE. [S]
Each PVC that you have been al located by the network vendor is
identified by a logical channel number in the range 1 to 4095.
You can give each PVC a name, when you associate the PVC with
you r I oca I OTE.
Enter a PVC used by this OTE in the fol lowing format:
name,channel
where
name

is the name you have chosen for the PVC,
(up to 6 characters long).

channel

is the logical channel number suppl ied by
the network vendor.

Answer with <RET> if you have specified al I PVCs.
For OLM PVC names, enter the OLM ci rcuit names.
The fol lowing OLM circuits have been generated:
(list of DIM circuits)

The same PVC name should not be used to identify more than one PVC.
>* 06.00 Enter a closed user group for this OTE. [S]
A closed user group is an optional faci I ity which al lows a OTE
belonging to the group to communicate with other members of the
same group, but not with any other DTEs. A bi lateral closed
user group is a special case of this, where the group consists
of only two OTEs.

NETGEN Dialog:

PSI

7-7

A DTE may belong to one or more closed user groups. Each closed
user group is identified by a name given it by the members of
the group, and a group number given it by the network vendor.
Specify a closed user group for this DTE in the fol lowing
format:
name,number[,BILATERAL]
where
name

is the name of this group (up to
6 characters).

number

is the number al located to this group
by the network vendor.

BILATERAL

means that the group is bi lateral.

If you have specified al I closed user groups for this DTE,
answer with <RET>.
>; <EOS> Do you want to:
>*
<RET> - Continue, R - Repeat section, P - Pause,
E - Exit.

7-8

DECnet-RSX Network Generation and Installation Guide

7.3

Specifying Incoming Call Destinations

This section defines incoming call destinations. A destination contains a set of parameter values describing the calls the destination can accept. Each destination is associated
with a task you have written to process a particular type of call.
When an incoming call arrives, the information it carries is compared with the parameter values for each destination. The call is processed by the task associated with the destination whose values match those of the call. If more than one destination can accept
the call, it goes to the destination with the highest priority.
Figure 7-3 illustrates the sequence of questions for specifying incoming call destinations.

NETGEN Dialog:

PSI

7-9

TW01S4

Figure 7-3:

7-10

PSI- Section 3 - Specify Incoming Call Destinations

DECnet-RSX Network Generation and Installation Guide

=============================================================
>; =============================================================
>;

>; PSI - Section 3 - Specify Incoming Cal I Destinations
Specify the X.25 destinations

NETGEN displays Questions 1.00 through 7.00 for each X.25 destination you want to
specify. To indicate you are done, press < RET> . NETGEN then repeats the same questions for you to specify X.29 destinations instead ofX.25.
> * 01. 00 Ide n t i f Y ani n c om i ng c a I Ide s tin a t ion. [ S R: 0 -6.] :
Each incoming cal I destination is identified by a destination
name. You specify the parameters in the destination and these
are used to decide if the incoming cal I wi I I be accepted by the
destination. One of the parameters in the destination also
names the task that the cal I wi I I be passed to, for processing.
You can specify a destination name up to 6 characters long.
If you have specified al I destinations, answer with <RET>.
If you specified X.29 support, NETGEN automatically defines a destination so X.29

calls use MeR as the command language interpreter. You do not need to specify such a
destination for Question 1.00.
>* 02.00 What is the priority? [0 R:O -255. 0:127]:
AI I destinations are placed on the destination I ist In order of
priority. When an incoming call arrives, the destination list
is searched for a suitable destination. The call IS sent to the
first (i.e. highest priority) destination that can accept it.
You can specify a priority in the range 0 to 255, where 0 is the
lowes t p rio r i t Y.
>* 03.00 What is the object-id? [S R:1. -6.]:
The object-id identifies the task which wi I I process the cal Is
accepted by this destination. You can specify a task name, (up
to 6 characters), or, if the object is a multicopy task, you
must specify the object's number, (in the range 1 to 255).
A multicopy task can process as many cal Is at one time as there
are copies of the task. If you are specifying a multicopy task
here, you must select multicopy task support when you bui Id the
X25ACP component.

To indicate a multicopy task, you must specify an object number instead of an object
name in Question 3.00.

NETGEN Dialog:

PSI

7-11

NETGEN displays Question 3.01 if you specified a multicopy task in Question 3.00.
>* 03.01 What is the task name for this object? [S]:
Specify the name of the multicopy task for this object. The
name must be in the form:
AAA$$$
where
AAA denotes one to three alphanumeric characters.
>* 04.00 What is the remote DTE address? [S R:O -15.]:
If you specify a remote DTE address, only incoming cal Is from
the specified remote DTE wi I I be accepted by this destination.
Specify the address assigned to the remote DTE in the format
requi red for this network, (see the PSI Reference Cards). Up to
15 digits are al lowed. An asterisk in any position wi I I match
any digit in the corresponding position in the incoming cal I
remote DTE address.
If the destination wi I I accept cal Is from any remote DTE,
answer with <RET>.
>* 05.00 What is the subaddress range? [S R:O -9.]:
If you specify a subaddress range, only incoming cal Is whose
subaddresses are within the specified range wi I I be accepted
by this destination.
Specify the subaddress range using one of the fol lowing formats:
LLLL - HHHH

where LLLL is the low end of the range
and HHH is the high end. LLLL and HHHH
are in the range 0 to 9999.

LLLL

where HHHH defaults to LLLL.

<RET>

where there is no subaddress range and
thus the destination accepts cal Is with
any subaddress.

>* 06.00 What is the closed user group name? [S R:O -6.]:
Specify the name of the closed user group whose cal Is wi I I be
accepted by this destination. If the destination wi I I accept
cal Is from any user group, answer with <RET>.
>* 07.00 What is the cal I mask? [S R:1. -32.]:
If you specify a cal I mask for this destination, the user data
value in the incoming call is subjected to a logical-and
operation. The value resulting from this masking operation is
compared with the destination's cal I value. An incoming cal I
wi I I only be accepted by the destination if its masked cal I
vaiue IS equai to the cai I vaiue of tne destination.

7-12

DECnet-RSX Network Generation and Installation Guide

If the destination's cal I mask is longer than the user data in
the incoming cal I, the cal I wi I I only be accepted by this
destination if the remainder of the cal I mask is zeroes.
Note that both cal I mask and cal I value are specified in hex.
If the destination has no cal I mask or cal I value, answer with
<RET>.
>* 07.01 What is the cal I value? [S R:1. -32.]:
Specify the cal I value that wi I I be compared with the masked
incoming cal I value to decide whether this destination wi I I
accept the cal I. The cal I value must be the same length as the
cal I mask you have specified.
>; <EOS> Do you want to:
>*
<RET> - Continue, R - Repeat section, P - Pause,
E - Exit.

NETGEN Dialog:

PSI

7-13

7.4

Specifying Logical Names for Remote OTEs

This section defines the logical names for all remote DTEs accessed by your system.

=============================================================
>; =============================================================
>;

>; PSI - Section 4 - Specify Logical Names for Remote DTEs
>* 01.00 Ident i fy a remote DTE. [S]:

A remote DTE is identified by its address, assigned by the
network vendor. You can also associate a logical name with
each remote DTE address.
Specify a remote DTE using the fol lowing format:
name,address
where
name

is the logical name you have chosen for
the DTE, (up to 6 characters).

address

is the address assigned to the DTE by the
netwo rk vendo r .

If you have specified al I remote DTEs, answer with <RET>.
>; <EOS> Do you want to:
>*
<RET> -Continue, R - Repeat section, P - Pause,
E - Exit.

7-14

DECnet-RSX Network Generation and Installation Guide

7.5

Specifying Required Components

This section defines the required PSI components. NETGEN displays Questions 1.00
through 5.00 only in component mode. If you are building X25ACP, NETGEN displays
Question 6.00. Figure 7-4 illustrates the sequence of questions for specifying required
components.

TW018~

Figure 7-4:

PSI - Section 5 - Specify Required Components

NETGEN Dialog:

PSI

7-15

>; =============================================================
>; PSI - Section 5 - Specify Required Components
>; =============================================================

>* 01.00 Do you want to bui Id the NW process? [D=N]

[YIN]:

The NW process interfaces with the X.25 level 3 protocol,
interfaces with user tasks via the QIO mechanism, val idates user
QIO requests, passes on non-data transfer requests to X25ACP,
and provides for segmentation and recombination of data.
Answer Y if you want to bui Id the NW process.
>* 02.00 Do you want to bu i I d the PL I process? [O=N]

[Y IN] :

The PLI process implements the X.25 level 3 protocol. It
transmits and receives data packets, multiplexes many virtual
circuits on one I ine, detects level 3 protocol errors, and
controls al location of logical channel numbers.
Answer Y if you want to bui Id the PLI process.

If you have specified SDP and/or SDV device driver processes, NETGEN displays Ques-

tion 3.00.
>* 03.00 Do you want to bui Id the LAB process? [O=N]

[YIN]:

The LAB process implements the X.25 level 2 protocol. It
transmits and receives data, starts up and shuts down lines,
detects and records errors, retransmits data if necessary, and
traces frames that cross the level 2 and device driver boundary.
You requi re this process if you have chosen the SOP andlor
SOV device drivers.
Answer Y if you want to bui Id the LAB process.

Question 04.00 is asked only in component mode.
>* 04.00 Do you want to bui Id the X25ACP task? [O=N]

[YIN]:

The X25ACP task supports the NW process by processing functions
that are not critical to system throughput. It connects and
disconnects cal Is, requests tasks for incoming cal Is, and
notifies tasks of network events.
Answer Y if you want to bui Id the X25ACP task.
>* 05.00 Do you want to bui Id the OLM process? [O=N]

[YIN]:

The OLM process al lows OECnet users to use X.25 to communicate
with remote OTEs on a PSON.
Answer Y if you want to bui Id the OLM process.

Multicopy task support is always included in a standard function network. If, therefore, you are in question and answer mode and you answered YES to Question 09.00 in
Section 1 of the NET portion of NETGEN, Question 06.00, following, is not displayed.

7-16

DECnet-RSX Network Generation and Installation Guide

NETGEN, however, will display Question 06.00 if you are building the X25ACP task in
component mode.
>* 06.00 Do you want multicopy task support? [D=N] [YIN]:
Multicopy task support al lows for multiple copies of a server
task connected by object type to be spawned. This feature
permits users to write single-threaded tasks that perform a
network service and have more than one request serviced at a
time. The multicopy attribute can be enabled/disabled on a
"per object" basis with the SET OBJECT command under NCP, CFE
or VNP.
>; <EOS> Do you want to:
>*
<RET> - Continue, R - Repeat section, P - Pause,
E - Exit.

NETGEN Dialog:

PSI

7-17

7.6

Specifying Optional Components

This section defines the optional PSI components that you may want to build. Figure 7- 5
illustrates the sequence of questions for specifying optional components.

VERIFY
KMX MICROCODE
DUMPER ANALYZER
SUPPORT

TW01S6

Figure 7-5:

7-18

PSI- Section 6 - Specify Optional Components

DECnet-RSX Network Generation and Installation Guide

>
>
>

=============================================================
PSI - Section 6 - Specify Optional Components
=============================================================

>* 01.00 Do you want to include the PSIFOR.OlB library? [D=N]

[YIN]:

If you want to write network tasks in FORTRAN, you must include
the PSIFOR.OlB I ibrary in your target system, so answer Y to
this question.
>* 02.00 Do you want to include the PSI.MlB library? [D=N]

[YIN]:

If you want to write network tasks in MACRO-II, you must include
the PSI.MlB I ibrary in your target system, so answer Y to this
question.

If you chose X.29 support in PSI Section 1 Question 2, NETGEN displays Question
3.00.
>* 03.00 Do you want to bui Id the X29ACP task? [D=N]

[YIN]:

The X29ACP task supports the X.29 protocol. It handles incoming
X.29 cal Is, and provides an interface with the ClI for
unsol iClted Input data.
If you need support for remote X.29 terminals, answer Y to this
question.
>* 04.00 Do you want to bui Id the TRA task? [D=N]

[YIN]:

The TRA task traces frames over a I ine and records the traced
frames on a disk. This helps you to find out what types of
packets are being transmitted and received on the I ine and thus
to monitor its performance.
Answer Y if you want to bui Id the TRA task.
>* 05.00 Do you want to bui Id the TRI task? [D=N]

[YIN]:

The TRI task interprets the data produced by the TRA task so
that you can p r i nt the traced frames. I f you want to wr i t e you r
own trace interpreter task you need not include TRI in the
system you are bui Iding.
Answer Y if you want to bui Id TRI.
>* 06.00 Do you want to bui Id the DUK task? [D=N]

[YIN]:

The DUK task dumps the KMX/KMY microcode onto a disk. You can
then run the KDA task to convert the dump to a printable format.
You require these tasks only if you have a KMX or KMY device.
Answer Y if you want to bui Id DUK.
>* 07.00 Do you want to bui Id the KDA task? [D=N]

[YIN]:

The KDA task converts a KMX/MKY microcode dump on a disk into a
printable format. You requi re this task only if you have a
KMX or KMY device.

NETGEN Dialog:

PSI

7-19

Answer Y if you want to bui Id KOA.
>* 08.00 How many X.29 terminals are supported [0 R:l. -40.]:
Specify the maximum number of X.29 terminals that you want to
operate simultaneously.

In response to Question 8.00, do not specify a value that exceeds the number of HT:
units you specified in DEC Section 6 Question 3.02.
>; <EOS> Do you want to:
>*
<RET> -Continue, R - Repeat section, P - Pause,
E - Exit.
>;

=============================================================

>; PSI - PSI CEX Product Generation Procedure
>;
All quest ions answered at time on date
NOTE

At this point, the network generation procedure proceeds to NET Section 8 (Chapter 5).

7-20

DECnet-RSX Network Generation and Installation Guide

8
Configuring a Pregenned RSX-11 M-PLUS End Node

This chapter applies only to users who are configuring a pregenned RSX-llM-PLUS
end node.

8. 1

Preliminary Considerations

Before running the DECnet configuration procedure, it is recommended that you
•

Make a copy of your target disk. The target disk will be write enabled during this
procedure.

•

Make sure you have enough free space (2200 blocks) to run this procedure.

8.2

Starting the NETGEN Procedure

To run the DECnet configuration procedure,
•

•

Mount the DECnet-llM-PLUS distribution disk
MCR> MOU dduu:ENDNODEKIT
Invoke the NETGEN command file by entering:
@dduu: [137,lO]NETGEN.CMD

The variable dduu refers to the kit disk.
The following paragraphs describe the NETGEN dialog format, organization, and operating modes.

8-1

8.2.1

NETGEN Dialog Format

The NETGEN dialog format consists of questions, optional explanatory text, formats
for user responses, default values, and responses for break questions.
8.2.1.1 Explanatory Text - During the NETGEN dialog, you can read expanded
information for a question. You elicit this text on a question by question basis by pressing the < ESCAPE> key after the question is displayed. This causes the NETGEN command file to issue an explanation and then reissue the question. For example, if you
press < ESCAPE> in response to the question:
> 02.00 Target system device? [dduu, D=SYOO:] [S]:

NETGEN prints the following explanatory text and then repeats the question:
This device wi I I contain al I fi les required to run DECnet. This
includes al I task fi les, databases, command fi les, and
generation work fi les. This must be a Fi les-ll disk device.
The trai I ing colon (e.g. SY:) is optional. The default device
is SY:.

NOTE
The explanatory text above and after each question in section 8.3, is a
word-by-word representation of what you will see on your screen
when you configure a pregenned RSX-IIM-PLUS end node. It may
not, however, be a line-by-line representation of what appears on your
screen.
8.2.1.2 Response Formats - Your response to NETGEN questions will be ope of
three types:
•

YES/NO [Y /N]

•

Numeric [N]

•

Character string [S]

NETGEN's default answer for a YES/NO question is NO unless otherwise stated.
NOTE
If you respond to most NETGEN questions using default answers, you
will generate a working system.

Numeric responses are octal or decimal. In the example [0 R: 1-377 D:5], 0 indicates
the octal number, R indicates the range, and D indicates the default. In the example
[D R: 1.-16. D: 1.], the first D indicates the decimal number, R indicates the range, and
the second D indicates the default. You do not need to include a decimal point after the
number you enter; it is assumed.

8-2

DECnet-RSX Network Generation and Installation Guide

Character string responses can include letters, numbers, and special characters. In the
example [dduu,D = DB01 :][S] , dd refers to 2 letters for the device, uu refers to 2 numbers for the unit, D refers to the default, and S refers to a string that can include letters,
numbers and special characters. See Table 8-1 for a summary ofNETGEN responses.
Table 8-1 :

Summary of NETGEN Responses

Response Type

NETGEN displays:

You enter:

YESINO

[D = N] [YIN]:

Y to indicate YES, N to indicate NO, or press
< RET> for default NO.

Numeric
Octal

[0 R:m-n]:

An octal number in the range of m through n.

[0 R:m-n D:d]:

An octal number in the range of m through n or
press < RET> to indicate that the default d should
be used.

[D R:m.-n.]:

A decimal number in the range of m through n.

[D R:m.-n. D:d.]:

A decimal number in the range of m through n or
press < RET> to indicate that the default d should
be used. You do not need to include a decimal
point after the number you enter.

[S]:

A character string of any length.

[f, D:d][S]:

A character string in the formf(such as ddu:), or
press < RET> to indicate that the default d should
be used.

[S R:m.-n]:

A character string from m through n characters in
length.

Numeric
Decimal

Character
String

8.2.2

Responses for End-of-section and Break Questions

!fyou want to interrupt normal processing, enter < CTRL/Z > . This displays the following break question:
>;<BRK> Do you want to:
>
<RET>-Continue, R-Repeat Procedure, P-Pause,
E-Exit[S] :

Respond by entering one of the following options:

< RET> Continuing with the dialog.
R

Repeating the dialog you just interrupted. Responses are discarded, and the
entire dialog is repeated.

Configuring a Pregenned RSX-11 M-PLUS End Node

8-3

Pausing from the dialog. This response temporarily suspends the indirect
command file, and gives you direct access to the operating system to perform
a task.
The system displays one of the following messages:
AT. --PAUSING. TO CONTINUE, TYPE "RES string"

or
AT. --PAUSING. TO CONTINUE, TYPE "UNS string"

After you have completed your task, you can continue processing by entering:
RES string

or
UNS string

Enter RES or UNS, depending upon which one of the above messages was displayed, and enter the same string contained within that message. After the
Pause, the break question is repeated.
E

8-4

Exiting from NETGEN. No responses are saved. Entering < CTRL/Z > in
response to an EOS question is identical to entering E.

DECnet-RSX Network Generation and Installation Guide

8.3

The Pregenned RSX-11 M-PLUS Dialog

The pregenned RSX-IIM-PLUS procedure displays the following dialog:
>; =============================================================
>; NETGEN - Installation Procedure For Pregenned RSX-IIM-PLUS
>;
End Nodes
Started at time on date
>;
>; =============================================================
>; 01.00 Do you want to see the NETGEN notes/cautions? [D=N]

[YIN] :

No~e the fol lowing:

. This procedure instal Is the DECnet V3.0 software that wi I I
run with the pre-sysgenned RSX-IIM-PLUS V3.0 system. The
software instal led by this procedure wi I I not run with
earl ier RSX-I1M-PLUS systems.
It is recommended that you back up your target disk before
continuing since the disk wi I I be write enabled during the
NETGEN sequence.
Your target disk must have a minimum of 4000 blocks free.
The fol lowing instal led tasks are required on your host system
whi Ie you are performing your NETGEN:
PIP is always required
LOA and UNL if device drivers require loading.
MOU and DMO if disks require mounting.
UFD if the required UICs do not already exist.
If these tasks are not already instal led, and you are logged
in as a privileged user, NETGEN will install them
automatically. If NETGEN requi res a task that it cannot
install, it wi II pause to allow you to correct the situation.
For questions with defaults, the default may be selected by
pressing a carriage return «RET» in response to the
question.
Additional explanatory text for each question may be obtained
by pressing the ESCAPE key.
It is possible to interrupt the network installation
procedure by typing <CTRL/Z> in response to any question.
You wi I I then be given the options of either continuing,
restarting the procedure from the beginning, temporari Iy
pausing, or terminating the procedure.
>* 02.00 Target system device? [dduu, D=SYOO:] [S]:
This device wi I I contain al I fi les required to run DECnet. This
includes al I task fi les, databases, command fi les, and
generation work fi les. This must be a Fi les-ll disk device.
The trai I ing colon (for example, SY:) is optional. The default
device is SY:.
Configuring a Pregenned RSX-11 M-PLUS End Node

8-5

>* 03.00 UIC Group Code for OECnet? [0 R:1-137 0:5] [5]:

5pecify the UIC Group Code under which al I OECnet fi les wi I I
reside. Work fi les, library fi les, object fi les, task fi les and
database fi les wi I I al I appear under various user codes within
this group (for example [5,*]). Fi les which already exist under
these UICs may be deleted or purged.
Note that the user must have privi leges to create, write, read,
and delete al I fi les under the chosen UIC Group Code.
>* 04.00 What is the ta rget node name? [5 R: 0-6]:

[5]:

Enter the node name for the OECnet node being generated. This
must be 1 to 6 alphanumeric characters, at least one of which
must be an alphabetic (A-Z).
Although it is not requi red, it is highly recommended that this
name be used by al I other nodes in the network to identify this
node.
Note that the node name can later be changed with CFE.
>* 04.01 What is the target node address [0 R:1.-1023. 0:1.]

Enter the node address that has been assigned for the node being
generated. The format is:
AREA.NOOE~UMBER

(Example: 3.17)

An AREA value must be entered if you want to specify the area in
which the node being generated wi I I be located. If no value is
specified for AREA, a value of 1 is assigned. AREA, therefore,
must be specified for nodes that you do not want to include in
area 1. AREA is a decimal value in the range of 1-63.
NOOE~UMBER specifies the address of the node within a given
area. It must be a unique decimal value within the area, in the
range of 1-1023.
The node address, consisting of the AREA and NOOE~UMBER fields,
wi I I be the address of this node within the network. If you are
entering values in both the AREA and NOOE~UMBER fields, you
must also enter the dot between the fields.
More efficient operation can be obtained by setting the node
numbers of the nodes in the area in the range 1 to the number of
nodes in the area. The highest node number would then be equal
to the number of nodes in the area.
Note that the node address may later be changed with CFE.
>* 04.02 Target node IO? [O=None]

[5 R:0.-32.]:

Enter the 10 string for the node being generated. This string
may be from 0 to 32 characters. Note that this string must not
contain any single-quote ('), double-quote (") or comma (,)
characters.

8-6

DECnet-RSX Network Generation and Installation Guide

The node 10 may be any arbitrary string which is useful in
identifying the node.
To enter a nul I node 10 string, type a single space, fol lowed by
a RETURN.
>* 05.00 Remote node name [<RET>=Done]

[S R:0-6]:

You may, at this time, define a series of node names which
correspond to nodes in your network. When the network is
loaded, these node names wi I I automatically be set. Enter
a RETURN when you have no more remote node names to define.
Each node name wi I I be assigned to a node address. Standard
procedure is to start with the node name for the lowest node
address used, and to progress from there to the highest node
address.
It is recommended that each node in the network assign the same
node name to each individual node, reducing the confusion of
moving from node to node.
Remote node names can be added or deleted using NCP, VNP, or
CFE.
>* 05.01 What

is the remote node address [0 R:1.-1023. D:1.]:

Enter the node address that node name has been assigned in
the network. The format is:
AREA.NODE~UMBER

(Example: 3.18)

Where AREA is the area number within the network and
NODE~UMBER is the node number within that area. The area must
be a decimal number in the range of 1 to 63. If only
NODE~UMBER is specified, the area that was specified for the
node being generated, wi I I be used as the default area number.
The node number within the area must be a decimal number in the
range 1 to 1023. If you are entering values in both the AREA
and NODE~UMBER fields, you must also enter the dot between
the fields.
Each node address defines a unique node in the network, and must
be assigned when the network is first set up.
The remote node address can be changed using NCP, VNP, or CFE.

The variable node-name refers to the name you specified in Question 04.00.
>* 06.00 Device Driver Process name [S R:0-3]:

The Device Driver Module (DDM) Process name identifies a process
which controls an 1/0 device.

Configuring a Pregenned RSX-11 M-PLUS End Node

8-7

The legal DECnet device driver process names for a UNIBUS
processor are:
Name
DMC
DMP
DUP
PCl
UNA

Physical device
DMCII or DMR11
DMPll
DUPll
PClll
DEUNA

Name
Dl
DZ
DH

Physical device
Dlll
DZ11
DHU11

The legal DECnet device driver process names for a Q-BUS
processor are:
Name
DMV
DPV
QNA

Physical device
DMVII
DPVll
DEQNA

Name
DlV
DZV
DHV

Physical device
DlVll
DZVll or DZQ11
DHV11

>* 06.01 CSR add ress fo r dev-x? [0 R: 160000-177774] :

Specify the Control/Status Register (CSR) address for the
controller.
Note that the CSR address can later be changed with NCP, VNP,
or CFE.
>* 06.02 Vector address for dev-x [0 R:4-774]:

Specify the interrupt vector address for the controller.
Note that the vector address can later be changed with NCP, VNP,
or CFE.

Question 6.03 is displayed for DLII devices only.
>* 06.03 Is dev-x a Dl11-E? [D=N] [Y /N] : ,

Answer the question YES only if the device is a DlII-E or
DlVII-E. These are the only Dlll options that al low the Clear
To Send signal to pass through the interface.

Questions 6.04 and 6.05 are displayed for asynchronous devices only.
>* 06.04 Is modem signa I mon i tor i ng requ i red for dev-x [D=N]? [Y /N] :

If the I ine is connected to a modem for a dial-up switched
ci rcuit service, you should select modem monitoring. Modem
signal monitoring checks the Carrier Detect signal (CD or CAR)
fo r asynch ronous lines.
If this signal is missing for two consecutive samples, taken at
one second intervals, the I ine wi I I be declared "down" without
waiting for a DECnet protocol to detect the fai lure.
If your modem supports this signal and you wish modem
monitoring, answer this question YES.

8-8

DECnet-RSX Network Generation and Installation Guide

>* 06.05 Does the modem switch carrier with Request To Send [D=N]?
[YIN] :

It is necessary to distinguish asynchronous I ine modems which
present continuous carrier once the connection is establ ished,
such as the BEll 103 or BEll 212, from those which switch
carrier along with Request To Send, such as the BEll 202.
Answer the question YES if you are using a BEll 202 modem or an
equivalent.
>* 06.06 Is dev-x a multipoint tributary [D=NJ? [YIN]:

Answer this question YES, only if the I ine is a tributary on a
multipoint line.
If you answer Question 6.06 YES, the procedure displays Question 6.07.
>* 06.07 What is the tributary address of dev-x? [YIN]:

Specify the tributary address which is to identify this
multipoint tributary to the multipoint master. Each multipoint
tributary must be identified by a unique tributary address.
Note that the tributary address can later be changed with NCP,
VNP, or CFE.

Question 06.08 is displayed for PCL devices only.
>* 06.08 What is the TOM bus address for dev-x? [YIN]:

The DECnet Satel I ite Support Components are:
Dll - The down-I ine system loader, host component
DUM - The up-I ine system dumper, host component
CCR - The Console Carrier Requester, host component
HlD - The down-I ine task loader, host component
If you answer YES to this question, al I of the Satel I ite Support
Components wi I I be generated. If you answer NO to this
question, none of the Satel I ite Support Components wi I I
be generated.

CCR will be displayed here only if you specified an Ethernet device in Question 6.00 of
this procedure.

Configuring a Pregenned RSX-11 M-PLUS End Node

8-9

>; =============================================================
>; NETGEN - Installation Procedure For Pregenned RSXII-M-PLUS
>.,
End-Nodes
>.,
Quest ion/answer sect ion completed at time on date
>; =============================================================

AI I the questions have been answered. The installation
procedure wi I I now move the required fi les onto your target
disk. This may take from five to ten minutes to complete,
depending on the system you are running on.
Copy ng the CEX Components
Copy ng the CEX Support Components
Copy ng the System Management Uti I ities
Copy ng the DECnet Communications Components
Copy ng the DECnet Network Management Components
Copy ng the DECnet Fi Ie Uti I ities
Copy ng the DECnet Terminal Uti I ities
Copy ng the DECnet Satel I ite Support Components
Creating NETCFE.CMD,
Creating NETCFG.TXT,
Creating NETINS.CMD,
Creating NETREM.CMD,

the Network Configuration Command fi Ie.
the Network Configuration Description fi Ie.
the Network Installation Command fi Ie.
the Network Removal Command fi Ie.

>; =============================================================
>; NETGEN - Installation Procedure For Pregenned RSXII-M-PLUS
>.,
End-Nodes
>.,
The f i I e move phase comp I eted at time on date
>; =============================================================

Your DECnet software should now be on your target disk. If
there were any errors reported whi Ie the software was
being moved to the target disk, correct the problem and
rerun the installation procedure. Your system configuration
is the fol lowing:
The I ine parameters are:
Dev i ce=device name, full or half duplex, synch or asynch
CSR=csr, Ve c tor =vector , P rio r i t y=device priority
Speed=speed, S tat e=ON
The Executor parameters are:
Node name=name, Add ress=node address
Max i mum I i nks=max links, Max i mum hops=max hops
The System parameters are:
Maximum control buffers=xxx, Minimum receive buffers=yyy
Maximum large buffers=zzz, Large buffer size=nnn
If you want to change any of these parameters in the permanent
database, use the fol lowing CFE commands:
DEFINE LINE
DEFINE CIRCUIT

- wi I I change the I ine parameters
- wi I I change circuit parameters (none were
listed above)
DEFINE EXECUTOR - wi I I change the executor parameters
DEFINE SYSTEM
- wi I I change the system parameters

8-10

DECnet-RSX Network Generation and Installation Guide

See the DECnet-RSX Guide to Network Management Uti I ities for
more detai Is on CFE or NCP commands.
The startup fi Ie that was provided with your RSX-llM-PLUS system
may be set up to deal locate the DECnet Communications Executive.
If this is sti II in your startup procedure, change it so the
DECnet Communications Executive is not deal located, the next
time your system is booted. The startup fi Ie also indicates how
to instal I DAPRES. If you want RMS access to remote fi les,
instal I DAPRES as indicated in the startup fi Ie.

To change the startup file:
1.

Use your favorite editor to
([ 1, 2]SYSPARM.DAT).

access

Change DECNET = NO to DECNET = YES.

the

startup procedure's

data file

3. If you want the resident DAP library, change DAPRES = NO to DAPRES = YES.
Once you are satisfied with the configuration, you may bring up
your network by:
Setting up the network UIC with the SET INETUIC=[xxx,54]
command, then invoke dduu: [xxx,OOl]NETINS.CMD. The network
UIC may also be permanently set with VMR.
>;
>;
>;
>;
>;

==============================================================
NETGEN - RSX-llM-PLUS Pregenned End Node Installation
Procedure
Stopped at time and date

==============================================================

>@ <EOF>

You should now proceed to Chapter 9 for information on the use of the NETINS. CMD
command procedure.

Configuring a Pregenned RSX-11 M-PLUS End Node

8-11

9
Installing DECnet

This chapter tells you how to install and load DECnet on various host systems. It also
tells you how to automate the installation procedure and how to modify the network
installation command file (NETINS.CMD) to meet certain needs.
NOTE

Before using the NETINS.CMD procedure to bring up the network, you should
set the network VIC. Vse the following VMR or MCR command:

SET INETUIC=[x,54]
If you set the wrong VIC, NETINS.CMD will use the set VIC and may load the

wrong network.

9.1

Steps Prior to Network Installation

To verify that you specified DECnet-RSX system information correctly during the
NETG EN process, use one of the following procedures:
•

Examine the network component parameters using the Configuration File Editor
(CFE)

•

List and examine NETCFE.CMD and NETCFG.TXT, both located under the network VIC [xxx, 1]

These techniques are described in the following sections.

9-1

9.1.1

Use of the Configuration File Editor (CFE)

CFE can be used to examine the data for the network configuration. (For more information on CFE, refer to the DECnet-RSX Guide to Network Management Utilities.) Use
the following CFE commands to determine how the component parameters were set
during NETGEN:
LIST SYSTEM
LIST EXECUTOR
LIST KNOWN LINES
LIST KNOWN CIRCUITS
LIST KNOWN LOGGING
LIST KNOWN NODES
LIST KNOWN OBJECTS
The following CFE commands are useful only in DECnet-RSX networks that interface
with a Packet Switching Data Network (PSDN) by means of the Packetnet System Interface (PSI) product:
LIST MODULE X25-ACCESS
LIST MODULE X25-PROTOCOL
LIST MODULE X25-SERVER
LIST MODULE X29-SERVER

9.1.2

Examination of NETCFE.CMD and NETCFG. TXT

The NETCFE.CMD file is a journal of all the commands entered and executed during
NETGEN. This file also lists the CFE commands executed in the selection of any of the
default values implicitly or explicitly specified during NETGEN.
The NETCFG. TXT file contains a narrative description of the system configuration as
formed during NETGEN.
Both files are a product of network generation. They accurately reflect the configuration of the network at NETGEN. To use the information in these files, list one or both
on a hard-copy device and verify that the results of the network generation reflect the
options that you intended to specify.

9-2

DECnet-RSX Network Generation and Installation Guide

9.2

Installing DECnet on RSX-11 M/M-PLUS Systems

If you are ready to install and load the network tasks and components as produced by

NETGEN, execute the network installation command file. To do this, enter:
@dduu: [x,l]NETINS.CMD

The variable dduu: refers to the target disk, and the variable x refers to the group code
you specified in NETGEN.
If you do not want to execute the NETINS.CMD file produced by NETGEN because you

want to automate the installation procedure, or because you want to modify this file to
meet certain unique needs, refer to Section 9.2. 1.
Figure 9-1 illustrates the installation procedure for an RSX-IIM/IIM-PLUS system
using the NETINS.CMD file as produced by NETGEN.

Installing OECnet

9-3

TW0187

Figure 9-1 :

9-4

Installation Procedure for an RSX-11 M/11 M-PLUS System

DECnet-RSX Network Generation and Installation Guide

NETINS.CMD displays the following six questions. If LAT is not included in your configuration, Question 3 is not displayed. If you do not have a license for PSI, Question 4
is not displayed.
1.

>* Do you want to instal I and load the CEX system? [YIN]

If you want to install all tasks and load the network partitions and Communications

Executive that are common to both DECnet and PSI, answer this question YES.
2.

>* Do you want to instal I and start DECnet? [YIN]

If you want to install all DECnet-specific tasks, answer this question YES. This

causes the network to be turned on. Any circuits marked for enable will be turned
on. If you have already loaded DEC, answer this question NO.
For more information on setting circuit states, see the DECnet-RSX Network Management Concepts and Procedures manual.
3.

>* Do you want to st art LAT? [Y IN]

If you want LA T started, answer this question YES. This causes the LCP program to

issue a START command which makes your service node available to LAT terminal
server users. If you have already started LAT, answer this question NO.
4.

>* Do you want to instal I and start PSI? [YIN]

If you want to use DECnet with data link mapping (DLM) or you want to use the PSI

User Interface, answer this question YES. Answering YES enables the PSI lines that
have been previously marked for enable, and enables the X.25 Server. If you have
already loaded PSI, answer this question NO.
5.

>* On what device are the network tasks? [D=dduu:]

Enter the name of the device for the NETGEN target disk where the network tasks
are located.

>* What is the network UIC group code? [D=xxx]

The variable xxx refers to the group code of the UIC you are currently in.
If you are bringing up the node for the first time and you want to make sure the node

was generated properly, consult Chapters 10 and 11 for the proper testing procedure.
For more information on the system management commands required to bring up a
node, see the DECnet-RSX Network Management Concepts and Procedures manual.

Installing DECnet

9-5

9.2.1

Modifying NETINS.CMD
CAUTION

In the standard NETINS.CMD file generated by NETGEN, the installation of
CFE and VNP has been disabled by placing a leading period and semicolon (.;)
on the line installing each of these programs. CFE and VNP are privileged utilities that should be used only by system managers. If you modify these lines to
allow CFE and VNP installation, any user can perform CFE and VNP operations.
If you want to modify the NETINS.CMD file to meet certain needs, you must edit

NETINS. CMD. Any of the following changes require you to edit NETINS. CMD.
•

Add a new component

•

Install a task in a different partition

•

Save DSR pool space

•

Change the way a task is installed

•

Install and start the network automatically

Adding a new component: If you add a new component such as FAL, NFT, or TLK in
component mode, you must add this component to the NETINS.CMD file. For example, to install TLK, enter a line such as the following:
INS XX: [371,054]TLK

Installing a task in a different partition: If you want to install a task in a partition
other than GEN, add /PAR = xxxxxx to the appropriate line. The variable xxxxxx refers
to the new partition for the installed task. For example, to install NETACP in the partition ACPPAR, modify NETINS.CMD as follows:
INS XX: [371,054]NETACP/PAR=ACPPAR

In this case, the group code, 371, is a variable. For you, this group code number must be
the value you specified in NETGEN.

Saving DSR pool space: If you want to save DSR pool space, you can prevent
NETINS.CMD from installing certain tasks. To do this, place a semicolon (;) or a period
and semicolon (.;) before the INS command on the line for that task. The semicolon (;)
form causes the line to print, while the period and semicolon (.;) form does not. This
procedure is preferable to deleting the line because you will retain a record of the original NET INS .CMD version. You can install these tasks later by using the appropriate
MCR commands.
Changing the way a task is installed: If you want to change the way a task is
installed, you may need to edit the NETINS.CMD file. If you specified memory-resident
RMS in an original network generation, NETINS. CMD will automatically check to see if
the memory-resident RMS library was installed. If you did a component mode generation and you generated a version of FAL that does not require memory-resident RMS,
you should remove this check by editing the NETINS.CMD file.

9-6

DECnet-RSX Network Generation and Installation Guide

Installing and starting the network automatically: If you want to install and start
the network automatically, you must edit the NETINS.CMD file and define the values
$CEX, $DEC, $LAT, and $PSI. If $CEX, $DEC, $LAT, or $PSI are defined and set TRUE,
NETINS.CMD automatically installs and loads these products. If $CEX, $DEC, $LAT, or
$PSI are defined and set FALSE, NETINS.CMD does not install or load these products.
After you define these values, you must also specify the device name for the disk where
the network tasks are located and the network UIC.

If you define the products, the device name, and the network UIC, NETINS.CMD performs all functions without displaying any questions. If any of these are not defined,
NETINS. CMD displays the appropriate question.
To modify the NETINS.CMD file, add the following lines to the beginning of the file:
1.

To define $CEX, $DEC, $LAT, and $PSI, enter:

.SETT $CEX
.SETT $DEC
.SETT $LAT
.SETT $PSI
In this example, all values are defined TRUE. You can change these values (TRUE or
FALSE) to meet your own needs. To assign a device name (XX:) for the target disk
device (dduu:), add the following line:

ASN dduu:=XX:
The network task device (dduu:) is the target device you specified in NETGEN.
2.

To define the network UIC group code, add the following line:

SET INETUIC=[x,54]
Or execute the following VMR command on the RSX-IIM/-IIM-PLUS system
image:

VMR> SET INETUIC=[x,54]
The variable x refers to the group code specified in NETGEN.
Finally, to invoke the NETINS. CMD file automatically, add the following line to the system startup command file:
@dduu: [x,l]NETINS.CMD

The variable x refers to the group code specified in NETGEN.
If you follow this procedure, NETINS.CMD installs and loads the network software

automatically.
After installing and loading the network for the first time, you should execute the testing procedures described in Chapters 10 and 11.

Installing DECnet

9-7

9.2.2

Using VMR to Install Network Tasks

An optional way of starting the network is to use the Virtual Monitor Routine (VMR)
commands to install network tasks. You can use individual VMR commands and/or you
can edit the SYSVMR.CMD file. (See Section 3.2.5.10 for information about
SYSVMR. CMD.)
Using VMR has three advantages over using the NETINS.CMD file:
•

VMR reduces startup time because the system already contains the tasks in installed
form when booted.

•

Installing tasks with VMR minimizes fragmentation in the Dynamic Storage Region
(DSR) pool.

•

There is no possibility of the RSX-11 M loading problem discussed in Section 3.2.2.

The commands for VMR are essentially the same as the MCR commands used in
NETINS. CMD. There are two exceptions - the ClI command and the load command. The ClI command and the /HIGH switch of the load command are not supported in VMR. Instead of the lOA NM:/HIGH command used by NETINS.CMD, use the
following command:

LOA NM:
The CLI command in NETINS.CMD that deals with RMTACP
(ClIlINIT = RMTACP ... ) should be deleted and issued as a command to MCR.
After making these two changes you should include the commands found in the
NETINS.CMD file in the SYSVMR.CMD file.
NOTE

OnRSX-IIM and RSX-I1M-PlUS systems, you cannot use the VNP SET EXECUTOR STATE ON command. If you wish the network to be loaded automatically, you must include the NCP SET EXECUTOR STATE ON command in your
system startup file.
For more information on Virtual Monitor Routine (VMR) commands, see the RSX-IIMI
M-PLUS System Management Guide.

9.3

Installing DECnet-11 S Systems

To install DECnet on an RSX-IIS system, use the virtual monitor routine with
NETINS.CMD (see Section 9.3.1). Then use VNP to load the network into the RSX-IIS
system image (see Section 9.3.2). For a description of down-line system loading or
down-line task loading, see the DECnet-RSX Network Management Concepts and Procedures manual.

9-8

DECnet-RSX Network Generation and Installation Guide

9.3.1

Installing Tasks Using VMR Commands

To install tasks using VMR commands, perform the following procedures:
1.

MCR> SET IUIC=[1,64]

2. Invoke the virtual monitor routine:
MCR> VMR

3. VMR then prompts you to enter the RSX-llS system image filename:
Enter fi lename: RSXllS.SYS
4.

Invoke the NETINS.CMD file:
VMR> @dduu: [x,l]NETINS
The variable dduu: refers to the target disk and the variable x refers to the group
code you specified in NET Section 1, Question 4.00. VMR executes the
NETINS.CMD file and produces a preinstalled, ready-to-be-booted system image,
but the network will not be usable unless you load it and turn it on with VNP before
booting.

Example 9-1 displays a NETINS.CMD file for an RSX-llS system generated with all
optional components. NTINIT and NETACP are required tasks. IfRCPl or MLD appear
in your command file, they are also required. Remaining tasks are optional. If you are
installing EVC, you should load NM:. If you are installing SLD, you should load OV:.
Example 9-1:

Typical NETINS.CMD for an RSX-11 S System

SET INETUIC=[xxx,64]
INS NTINIT/FIX=YES
INS MLO/FIX;"'YES
INS EVC/FIX=YES
INS NCP/FIX=YES
LOA NM:
INS NETACP/FIX=YES
INS RCP1/FIX=YES
INS NICE/CKP=NO/FIX=YES
INS EVR/FIX=YES
INS NTO/CKP=NO/FIX=YES
INS NTOEMO/CKP=NO/FIX=YES
INS MIR/CKP=NO/FIX=YES
INS SLO/FIX=YES
LOA OV:
INS FAL/CKP=NO/FIX=YES
INS RMT/CKP=NO/FIX=YES
INS RMTACP/FIX=YES
LOA HT:
INS RMHACP/CKP=NO/FIX=YES
INS TLK/CKP=NO/FIX=YES
INS LSN/FIX=YES
INS TCL/CKP=NO/FIX=YES

Installing DEenet

9-9

If you want to change the way tasks are installed, you must either modify the

NETINS.CMD command file before using it with VMR, or at a later time, modify the system image with VMR.
•

If you are not installing SLD and loading OV:, you cannot down-line load or up-line

checkpoint tasks. Tasks should be fixed in memory using the FIX = YES switch and
installed as noncheckpointable. For example:

INS RMT/CKP=NO/FIX=YES
•

If you are installing SLD and loading OV:, you can down-line load and/or up-line

checkpoint tasks. In either case, do not use the FIX switch. NETGEN builds all tasks
that can be checkpointed as checkpointable. If you want to install a task as
checkpointable, do not specify a switch. For example:

INS RMT
If you want to install a task as noncheckpointable, specify /CKP = NO. For exam-

ple:

INS RMT/CKP=NO

9.3.2 Loading the Network Using VNP Commands
After using VMR to install your tasks, use the Virtual Network Processor (VNP) commands to load the network. To load tasks using VNP commands, perform the following
procedures:
1.

Enter:

RUN dduu: [x,64]VNP
The variable dduu: refers to the target disk and the variable x refers to the group
code you specified in NET Section 1, Question 4.00.
2.

VNP then prompts you to enter the RSX-ll S system image file name:

Enter filename: RSXllS.SYS
3.

The following SET command will start the network and enable circuits that were
requested to be enabled during NETGEN:

VNP> SET SYSTEM
VNP> SET EXECUTOR STATE ON
4.

To load lines not previously set to ON, enter:

VNP> SET LINE line id ALL
5.

To enable circuits not previously set to ON, enter:

VN P> SET CI RCU I T circuit id STATE ON
6.

If YOll want to check the states of your lines and circuits, enter the following two

commands:

VNP> SHOW LINE line id
VNP> SHOW CI RCU I T circuit id

9-10

DECnet-RSX Network Generation and Installation Guide

For more information on using these commands, see the DECnet-RSX Guide to Network Management Utilities.

After preparing the system image as just described, you have a ready-to-be-booted system image. If you need more information on how to boot or down-line load an RSXlIS system, refer to the DECnet-RSX Network Management Concepts and Procedures
manual.
After installing and loading the network for the first time, you should execute the testing procedures described in Chapters 10 and 11. If you wish to perform these tests you
should also set up a loop node for the circuit(s) you will be testing. To set up a loop
node for a circuit enter the following command:
VNP> SET NODE loop-name C I RCU IT circuit-id

where loop-name is the name of the loop node and circuit-id is the name of the circuit
you will be testing. You must do this with VNP prior'to running the loop tests because
RSX-11S NCP does not support setting up a loop node. For more information on loop
node testing, see the DECnet-RSX Network Management Concepts and Procedures
manual.

9.4

Installing a DECnet-11S System on a VAX/VMS System

This procedure (intended for users who are installing DECnet-11S on a VMS Version
4.0 or later system) is very similar to the procedure described in Sections 9.3.1 and
9.3.2. There are three exceptions:
•

You must have certain privileges set prior to running VMR or VNP.

•

You must login using MCR as your CLI.

•

You must use a different command to set your VIC group code.

9.4.1

Establishing VAX/VMS Privileges

You must be logged in to the VAX/VMS system with certain privileges in order to set up
and run the RSX-11S network generation. You should have access to the following
required privileges:
•

Change mode to kernel privilege (CMKRNL). This allows you to change default
VICs and directories. You must have this privilege before you can set the remaining
privileges.

•

Logical I/O privilege (LOG_IO). This privilege permits RSX-IIM VMR to do logical I/O on task image files.

•

System protection privilege (SYSPRV). This privilege allows you to create files in
directories not owned by the current VIC.

Installing DECnet

9-11

If you have the SETPRV privilege, you can establish the required privileges with the fol-

lowing MCR command:

> SET PROC/PRIV=(privi lege[, ... ])
To find out what privileges you have, use the following MCR command:

> SHO PROC/PRIV
For more information on using these privileges, see the VAXNMS System Manager's
Guide.

9.4.2

Installing Tasks Using VMR Commands

After you have set your privileges, you can install tasks using VMR commands. The following steps permit you to begin this procedure on a VAX/VMS system.
1. Log in using the /CLI = MCR qualifier so that you can use MCR -

for example,

Username: SYSTEM/CLI=MCR
Password:
2.

Set up the required privileges (Section 9.4.1).

3. Mount the RSX-11S system disk and NETGEN target disk.
> M~OUNT dduu: label
4.

Set your default disk and VIC to [x,64] on the target disk -

for example,

> SET DEF dduu: [1,64]
The variable x refers to the VIC group code you specified in NET Section 1 Question 4.00.
5.

Invoke the virtual monitor routine:

> VMR

6. VMR then prompts you to enter the RSX-11S system image filename:
Enter fi lename: RSX11S.SYS
7.

Invoke the NETINS.CMD file:

VMR> @dduu: [x,l]NETINS
The variable dduu: refers to the target disk and the variable x refers to the group
code you specified in NET. VMR then executes the NETINS.CMD file.

9-12

DECnet-RSX Network Generation and Installation Guide

9.4.3

Loading the Network Using VNP Commands

After setting up the required privileges and using VMR to install your tasks, use the Virtual Network Processor (VNP) commands to load the network. To load tasks using VNP
commands, perform the following procedures:
1.

Enter:
RUN dduu: [x,64]VNP

The variable dduu: refers to the target disk and the variable x refers to the group
code you specified in NET Section 1, Question 4.00.
2.

VNP then prompts you to enter the RSX -11 S system image filename:
Enter fi lename: RSXllS.SYS

The following SET command will start the network and enable circuits which were
requested to be enabled during NETGEN:
VNP> SET SYSTEM
VNP> SET EXECUTOR STATE ON

To load lines not previously set to ON, enter:
VNP> SET LINE line-id ALL

To enable circuits not previously set to ON, enter:
VN P> SET C I RCU IT circuit-id STATE ON

If you want to check the states of your lines and circuits, enter the following two

commands:
VNP> SHOW LINE line-id
VN P> SHOW C I RCU IT circuit-id

For more information on using these commands see the DECnet-RSX Guide to Network Management Utilities.

After preparing the system image as just described, you have a ready-to-be-booted system image. For a description on how to boot or down-line load an RSX-IIS system,
refer to the DECnet- VAX System Manager's Guide.
After installing and loading the network for the first time, you should execute the testing procedures described in Chapters 10 and 11. If you wish to perform these tests, you
should also set up a loop node for the circuit(s) you will be testing. To set up a loop
node for a circuit, enter the following command:
VN P> SET NODE loop-name C I RCU IT circuit-id

where loop-name is the name of the loop node and circuit-id is the name of the circuit
you will be testing. You must do this with VNP prior to running the loop tests because
RSX-IIS NCP does not support setting up a loop node. For more information on loop
node testing, see the DECnet-RSX Network Management Concepts and Procedures
manual.

Installing DEenet

9-13

10
Installation Testing

This chapter describes procedures you can use to check your RSX-11S, RSX-11M, or
RSX-11M-PLUS node. Certain restrictions, as outlined throughout this chapter, apply
to testing an RSX-11S node. The checkout procedures consist of three types of tests:
•

The first test determines if your local node software is operational.

•

The second test determines if your local DECnet node and communications hardware are functioning properly.

•

The third test checks your local node's ability to communicate with a remote node
(see Figure 10-1).

All procedures are designed to be initiated and executed from the local node. However,
an operator should be available at each remote node to assist in problem isolation. If
you selected event logging during NETGEN and have disabled event logging using CFE
or NCP, you should enable console event logging prior to performing the tests
described in this chapter. You can enable event logging by using the following command:

NCP>SET LOGGING CONSOLE STATE ON
For more information on the Network Control Program (NCP) commands mentioned
in this chapter, refer to the DECnet-RSX Guide to Network Management Utilties.
If you are testing an RSX-11S node you must have installed NCP and NICE in the RSX11 S system image using VMR. The procedure for installing the network software on an
RSX -11 S system is given in Chapter 9.

10-1

,..----,r-

I QI;E SECTION 10.1
~O~E~L!-..J

~______~__---------J

r;E-;E~:70.:..r;"1
________
______
• FOR DETAILS

~-----------J

,..----':-1

I $EE SECTION 10.3 r~O~E~L!- ..J

~______~__________-J

TW0188

Figure 10-1:

10-2

Installation Test Flowchart

DECnet-RSX Network Generation and Installation Guide

NCP
SETUP
COMMANDS

NONE

USER
LAYER

0
0

E
B

0
S

NETWORK
MANAGEMENT LAYER
NETWORK
APPLICATION LAYER
NSP/SESSION
CONTROL LAYER
TRANSPORT*
LAYER
DATA LINK
LAYER
PHYSICAL LINK
LAYER

COMMUNICA TION
HARDWARE

PHYSICAL LINK
LAYER
DATA LINK
LAYER

0
0

E
T
R

TRANSPORT*
LAYER
NSP/SESSION
CONTROL LAYER
NETWORK
APPLICATION LAYER
NETWORK
MANAGEMENT LAYER
USER
LAYER

COMMENTS

PROGRAM A sends
data toPROGRAM B
which returns data
to PROGRAM A
Target node IS local
BOS

Legend:

0--0

User program or DECnet module
Path travelled by test data
The Transport layer uses IU
routing algOrithm to determine
on which path to loop the test
data.

Figure 10-2:

Local Node Software Test

Installation Testing

10-3

10.1

Local Node Software Testing

The local node test described in this section requires the use of the indirect command
file processor. RSX-llS does not support this command file processor, therefore, for
RSX-11 S systems you should skip ahead to Section 10.2.
This test determines whether the five layers of software are working: the User layer,
the Network Management layer, the Network Application layer, the End Communications layer, and part of the Routing layer. This test does not check the device driver or
the communications device. See Figure 10-2 for the path the test data travels. You may
use other types of tests, but the NTEST.CMD procedure is preferred because it is auto- '
mated. The NTEST.CMD procedure tests and installs the network tasks needed for testing and brings up the network.
If the requirements of Section 10.1.1 cannot be met (for example, a 16K-word system
controlled partition), the TLK and LSN utilities can be substituted for the NTEST .CMD
test.

NTEST.CMD executes two sets of programs: DECnet test sender (DTS) and DEC net test
receiver (DTR), and Network File Transfer (NFT) and File Access Listener (FAL). DTSI
DTR perform message transfer operations using task-to-task communication. NFT IFAL
perform various file transfer operations. Details of the DTS/DTR programs are in
Appendix D. Details ofNFT and FAL are found in the DECnet-RSX Guide to User Utilities.

10.1.1

Preparing to Run the Test

The following conditions must be met before the local NTEST. CMD procedure is executed:
•

The partition in which the network software and the software tools will run must
be system controlled and at least 16K words in length.

•

A NETGEN must be completed.

•

NETGEN must have moved the NTEST.CMD file onto your disk under the DIC
[xxx,24], where xxx is the group code chosen during NETGEN.

•

You must have installed the network using the NET INS procedure or NCP commands. For more information on installing the network, see Chapter 9.

•

You must be logged into a privileged account (if your system supports multiuser
protection).

When you have satisfied the conditions listed above, proceed to the next section and
run NTEST .CMD.

10-4

DECnet-RSX Network Generation and Installation Guide

10.1.2

Running the Test

Invoke the command file by entering the following:
>@[xxx,24]NTEST.CMD <RET>

where xxx is the network software DIC group code specified during NETGEN.
NTEST. CMD begins to execute and prompts you for any input it requires during execution. Example 10-1 is a sample listing of the NTEST.CMD procedure.
NOTE

Your answers to the questions displayed by the NTEST.CMD procedure may
be different from those shown in the sample listing. For example, you may
specify device names, YES and NO options, and group DIC codes that differ
from those shown.
NOTE

The NTEST. CMD procedure is used twice in the testing procedures outlined in
this chapter. For the test described in this section, you should answer YES to
the question "Is this an internal node test?". This same procedure will be used
in the test described in Section 10.3 at which time you should answer NO to
this question.

Installation Testing

10-5

@NTEST.CMD
>SET /BUF=TI: 80
>;

>;===============================================================
>;
RSX-11M/M-PLUS Network Installation Procedures
>;===============================================================
>;

>; Copyright (C) 1979, 1980, 1981, 1982, 1983, 1984, 1985 by
>; Digital Equipment Corporation, Maynard, Mass.
>;

>SET /UIC=[1,1]
>; Test procedure performed at 08:02:34 on 01-JAN-85
>;

>; This procedure is designed to al Iowa user to perform the fol lowing
>; tests on a DECnet-11M/M-PLUS system:
>.,

>;
>;
>;
>;
>;
>;
>;

>.,
>;

Internal node level -- Test the local node without using any
communication circuits.
Circuits level -- Test the local node with a communications line
attached to a turnaround/loopback device or a modem with loopback
capabil ities.
Remote node level -- Test the local node by attempting to communicate with a remote node.

>; The procedure wi I I execute DTS/DTR tests and wi I I use NFT/FAL to
>; transfer and execute fi les.
>;

>; Before running this procedure, you should have access to the fol lowing
>; info rmat ion:
>;

>;
>;
>;
>;
>;
>;

The device and group code of the UIC under which the network
tasks are stored.
The target node name (for local tests, this is the local node
name) .
. The necessary UIC and password to access fi les residing on the
target node, for NFT tests.

>* Do you want to: <RET>-continue E-exit P-pause [S]:
>;

(continued on next page)

Example 10-1:

10-6

NTEST.CMDSample Listing

DECnet-RSX Network Generation and Installation Guide

>;======================================================================
>;
Test Parameters
>;======================================================================
>;
>;
>;
>;

If you do not choose long dialogue mode, explanatory text for each
question is avai lable by hitting the "ESCAPE" key.

>* Do you want
>;
>;
>;
>;

long dialogue mode? [Y/N]: Y

Enter the UIC group code under which the network fi les were stored
for this node's network.

>* What group UIC is the network stored under? [0 R:1-377]:
>;
>;
>;
>;

100

Specify the device on which the network fi les are stored. The default
device is "SY:".

>* On what device are the network fi les (DONN)? [S R:0-5]: OM1:
>;
>;
>;
>;
>;

Answer "YES" to the next question if this is an internal node test,
i.e. the test does not involve conmunications I ines and/or
remot e nodes.

>* Is this an
>;
>;
>;
>;

internal node test? [Y/N]: Y

Enter the target node name (1-6 alphanumeric characters). This is the
node name of the remote node. If this is an internal level test, then
the local node name should be used. If this is a line level test,then

>; the loopback node name should be used.
>;
>* What is the local node name? [S R:1-6]: NOOEA
>;
>; OTS wi II now be invoked to perform three tests:
>;
1.) OTS wi I I issue a connect request to OTR caus i ng
>;
OTR to be loaded, answer the request and task
>;
exit
2.) OTS/OTR wi I I execute a 1 minute pattern test
>;
uSing a 10 byte message size
>;
>;
3.) OTS/OTR wi I I execute a 1 minute pattern test
using an 1100 byte message size.
>;

>.,

>;
>;
>;
>;

The 10 byte message size was selected because this size is smaller
than ECl's segment size ... thus there wi I I be 1 segment for each
message. The 1100 byte message size is larger than the segment size;
therefore, there wi I I be mUlti-segments for each of these messages.
(continued on next page)

Example 10-1 (cont.):

Installation Testing

NTEST.CMD Sample Listing

10-7

>NCP ZERO EXECUTOR COUNTERS
>NCP ZERO SYSTEM COUNTERS
>DTS @DTSTST
LOAD/EXEC DTR VIA CONNECT TEST
DTS -- Test finished at 8:03:30
PAT/MSG=10/TIME=lM
DTS -- Test started at 8:03:30
DTS -- Test finished at 8:04:30
3013 Messages sent
502 Characters/second
4016 Baud
PAT/MSG=1100/TIME=lM
DTS -- Test started at 8:04:31
DTS -- Test finished at 8:05:31
876 Messages sent
16060 Characters/second
128480 Baud
>;
>;
>;

>; In order to execute NFT fi Ie transfer tests, you must supply val id
>; access control information for the target node NODEA. This
>; in format i on IS
>;
(1). A user identification (the format depends on what type of
system NODEA is.
>;
>.,
(2). The password for that user-id.
>;

What is the user-id for NFT tests on node NODEA [S R:l.-16.]:[200,200]

>* What is the corresponding password [S R:0.-8.]: PRIV
>PI P NFTTSI. ZZZ; IDE/NM, NFTTS2; ,NFTTS3; ,NFTTS4; ,NFTTS5;
>;

>; NFT wi I I now be invoked. A copy of the command fi Ie NTEST.CMD wi I I be
>; transferred to node NODEA with the fi lename TMPNETTST.ZZZ using the
>; access control information you specified. NFT wi I I then delete the
>; f i Ie.
>;

>NFT @NFTTSl.ZZZ/TR
NFT>NODEA/[200,200]/PRIV: :TMPNETTST.ZZZ;l=DMl: [100,24]NTEST.CMD
(continued on next page)

Example 10-1 (cont.):

10-8

NTEST.CMD Sample Listing

DECnet-RSX Network Generation and Installation Guide

>; The fi Ie TMPNETTST.ZZZ wi I I now be deleted using NFT.
>;

>NFT @NFTTS2.ZZZ/TR
NFT>NODEA/[200,200]/PRIV: :TMPNETTST.ZZZ;1/DE
>;
>;

>NCP SHOW EXECUTOR COUNTERS
Node counters as of 01-FEB-85 09:24:12
Executor node = 14 (NODEA)
308 Seconds since last zeroed
1009182 Bytes received
1009182 Bytes sent
17658 Messages received
17658 Messages sent
8 Connects received
8 Connects sent
o Response timeouts
o Received connect resource errors
2 Total maximum logical I inks active
o Aged packet loss
o Node unreachable packet loss
o Node out-of-range packet loss
o Oversized packet loss
o Packet format error
o Pa rt i a I rout i ng update loss
o Verification reject
2 Node maximum logical I inks active
o Total received connect resource errors
>NCP SHOW SYSTEM COUNTERS
System counters as of 01-FEB-85 09:25:03
368 Seconds since last zeroed
o Control buffer al location fai led
o Sma I I buffer al location fai led
o Large buffer al location fai led
o Receive buffer al location fai led

>.,

>; If you have reached this point with no errors, then you can be
>; confident that the node has been created properly.
>;

>; End of test at 08:07:12
>;

>;======================================================================
>;
End of Installation Test
>;======================================================================
>SET IUIC=[100,24]
>@ <EOF>

Example 10-1 (cont.):

Installation Testing

NTEST.CMD Sample Listing

10-9

The counter displays illustrated in Example 10-1 reflect an error-free test execution.
Although the values of the counters you actually see may vary slightly, you should verify that the bytes sent counter equals the bytes received counter. Also verify that all
counters shown as having a value of zero have a value of zero in your display. Note that
on a nonrouting node not all the executor counters shown will be displayed.
Check the counters displayed by NTEST for errors. If there are any errors, proceed to
Section 10.1.3.
If there are no errors, proceed to Section 10.2, Node Level Hardware Loopback Circuit
Testing.

10.1.3 NTEST.CMD Failure Conditions and Handling
The most common failure conditions encountered during this test are resource allocation failures or network tasks not installed. The symptoms most commonly seen under
these failure conditions are described below. Identify which class of failure is occurring, take appropriate corrective action, and then repeat the test.
Common symptoms of this class of failure are:
•

Inability to execute the NCP SET SYSTEM command

•

Inability to execute the NCP SET EXECUTOR STATE ON command

•

Inability to establish a logical link

Corrective action. If, at any time, the operation appears to be hung, use the NCP
SHOW SYSTEM COUNTERS command. If any of the buffer allocation failed counters
are incrementing, there is a resource allocation problem. The network should be
unloaded, and the resource allocations should be increased using CFE. Once the system
resources have been increased, reload the network with the NCP SET SYSTEM command. Then set the executor state ON with the NCP SET EXECUTOR STATE ON command.
Allocation failures. Resource allocation failures include one or more of the following
conditions:

•

Insufficient number of large buffers (LD Bs)

•

Insufficient number of control buffers (CCBs)

•

Insufficient number of small buffers (SDBs)

•

Insufficient RSX system pool size (DSR)

•

Insufficient number of logical links

10-10

DECnet-RSX Network Generation and Installation Guide

Other failures. Other reasons for failure can include one or more of the following con-

ditions:
•

Insufficient checkpoint space

•

Fragmented system-controlled partition

•

Hardware problems, including disk errors that can result in nonreproducible system behavior

If there is insufficient checkpoint space, RCP! (routing control process) may not be

able to allocate space for its database and you would receive the following message:
RCP! - Cannot al locate space for database
If NTEST.CMD fails while running DTS/DTR or NFT/FAL, refer to Appendix D (for
DTS/DTR) or to the DECnet-RSX Guide to User Utilities (for NFT/FAL).

Installation Testing

10-11

SET NODE
TESTER
CIRCUIT DMC·O

NCP
SET UP
COMMANDS

USER
LAYER

0
0

E
B

0
S

NETWORK
MANAGEMENT LAYER
NETWORK
APPLICATION LAYER
NSP/SESSION
CONTROL LAYER
TRANSPORT
LAYER*
DATA LINK
LAYER
PHYSICAL LINK
LAYER

I
I
COMMUNICATION
HARDWARE

I
I
I

~
~
\.2...1

I
I
I
I
I

PHYSICAL LINK
LAYER
DATA LII'<IK
LAYER

0
0

E
T

R
N

TRANSPORT
LAYER*
NSP/SESSION
CONTROL LAYER
NETWORK
APPLICATION LAYER
NETWORK
MANAGEMENT LAYER

USER
LAYER

COMMENTS

Notes

Legend:

0--0

PROGRAM A sends
data to PROGRAM B
whIch returns data
to PROGRAM A
Target node IS
loopback name
TESTER

User program or DECnet module
Path travelled by test data

At controiler
At lone loopback connector
At modem

The Transport layer uses Its
routing algorothm to determone
on wh,ch path to loop the test
data.

Figure 10-3:

10-12

Node Level Hardware Loopback Circuit Test

DECnet-RSX Network Generation and Installation Guide

10.2

Node Level Hardware Loopback Circuit Testing

This test verifies that the Routing layer, the device drivers, and the communications
hardware are functioning properly. In addition to the layers previously tested by the
local node software tests (Section 10.1), the Data Link and Physical Link layers and associated hardware are tested in this procedure.
This test transfers data over the communications circuit through the loopback connector to the associated loop node name. Figure 10-3 shows the path the test data travels.
As illustrated in Figure 10-3, the test can be run to the controller, to a loopback connector, or to the local modem.
The following devices cannot perform the hardware loopback tests described in this
section. You should observe the special instructions given for these devices.
•

For UNA and QNA devices, proceed to Section 10.3.

•

For DLM, run the PSI checkout procedures first (see Chapter 11), including NCP
LOOP LINE; then proceed to Section 10.3.

•

For multipoint devices (controller or tributary) using software DDCMP, proceed to
Section 10.3.

The following devices must be used in a special manner . You should observe the special instructions given before performing the tests described in this section.
•

For PCL, PCL-n.O is assigned as the local TDM address. This circuit should be used
for the tests specified in this section.

•

For half duplex devices (point to point), temporarily set up for full duplex operation to perform the tests described in this section. If the tests are successful, switch
back to half duplex and proceed to Section 10.3.

10.2.1

Preparing to Run the Test

You should perform four steps before running the test:
Step 1:

Perform one of these three actions (for more information, see the chapter entitled
"Testing the Network" in the DECnet-RSX Network Management Concepts and Procedures manual):
•

Install the hardware loopback connector using the correct connector.

•

Use the analog loopback button on the modem.

•

Set the device (DMC/DMR, or DMP) to controller loopback with the following NCP
commands:

NCP>SET LINE DMC-O
NCP>SET LINE DMC-O CONTROLLER LOOPBACK

Installation Testing

10-13

Step 2:

Use the following NCP commands to set the line and circuit on (assuming that the line is
not already loaded). For example, enter
NCP>SET LINE DMC-O DUPLEX FULL
NCP>SET CIRCUIT DMC-O STATE ON
The event logger prints a message on the console if logging is enabled. A sample message is shown below.
Event type 4.10, Ci rcuit up
Occurred 01-FEB-85 09:29:17 on node 14 (NODEA)
C i r CUI t DMC-O
Node address = 14 (NODEA)
Step 3:

Use the NCP SHOW command to verify that the circuit state has changed to ON. Note
that the executor node is now also the adjacent node on the test circuit.
NCP>SHOW CIRCUIT DMC-O
A sample of the summary displayed is shown below:
Ci rcuit summary as of 01-FEB-85 09:42:56
C i r cui t = DMC-O
State = On
Adjacent node = 14 (NODEA)
If the state of the circuit is still ON-STARTING, proceed to Section 10.2.4 (On-Starting

Substate Error Handling).

Step 4
Zero all system, executor, and circuit counters using the NCP ZERO SYSTEM COUNTERS, ZERO EXECUTOR COUNTERS, and ZERO CIRCUIT COUNTERS commands.

10.2.2

Running the Loop Node Test

Use the SET NODE command with the CIRCUIT parameter to establish a loop node
name. For example,
NCP>SET NODE TESTER CIRCUIT DMC-O
establishes circuit DMC-O as the circuit over which loop testing takes place. The loop
node name is necessary because, under normal operation, DECnet routing software
decides what path to use. The loop node name overrides the routing function to route
information over a specific circuit. For RSX-IIS systems you must establish this loop
node using the VNP SET NODE command before loading the system image.

10-14

DECnet-RSX Network Generation and Installation Guide

Next, use the NCP command to loop messages over the line through the loopback
device to the loopback node name TESTER.
NCP>LOOP NODE TESTER COUNT 10

Now show the circuit counters by entering:
NCP>SHOW CIRCUIT DMC-O COUNTERS

A sample of the counters displayed is shown below:
Circuit counters as of 01-FEB-85 09:54:15
C i r cui t

DMC-O

226 Seconds since last zeroed
87 Terminating packets received
87 Originating packets sent
o Transit packets received
o Transit packets sent
o Transit congestion loss
o Circuit down
o Initialization failure
4378 Bytes received
4378 Bytes sent
143 Data blocks received
143 Data blocks sent
o Data errors inbound
o Data errors outbound
o Remote reply timeouts
o Local reply timeouts
o Local buffer errors

The last five counters are error counters and should be zero. If any of the last five counters shows a nonzero value, see Section 10.2.5; otherwise enter the following command to show system counters:
NCP>SHOW SYSTEM COUNTERS

A sample of the counters displayed is shown below:
System counters as of 01-FEB-85 09:56:42
1860 Seconds since last zeroed
o Control buffer allocation fai led
o Sma I I buffer al location fai led
o Large buffer allocat ion fai led
o Receive buffer al location fai led

Errors indicated by the system counter summary can be corrected by increasing the
number of buffers whose allocation failed. Use CFE to modify system buffer parameters.
Next, enter the command to show the executor node counters:
NCP>SHOW EXECUTOR COUNTERS

Installation Testing

10-15

A sample of the counters displayed (for a routing node) is shown below:
Node counters as of 01-FEB-85 09:56:23
Executor node = 14 (NODEA)
1862 Seconds since last zeroed
1011764 Bytes received
1011764 Bytes sent
17745 Messages received
17745 Messages sent
9 Connects received
9 Connects sent
o Response timeouts
o Received connect resource errors
2 Total maximum logical I inks active
o Aged packet loss
o Node unreachable packet loss
o Node out-of-range packet loss
o Oversized packet loss
o Packet format error
o Pa rt i a I rout i ng update loss
o Verification reject
2 Node maximum logical links active
o Tot a I rece i ved connect resou rce e r ro rs

Check the executor node counters and the circuit counters. If the counters were properly zeroed before the test, then the counters should show equivalent values for the following counter pairs:
Circuit Counter Pairs

Node Counter Pairs

Terminating packets received
Originating packets sent

Bytes received
Bytes sent

Transit packets received
Transit packets sent

Messages received
Messages sent

Bytes received
Bytes sent

Connects received
Connects sent

Data Blocks received
Data Blocks sent

If counter pairs show equal values, proceed to restore your node as described in the
next section. If you see unequal values for counter pairs, there is probably a hardware

problem; follow the steps described in Section 10.2.5 for line and circuit error handling.

10.2.3

Restoring the Node

Now that the test is complete, the circuit must be restored to its normal state. This
involves removing the loopback name and restoring the hardware to its normal state
(removing the hardware loopback connector, setting the modem to normal, or setting
the line to controller normal from controller loopback).

10-16

DECnet-RSX Network Generation and Installation Guide

Use the NCP CLEAR NODE command to remove the loopback name. For RSX-IIS systems, the loopback name must be left set or removed from the system image file with
the VNP CLEAR NODE command. The loopback name is TESTER so enter:

NCP>CLEAR NODE TESTER CIRCUIT
Next, remove the hardware loopback connector or set the modem back to normal.
Note that the circuit status returns to the ON-STARTING state. The time it takes to
return to the ON-STARTING state depends on the listener time for the circuit. The
default value specified in NETGEN is 30 seconds. A sample of the event message displayed is shown below:

Event type 4.7, Ci rcuit down - ci rcuit fault
Occurred 01-FEB-85 10:05:16 on node 14 (NODEA)
C i r cui t DMC-O
Adjacent node I istener receive timeout
If a DMC/DMR, DMP, or DMV device was previously set to controller loopback, you

must set the device back to controller normal (the default state) with the following commands:

NCP>SET CIRCUIT DMC-O STATE OFF
NCP>SET LINE DMC-O CONTROLLER NORMAL
NCP>SET CIRCUIT DMC-O STATE ON
The local node level communications circuit testing is now complete. You can be confident that the local software and hardware are functioning properly and can connect to
a remote node. Proceed to Section 10.3.

10.2.4

On-Starting Substate Error Handling

The test cannot be run if the circuit state is ON-STARTING. Examine the hardware
loopback connector. Be sure it is properly installed and conforms to the specifications
described in the DECnet-RSX Network Management Concepts and Procedures manual. If you are using a modem with the analog loopback capability, see if special strapping is required.
Next, use the NCP SHOW LINE line-id CHARACTERISTICS command (use VNP for
RSX-IIS nodes) to check that the line has been set to full duplex mode (even though
you may have selected half duplex during NETGEN). Half duplex mode may have been
specified in the network generation procedure. Use NCP commands (VNP commands
for RSX-IIS nodes) to change the controller. For example, the following NCP commands set up a DMC/DMR line for testing using the controller loopback capability of
the DMC/DMR device:

NCP>SET CIRCUIT DMC-O STATE'OFF
NCP>CLEAR LINE DMC-O ALL
NCP>SET LINE DMC-O DUPLEX FULL
NCP>SET LINE DMC-O CONTROLLER LOOPBACK
NCP>SET CIRCUIT DMC-O STATE ON

Installation Testing

10-17

If the circuit could not be set ON, check

•

For resource allocation problems. NCP > SHOW SYSTEM COUNTERS should show
no errors.

•

That CSRs and vectors were properly specified during NETGEN. If they were not,
correct them by means of NCP or CFE.

•

If a device was included in the RSX system during SYSGEN and also included at
NETGEN. If it was, the following message will appear when loading the network:

NTL-Vector not avai lable
If this condition occurs, you need to change the hardware vector and CSR of the

device, as well as modify the network permanent database, (or re-SYSGEN). Otherwise check with field service.
•

If you specified too low a value for the maximum vector parameter during

SYSGEN, the following message will appear when loading the network:
NTL-Vector not in system
If this condition occurs, you need to change the hardware vector and CSR of the

device, as well as modify the network permanent database, (or re-SYSGEN). Otherwise check with field service.
See Chapter 3 for detailed SYSGEN requirements. If the circuit is still in the 0 N-START ING state, use the NCP SHOW CIRCUIT COUNTERS command. Check the summary. If
bytes received and bytes sent are zero, the communications hardware is at fault. Have
Digital field service verify that the device and connector have been properly installed.
Three common hardware failures that occur are:
•

The communication device fails to interrupt.

•

The modem and/or line corrupts the data being transmitted or received.

•

Communication device addresses (CSR and vector) are improperly set on the interface.

Common symptoms of this type of failure are:
•

Inability to turn on a circuit

•

Inability to complete the transport initialization operation

Most other types of failures are documented by appropriate error messages. You
should take advantage of the extensive status and error counters maintained by the
DECnet-RSX system to aid in trouble shooting. Refer to the DECnet-RSX Network
Management Concepts and Procedures manual for descriptions of the various counters and procedures for displaying them.

10-18

DECnet-RSX Network Generation and Installation Guide

10.2.5

Line and Circuit Counter Error Handling

The following errors indicate problems requiring Digital field service:
•

Errors indicated by circuit counters with receive overrun error qualifiers can indicate that the communication device is installed at the wrong device priority. For
example, a bus request plug may not be installed at BR5.

•

Data errors inbound or outbound indicate a hardware device, modem, or line problem. If possible, verify the line quality by using standard modem test procedures
(see the owner's documentation for your modem) before calling Digital field service.

Installation Testing

10-19

NCP
SET UP
COMMANDS

NONE

USER
LAVER

0
0
E
B

NETWORK
MANAGEMENT LAVER
NETWORK
APPLICATION LAVER
NSP/SESSION
CONTROL LAVER
TRANSPORT"
LAVER
DATA LINK
LAVER
PHVSICAL LINK
LAVER

I
COMMUN ICA TlON
HARDWARE

I
I

I
I
I
I

I
I

I
PHVSICAL LINK
LAVER
DATA LII~K
LAYER

0
0
E
T
R
N

TRANSPORT"
LAVER
NSP/SESSION
CONTROL LAVER
NETWORK
APPLICATION LAYER
NETWORK
MANAGEMENT LAVER

USER
LAVER

COMMENTS

Same as test In
F,gure 10-2. except
PROGRAM A and
PROGRAM B resIde
In seperate nodes.

Legend

0--<:>

User program or DECnet mo:Jule
Path travelled by test data
The Transport layer uses ItS
routIng algorothm to determine
on whIch path to loop the test
data.

Figure 10-4:

10-20

Remote Node Software Test

DECnet-RSX Network Generation and Installation Guide

10.3 Remote Node Software Testing
In this test, all the network layers are tested. Figure 10-4 depicts the path taken
through the various layers. This test involves testing the software in your node against
software in a remote node. Both nodes should have been verified locally before the
tests described in this section are performed.

10.3.1

Preparing to Run the Test

You should perform these three steps before running the tests described in this section:

Step 1:
Verify that the circuit to be used for the test has been connected to a properly configured Phase III or Phase IV remote node. Verify that both the local and remote nodes
are set up with the same line and circuit characteristics.

Step 2:
Use the NCP SHOW KNOWN CIRCUITS command. If your circuit is in the ON-START1NG substate, proceed to Section 10.3.3 for non-Ethernet devices and Section 10.3.4
for Ethernet devices. Follow the procedures in those sections to get the circuit state to
ON. If the circuit state is ON, proceed to the following step. For Ethernet devices, you
can perform the test to any adjacency (remote node on the Ethernet) that is reachable.
Use the NCP SHOW NODE nodeid command to determine if an adjacency is reachable.

Step 3:
Zero all the counters used in previous tests by entering the NCP ZERO COUNTERS commands, including system, executor, and circuit. If you are using the NTEST. CMD procedure to perform the test, the procedure will zero the counters for you.
NOTE

For the remote tests described in this section, both DTR and FAL must be
installed on the remote node before you can run NTEST. CMD.

10.3.2 Running the Test
There are two methods of testing. One method uses the NTEST.CMD file (not available
on RSX-llS), the other method uses DECnet utilities. If you are using the DEenet utilities, when testing is completed, display all counters using NCP SHOW commands. Be
sure to include circuit, node, executor, and system counters. If you are using the
NTEST .CMD procedure, the procedure will display the counters for you. Pairs of counters will no longer match on the local node, but errors should be noted and corrected.
Refer to the DECnet-RSX Network Management Concepts and Procedures manual for
information on correcting errors indicated by the network counters.

Installation Testing

10-21

10.3.2.1 Using the NTEST.CMD Procedure - This test (not available on RSX-IIS
nodes) is the same as the one described in Section 10. 1.2, with one exception. Instead
of answering the question, "Is this an internal node test?", with a Y, answer with an N.
Then follow the procedures described in that section.
10.3.2.2 DECnet Utility Tests - When all circuits have been verified to be in the ON
state, run one of the DECnet utilities listed in Table 10-1. Utilities are listed in descending order of preference. The DTS/DTR test is more thorough than the others, which is
why it is the preferred test. The table specifies the availability of each test on each of
the DECnet products.
For RSX-IIS systems, the TLK and NCP LOOP tests are the only tests available on the
local node. However, you can use the NFT utility on another node to transfer files to
and from RSX-11 S unit record devices if you have installed FAL in the RSX-11 S system
image using VMR. In addition, you can install DTR on an RSX-IIS system and perform
the DTS/DTR test shown below from another node to the RSX-IIS system.

Table 10-1:

Utility Availability on DECnet Nodes

TEST

REMOTE NODE TYPE
DECnet-RSX DECnet-20

DECnet-RT DECnet-E DEC net-VAX DECnet-IAS

DTS/DTR

Yes

Yes*

Yes

NFT/FAL

Yes

TLK/LSN

Yes

PHO

Yes* *

Yes

NCPLOOP

Yes

* Available from software specialist
* * RSX-IIM-PLUS and Micro/RSX only

DTS/DTR. In order for the DTS/DTR utility to run, DTR must be installed at the remote
node. The general format ofDTS commands is shown below, followed by an example.
Full details on DTS/DTR are given in Appendix D.
Format:

> DTS

[nodename[aci]]::[/TEST = type][/PRINT = choice]

(aci is optional access control information.)

Example:

>DTS REMNOD: :/TEST=DATA

10-22

DECnet-RSX Network Generation and Installation Guide

NFT/FAL. In order for NFT to perform a file operation to a remote node, FAL must be
installed at the remote node.
Use NFT to transfer a copy of the LB:[1,2]STARTUP.CMD file to the remote node and
then back to the local node. Then, use PIP to output the file and determine whether the
file was transmitted and received correctly. (You can use the CMP utility to compare
the two versions of the file.) An example is shown below.

>NFT
NFT>REMNOD/IO,lO/PASWD: :=LB: [1,2]STARTUP.CMD
NFT>=REMNOD/IO,lO/PASWD: : STARTUP. CMD
<CTRL/Z>
>PIP TI:=STARTUP.CMD
Note that the file is transferred from the local node's UIC[1,2] to the remote node's
UIC[10, 10]. The file is then returned to the local node's UIC. Access control information is given in the example, but can be defaulted if an alias node name is specified.
For more information on NFT, refer to the DECnet-RSX Guide to User Utilities. There
may be special file considerations, if the remote node is not RSX, as the syntax of the
specified file can differ from that expected by NFT. (Check the appropriate system documentation for details.)
TLK/LSN. In order for the TLK/LSN utility to run, LSN must be installed at the remote
node, and an operator must be present to respond to the message. To use TLK on an
RSX-l1 S you must have installed TLK using VMR. The command syntax for single message mode is shown below, followed by an actual example.
Format:

> TLK [targetnodespec:: ][TTn:], message
(TTn: defaults to the console device.)
Example:

>TLK REMNOD: : IF THIS MESSAGE IS RECEIVED, WE HAVE SUCCEEDED.
For information on the use of TLK on RSX systems, refer to the DECnet-RSX Guide to
User Utilities. When using TLK on other systems, check the appropriate system documentation.
PHO. In order for the phone communications utility to run, the phone utility must be
installed on the remote node. Phone is only available on RSX-IIM-PLUS and Micro/
RSX systems and only at video terminals.
Use PHO to display a list of phone users on a remote system. For example,

PHO DIRECTORY REMNOD
displays a list of phone users at the node REM NOD .

Installation Testing

10-23

NCP LOOP Command. In order for the NCP LOOP Command to work, the loopback
mirror task (MIR) must be installed on the remote system. A complete description of the
use of the NCP LOOP command is given in the DECnet-RSX Network Management
Concepts and Procedures manual. An example of its use is shown below.

NCP>LOOP NODE REMNOD COUNT 100 LENGTH 200

The example causes a message 200 bytes in length to be looped 100 times.

10.3.3 Non-Ethernet Error Handling
The circuit state must be ON in order to run remote tests. If the circuit has a substate of
STARTING, the remote operator should make sure that the remote node's corresponding circuit also has been set to the ON state. If the STARTING substate persists upon
reexecution of the NCP SHOW KNOWN CIRCUITS command, use the NCP SHOW
EXECUTOR CHARACTERISTICS command (shown below) to see whether transmit
and receive passwords have been set on the local system. If passwords have been set,
make sure that the passwords are the same on the remote node. An example using the
executor node follows.
NCP>SHOW EXECUTOR CHARACTERISTICS

This command displays the following information (actual display may vary depending
on your node type):
Node characteristics as of 01-FEB-85 13:34:15
Executor node = 4.16 (EOWYN)
Identification = EOWYN DECNET SYSTEM, Management version = 4.0.0
Host = 4.16 (EOWYN), Loop count = 1, Loop length = 40
Loop with = Mixed, Incoming timer = 15, Outgoing timer
10
NSP version = 4.0.0
Maximum I inks = 10, Delay factor = 1, Delay weight = 1
Inactivity timer = 30, Retransmit factor = 5
Routing version = 2.0.0, Type = Routing IV, Routing timer = 300
Broadcast routing timer = 40, Maximum address = 400, Maximum ci rcuits
Maximum cost = 1022
Maximum hops = 10, Maximum visits = 20
Maximum broadcast nonrouters = 32
Maximum broadcast routers = 10
Segment buffer size = 576, Verification state = On
If passwords have been set, they must have been set appropriately on both nodes. If the

passwords are not identical, or if a password was set at one node and not at the other,
the following event messages are generated at the two nodes (provided that the event
logger has been enabled):
•

Message at node requesting password:
Event Type 4.6 Verification Reject
From Node 4.9 (HARTFD) Occurred 05-Feb-85 17:16:21

10-24

DECnet-RSX Network Generation and Installation Guide

•

Messages at node with incorrect or missing transmit password:
Event Type 4.10, Circuit Up
From Node 4.4 (BOSTON) Occurred 05-Feb-85 17:16:55
C i r cui t DMC-1
Event Type 4.7 Circuit Down, Circuit Fault
From Node 4.4 (BOSTON) Occurred 05-Feb-85 17:17:12
C i r cui t DMC-1

If a requesting node does not receive a password or receives an incorrect one, it restarts

the node initialization process.
To ensure that passwords are set correctly, reset them on both the local and remote
(RSX) nodes by using the following commands:
NCP>SET EXECUTOR TRANSMIT PASSWORD password
NCP>SET EXECUTOR RECE I VE PASSWORD password

where password is 1 to 8 ASCII characters.
NOTE

The transmit password on the local node must be identical to the receive password on the remote node. Some non-RSX operating systems may require a full
8-character ASCII password for connection to an RSX operating system. For
VMS nodes, passwords are set on a per circuit basis.
If the circuit is still in the STARTING substate:

•

Check that the hardware and software device line speed settings and duplex mode
are the same on both nodes.

•

On local coaxial cable links, check that the transmit cable is connected to the
receive cable on the remote node.

•

If a null modem cable is being used for EIA asynchronous communication, check to
make sure the cable conforms to the null modem wiring diagram given in the chapter entitled "Testing the Network" in the DECnet-RSX Network Management
Concepts and Procedures manual.

•

If your device is a D L11 type, check the NETG EN specification for that device. If it

was NETGENed as a DLII-E/DLVII-E, that DLll must have all EIA signals including request to send (RTS) and clear to send (CTS).
•

For connecting DMP, DMV or DMR devices to a DMC device, check that the nonDMC device is set to DMC compatibility mode. When connecting DMP, DMV, and
DMR devices to other DMP, DMV, or DMR devices, both devices in the pair must
have DMC compatibility mode enabled or disabled.

Installation Testing

10-25

•

Check the cabling installed in your node against the types listed in the chapter entitled "Testing the Network" in the DECnet-RSX Network Management Concepts
and Procedures manual. If your cable is not listed, the network software will not
function correctly.

•

If your device supports switch settings that supersede software settings, check
these options (for example, the mode switches on DMP and DMV devices).

When the nodes are physically connected and the hardware communications devices
are functioning properly, the circuit counts (bytes sent and bytes received) increment.
If the STARTING substate persists and the circuit count (bytes received) is not
incremented, the hardware is probably at fault.
If DDCMP line communication has been established, the bytes sent and bytes received
counters are incrementing, and line errors are not occurring, check other sources of
possible software incompatibilities (for example, maximum node address or large data
buffer sizes).

Local software resource allocation failures can cause line synchronization to be lost.
Use the SHOW SYSTEM COUNTERS command to see if allocation failures have
occurred. Then, increase the number of buffers that have allocation failures.
It is important to note that the DECnet local node software and hardware verification
procedures must have been followed on both nodes (see Sections 10.1 and 10.2).
Events logged on the console help locate the problem (provided console logging has
been enabled).
When all lines have been verified to be in an ON state, return to Section 10.3.2 and run
the test.

10.3.4 Ethernet Error Handling
The circuit state must be ON and the adjacency (the remote node on the Ethernet) must
be reachable in order to run remote tests. If the adjacency is not reachable, the remote
operator should make sure that the remote node's corresponding circuit has been set to
the ON state.

10.3.4.1 Problems Connecting to the Ethernet - If the circuit is still in the START1NG substate,
•

Check that the CSR and VECTOR set in the software match the CSR and VECTOR
set in the hardware

•

Check that the transceiver cable is connected properly on local cable links

•

Perform one of the circuit-Ievelloopback tests described in the DECnet-RSX Network Management Concepts and Procedures manual.

10-26

DECnet-RSX Network Generation and Installation Guide

When the nodes are physically connected and the hardware communications devices
are functioning properly, the circuit counts (bytes sent and bytes received) increment.
If the STARTING substate persists and the circuit count (bytes received) is not
incremented, the hardware is probably at fault.
Events logged on the console help locate the problem (provided the logging console
has been enabled). The following event confirms when the connection to the Ethernet
has occurred:
Event type 4.10, Circuit up
Occurred 1-FEB-85 11:22:05 on node 4.16 (EOWYN)
C i r cui t UNA-O

10.3.4.2 Problems Connecting to the Adjacency - If communication has been
established, the bytes sent and bytes received counters are incrementing, and line
errors are not occurring, check other sources of possible software incompatibilities
(for example, maximum node address or large data buffer size).
Local software resource allocation failures can cause line synchronization to be lost. If
this occurs, use the SHOW SYSTEM COUNTERS command to see if allocation failures
have occurred. Then, increase the numbers of buffers that have allocation failures.
It is important to note that the DECnet local node software procedures have been fol-

lowed on both nodes (see Section 10. 1). You should also perform the circuit-level loopback tests described in the DECnet-RSX Network Management Concepts and Procedures manual to help isolate hardware problems.
If passwords have been set for your local node as well as for the adjacency that you are

trying to reach, you must verify that the passwords have been set appropriately on both
nodes. See Section 10.3.3 for a discussion involving problems related to passwords.
The following event is notification that the adjacency is reachable:
Event type 4.15, Adjacency up
Occurred 1-FEB-85 11:22:06 on node 4.16 (BOSTON)
Ci rcu i t UNA-O
Adjacent node = 4.19 (HARTFO)

10.4 Summary
Figures 10-5 and 10-6 illustrate and summarize the kinds of software and hardware
tests that are available on a DECnet-RSX node. After performing the tests in this chapter, you have done Test 2 in Figure 10-5 and Tests 1,2,4, and 5 in Figure 10-6. Note
that the tests can be run using network management utilities or software that runs at the
User layer (such as NFT/FAL or DTS/DTR). For more information on running other
node-level and circuit-level tests, see the DECnet-RSX Network Management Concepts and Procedures manual.

Installation Testing

10-27

After performing the tests described in this chapter you have verified your DECnetRSX installation. For more information on monitoring, controlling, and testing of your
DECnet-RSX node, see the DECnet-RSX Network Management Concepts and Procedures manual. For more information about the use of the user utilities available, see the
DECnet-RSX Guide to User Utilities. Information about the use of DECnet-RSX programming interfaces is contained in the DECnet-RSX Programmer's Pocket Guide.

10-28

DECnet-RSX Network Generation and Installation Guide

NCP
SET UP
COMMANDS

TEST 1
NONE

0
0

E
B

0
S

USER
LAYER

TEST 2
TEST 3
SET NODE LOOPER SET NODE LOOPER
CIRCUIT CIr-/d
CIRCUIT CIf-,d

0 0

NETWORK
MANAGEMENT LAYER

TEST 4
NONE

0 0

NETWORK
APPLICATION LAYER
NSP/SESSION
CONTROL LAYER
TRANSPORT
LAYER

DATA LINK
LAYER
PHYSICAL LINK
LAYER

I
COMMUNICATION
HARDWARE

I
I

I
I
I

I
I
I
I

I
I

~
~
4

I
I

I
I
I

I
I

I
I
I

PHYSICAL LINK
LAYER
DATA LINK
LAYER

0
0

E
T
R
N

TRANSPORT
LAYER

'-..-I

NSP/SESSION
CONTROL LAYER

NETWORK
APPLICATION LAYER
NETWORK
MANAGEMENT LAYER

USER
LAYER

COMMENTS

The command loops
data within the local
node.

The command loops
data via loopback
node LOOPER
assOCiated with a
circuit containing a
loopback deVice

The command loops The command loops
data via loopback
data through NODE
node LOOPER. The TRN where the
circuit associated
Network Apphcatlo~
with LOOPER leads layer returns data
to adjacent node
TRN where the
Transport layer
returns the data.

See DEener RSX
Network Management
Concepts and
Procedures

Legend.

S
o

Notes.
The module that accepts command Inputs
The module that loops or receives test data
Path travelled by test data

Test data IS looped Internally by Transport

At controller

At hne loopback connector

At modem

Interface between nodes

Figure 10-5:

Node Level Loopback Tests Using DECnet-supplied Software

Installation Testing

10-29

NCP
SET UP
COMMANDS

TEST 1
NONE

USER
LAYER

e E~

NETWORK
APPLICATION LAYER

NETWORK
MANAGEMENT LAYER

0
5

TEST 2
TEST 3
SET NODE LOOPER SET NODE LOOPER
CIRCUIT CIf·,d
CIRCUIT CIT Id

NSP/SESSION
CONTROL LAYER
TRANSPORT
LAYER

TEST 4
NONE

TEST 5
NONE

DATA LINK
LAYER
PHYSICAL LINK
LAYER

I
COMMUNICATION
HARDWARE

I
I

I
I
I
I

I
I
I

~
~
4

I
I

I
I
I
I

I
I
I
I
I
I

PHYSICAL LINK
LAYER
DATA LINK
LAYER

0
D

E
T

R
N

TRANSPORT
LAYER
NSP/SESSION
CONTROL LAYER

NETWORK
APPLICATION LAYER
NETWORK
MANAGEMENT LAYER

USER
LAYER

COMMENTS

PROGRAM A sends PROGRAM A uses
data to PROGRAM B File Access Routones
whIch returns data to (NFARs) and FIle
PROGRAM A.
Access LIstener (FAL
Target node IS local to store and then
retrieve fIle data.
BOS
LOOPER (the target
node) IS a loopback
node name assocIated
WIth a cIrcuit
connected to
loopback hardware.

Legend:

o
~

Same as Test 1
except data
transmIssIon loops
to NODE LOOPER
through NODE TRN

Same as Test 2.
Same as Test 1.
except PROGRAM A
except the target
node IS TRN and
and PROGRAM B
reSIde on separate
WIthout hardware
loopback connector. nodes.

40)

Notes'
User program (such as DTS/DTR)

1. The data IS looped onternally by Transport

or DECnet module

2. At controller

Path traveled by test data

4 At loopback connector

3. At modem

Interface between nodes

Figure 10-6:

10-30

Node Level Loopback Tests Using User-supplied Software

DECnet-RSX Network Generation and Installation Guide

11
Testing the RSX-11 PSI Installation

This chapter describes the procedure to be observed by users of a combined DECnet/
PSI configuration. The procedure verifies that the RSX-ll PSI system has been generated and installed correctly. The chapter also describes how you can test the X.29 software to ensure that it has been installed correctly. Section 11.3 contains information
on verifying data link mapping (DLM) circuits.

11.1

The Installation Checkout Procedure

The procedure uses two programs: XTR (X.25 test receiver) and SCPXTS (X.25 test
sender). They verify that your PSI system can set up a call and transfer data between
your local DTE and a remote DTE. SCPXTS is provided as a prebuilt task in your distribution kit, and it communicates with XTR running on a remote system.

11.1.1

Preparing to Run SCPXTS

Before you can run SCPXTS, you must have generated and installed your RSX-ll PSI
system. A checklist of the steps you must have performed is given below, and full
details can be found in the RSX-11 PSI Generation Guide:
•

Run SYSGEN (if necessary)

•

RunPREGEN

•

RunNETGEN

•

Install the system

After you have performed these steps, the RSX-ll PSI system is up and running. To
run SCPXTS, your generated system must subscribe to the open user group and must
have at least one switched virtual circuit (SVC).

11-1

11.1.2

Running SCPXTS

To run SCPXTS, use the following format:

> RUN ddu:[uic]SCPXTS
where

ddu:[uic] is the location of SCPXTS after NETGEN (see the RSX-JJ PSI Generation
Guide). SCPXTS asks for the address and subaddress of the remote DTE, as follows:

Input fu I I OTE address :

Enter the 1- to 14-digit DTE address immediately followed by the 0- to 4-digit
subaddress.
The remote DTE must be a machine running XTR under the PSI product. This can be
one of your own machines or any other suitable machine. If you are not using one of
your own machines, enter the address provided by your field software support specialist.
SCPXTS will then run for 5 to 15 minutes, depending upon the configuration of your
system and on any other tasks running at the same time. The procedure tests the installation of the RSX-ll PSI product by setting up a call between SCPXTS and XTR. These
programs transmit and receive 500-byte long data packets continually for three minutes. When the tests stop, the call is cleared. When the tests have completed, SCPXTS
produces a summary. An example of a summary from a normal run follows:
Test Finished
Test Statistics
Number of transmits
180
Number of receives
180
Interrupts completed
0
Interrupts received
0
Interrupts confirmed
0
Resets completed
0
Resets received
0

SCPXTS and XTR can run at different speeds. Therefore, the number of transmits and
receives reported by the SCPXTS summary may not be equal. However, the number of
transmits and receives should always be approximately equal and should exceed 170. A
summary showing fewer than 170 transmits and receives does not mean that the test
procedure failed, but it could imply PSDN problems such as noisy lines or network congestion.

11 ~ 1.3

Error Messages

If your PSI system is not working correctly, SCPXTS prints error messages in the follow-

ingformat:
% ICP-E-Errorx occurred

where x is an error number. The following error numbers can occur:

11-2

DECnet-RSX Network Generation and Installation Guide

Number

Meaning
Error in address/user group
Information message. The address of the remote DTE was not specified correctly.

Could not send set-up message
Fatal error. A data transmission error occurred, or SCPXTS failed to reach XTR.

Set-up message was not same when polled
Fatal error. A data transmission error occurred, or SCPXTS failed to reach XTR.

Could not assign network device
This error can occur ifXTR is not set up correctly. Try the test again in case the network was the cause of the error, and if the error recurs, ask your software support
specialist to make sure that XTR is fully operational.

Received diagnostics code not as expected
Warning message. The consistency check on a clear or reset diagnostics code failed.
This means the clear or reset was generated by the network.

Received cause code not as expected
Warning message. The consistency check on a clear or reset cause code failed. This
means the clear or reset was generated by the network.

Data received not as expected
Warning message. The consistency check on incoming data failed due to a data transmission error.

Clear data not as expected
Warning message. The consistency check on the clear data failed. This means the
clear was generated by the network.

Failed to set up with XTR
Fatal error. SCPXTS failed to reach XTR.

You are advised to take advantage of the extensive status and error counters maintained by RSX-ll PSI to aid in diagnosing problems (see the RSX-l1 PSI System Manager's Guide). You may also need to test the physical line using loopback tests (see the
RSX-ll PSI System Manager's Guide).

For more details on SCPXTS, see the RSX-l1 PSI Generation Guide.

Testing the RSX-11 PSI Installation

11-3

11.2 Verifying X.29 Installation
To verify that the X.29 terminal driver has been correctly installed, use a portable terminal and acoustic coupler. Perform the following steps:
1.

Dial an X.29 PAD (packet assembly/disassembly) facility.

Log into an account on the local DTE.

Engage the DTE in a simple dialog by using a few standard commands, such as:
PIP/LI
EDI

Log out.

See the RSX-ll PSI Network-specific In/ormation manual for further details on these
steps.

11.3 Verifying Data Link Mapping Installation
After you have established that your local DTE is fully operational by following the procedures described in Section 11. 1, check any D LM circuits that may be in the network
in the following manner:
1.

Ensure that the DTE is ON. Enter the following NCP command:
NCP>SHOW MODULE X25-PROTOCOL DTE dte-address

The system response will show you whether the DTE you are testing for is ON or
OFF. If it is OFF, turn it ON by means of the NCP command:
NCP>SET MODULE X25-PROTOCOL DTE dte-address STATE ON

For PVCs, ensure that the channel numhers, maximum window size, and maximum
packet size defined are the same as that allocated to you by the PSD N. To determine
this information, enter the following CFE command:
CFE>LIST CIRCUIT DLM-x.y CHARACTERISTICS

3. For SVCs, check that the usage parameters for the DLM circuits are consistent on
each node (one side must be outgoing and the other side incoming). Enter the following CFE command for each circuit:
CFE>LIST CIRCUIT DLM-x.y CHARACTERISTICS

The node response will show whether the circuit is incoming or outgoing. If a correction is required, make it by means of the following CFE command for the appropriate circuit:
CFE>DEF I NE C I RCU IT circuit-id USAGE usage

where usage = INCOMING or OUTGOING

11-4

DECnet-RSX Network Generation and Installation Guide

For SVCs you should also verify whether or not the maximum packet size and maximum window size will be negotiated by the software. If the packet size and/or window size is a value other than zero, then either one (or both) of these will be negotiated. The default is zero (which implies no negotiation with the network).
Therefore, if your network does not support negotiations, the values for the packet
size and/or window size must be zero. Use the following commands to specify the
maximum packet and window sizes:
CFE>DEFINE CIRCUIT DLM-x.y MAXIMUM DATA byte-count
CFE>DEFINE CIRCUIT DLM-x.y MAXIMUM WINDOW block-count

For outgoing circuits, the subaddress of the circuit must be within the range specified at the remote node in the PSI portion of NETGEN in response to the question
WHAT IS THE TRANSPORT SUBADDRESS RANGE?

If it is not, correct it by entering the CFE command
CFE>DEFINE CIRCUIT DLM-x.y NUMBER dte-address

The variable dte-address consists of up to 15 digits made up of two parts:

•

The DTE's unique address within the network.

•

Subaddress. The length of dte-address depends on the PSDN to which you are
connecting. The subaddress is the low-order portion of the dte-address.

For incoming circuits, check that the DTE accepting the calls is properly set up to
accept calls from the calling DTE. To determine this information, enter the following CFE command:
LIST CIRCUIT DLM-x.y

If the DTE is improperly set up, change the information by using the following corp.-

mand (for more information about dte-address, see Step 4):
DEFINE CIRCUIT DLM-x.y NUMBER dte-address

6. In testing SVC circuits, be aware of the possibility of exceeding the maximum number ofre-calls. If this number is exceeded, the circuit will go to the ON-FAIL state.
This would be reported in response to the NCP command:
NCP>SHOW C I RCU IT circuit-id

If the circuit is ON- FAIL state, first turn it OFF and then ON by using the following

NCP commands:
NCP>SET C I RCU IT circuit-id STATE OFF
NCP>SET CIRCUIT ci rcuit-id STATE ON

When the circuit has been set to the ON state, run NTEST.CMD as a remote node
test (see Section 10.3).

Testing the RSX-11 PSI Installation

11-5

A
Sample NETGEN Dialog

The following listing reflects a sample NETGEN for a combined DECnet-llM/PSI system using the UNIBUS. If you do not have a license for the PSI product, the PSI section
of the procedure is not displayed.
>aDL1:[137,101NETGEN

>;
>;
>;

=.a=============================azz=====================a============
NET - RSX-11"/S CEX Syste. Generation Procedure
Started at 15:53:01 on 01-"AY-85

=====================================================================

>; Digital Equip.ent Corporation, "aynard, "ass.

>;
>;

>; The following CEX Products .ay be generated
>;
DECnet
>;
PSI

>;
If
>; The distribution kits .ust be up to the current patch level.
>; necessary, you should stop the generation and apply the necessary
>; patches.
>;
>; The logical device na.e LB: should already be assigned to the device
>; which contains the library files for the syste. you are generating.
>; These files are EXELIB.OLB, EXE"C."L&, RSX"AC.S"L, SYSLIB.OLS, and,

>. for RSX-11S, 11SLIB.OLB.

>; The logical device na.es IN:, OU:, lS: and "P: should not be assigned
>; to any devices before starting the generation. These logical device
>; na.es are used by NETGEN to refer to various disks used during the
>; generation.

>.
>;
>. <EOS>
>*

>;
>;

Do you want to:
<RET>-Continue, E-Exit (S]:

(continued on next page)

A-1

>; ======================================================:====:a:=======
>;
NET - Section 1 - General Initialization
>; =====================================================================

>*
>*
>*
>*

01.00 00 you want to see the NElGEN noteslcautions [O=N]? [YIN]:
02.00 Target syste. device tdduu, O=SYOO:] [5]: OLD:
03.00 Listing/.ap device [dduu, O=None] [5]: OLO:
04.00 UIC Group Code for NETGEN output [0 R:1-377 O:Sl: 110

>;
>; Checking for required library files, tasks, and UICs.
>;
>;
>; Creating UFD for OLOO:C11J,0011
>; Creating UFO for OLOO:[110,010]

>;
>; 07.00 User 10 to be used to identify your new responses
[O=None] [S R:O.-30.): OECNET/PSI
>*
>* OB.OO Is this generation to be a dry run [O=N]? (YIN]:
>* 09.00 00 you want a standard function network [O=N]? [YIN]:
>* 10.00 Should all components be generated (D=N]? [YIN]: Y
>* 11.00 Should old files be deleted CO=N]? [YIN):

>;
>; <EOS>
>*

Do you want to:
<RET>-Continue, R-Repeat section,

P-Paus~,

e-exit [S]:

>;
>i ========================================================:=::a=:=:=:::
>;
NET - Section 2 - Define the target syste.
>; ==================================================a==================

>* 02.00 R5XMC.MAC location (ddu:[9,.1, D=OLOO:[011,010]) [5]:

>; DLOO:(011,010)RSXMC.MAC is being copied to OLOO:[110,010]RSXMC.MAC

>;
>; OLOO:C110,010)RSXMC.MAC is being scanned to define your target syste ••
>; This may take uP to several .inutes.

>; The target is an RSX-11M syste., with •••
>;
A 20K-Execut;ve
>;
Multi-user protection
>;
Extended instruction set (EIS)
>;
Powerfail recovery support
>;
Dyna.ic task checkpoint allocation
>;
PLAS support
>;
Queue Manager (QMG) support
>i
>; Checking for required library files and tasks.

>;
>;

>; Creating UFO for DLOO:[110,024]
>; Creating UfO for OLOO:[110,034]
>; Creating UfO for OLOO:[110,054)
>i

>* 04.00 RSX11M.STB location (ddu:[g,.], D=OLOO:[001,054]) [S]:
>; DLOO:[001,054)RSX11M.STB is being copied to OLOO:[110,054]RSX11M.'TB

>* 06.00 Should tasks link to the Memory Resi~ent FCS library (D=N]? ("N]:
>* 07.01 Generate the DECnet product [D=N]? [YIN]: Y

>; Routing or non-routing nodes can be generated fro. your distribution kit.

>* 07.02 Generate the PSI prOduct [O=N]? [YIN):

>; <EOS>
>*

Do you want to:
<~ET>-Continue,

R-Repeat section, P-Pause, E-Exit [S):

(continued on next page)

A-2

DECnet-RSX Network Generation and Installation Guide

>; =====================================================================
>; NET - Section 3 - Oefine the system lines
>; =====================================================================

>* 01.00 Oevice Driver Process name «RET>=Oonel (S

R:O-~l:

>; The Device Driver ~odule (OD~) Process name identifies a process
>; which controls an 1/0 device. Enter a carriage return when there are
>; no more devices to support.

>; The legal DECnet device driver process names for UNIBUS orocessors:

>;
>; Name Physical device
>;
>; O"'C OMC11 or 0"'~11
>; DMP OMP11
>; OU
DU11
>; DUP DUP11
>; DY
DY11
>; KDP I(MC11/0UP11
>; UNA DEUNA
>; DLM PSI Oata link .aDPing
>;

---------------

Nalle
Dl
OZ
DHU
KDI
PCl

Physical device

--------------0L11
OZ11
OHU11
KMC11/DI11
PCL11

>; The legal OECnet device driver process names for Q-BUS processors:
>;
>; Na.e Physical device
Physical device
Name
>;
----------------------------Dly
>; DMY OMY11
OlV11
>; DPY OPV11
DZY
OZV11 or OZQ11
>; DUV DUV11
OHV
OHY11
>; QNA OEQNA
>; OLM PSI Data link .apping
>;
>;
>; The legal PSI device driver process nalles are:

>; Nail.

Physical device

>; KIllY
>; SOY

OUP11
I("S11-130/8E
KfIIIS11-PXI PY
DPV11 (Q-Bus device,e.g.

>;
>; SOP
>; KMX
>;

11/2~,11/2~b)

>* 01.00 Device Driver Process nail. «RET>=Oone1 (5 R:0-31: OfillC
02.00 How .any OMC controllers are there [0 R:1.-16. 0:1.1: 2
>* 02.01 Are all OMCs installed at the sa.e device priority [0=N1? [YIN]: Y
>* 02.02 What is the DMC device priority [0 R:4-6 0:51:
>* 02.03 Will all the DIIIIC lines run at the same line speed (O=N]? [YIN]: N
>* 02.05 Will all OMC lines run as FULL duplex lines [D=H]? [Y/H): Y
>*

>* 03.01 CSR address for OflllC-O [0 P:160000-177777 0:177777]: 160100
>* 03.02 Vector address for OflllC-O [0 R:O-774 0:0]: ~20

>* 04.01 line speed for OflllC-O [0 0:50.-56000. 0:56000.J:
>* 04.07 Set the state for OMC-O ON when loading the network (O=N)? (YIN):
>; The line owner is OECnet

>;
>;

>* 03.01 CS~ address for D~C-1 [0 R:160000-177777 0:160110]:
>* 03.02 Vector adOress for OflllC-1 [0 R:O-774 0:330]:

>* 04.01 line speed for 0"C-1 (0 A:50.-S6000. 0:56000.): 19200
>* 04.07 Set the state for O~C-1 ON when loading the network (O=N]? (YIN): Y

>;
>; The line owner ;s OECnet
>;

(continued on next page)

Sample NETGEN Dialog

A-3

>i
>*
>*
>i
>*
>*
>~

01.JO Device Driver Process name [<~ET>=oone) [5 P:0-3]: CUP
02.00 HOw many DUP controllers are there (0 R:1.-16. 0:1.):
03.01 CSR address for OU?-O (0 R:16000Q-177777 0:177777):
03.02 Yector aodress for Ouo-O (0 P:O-774 0:0]:
03.03 Device priority for OUP-O (0 R:4-~ 0:5):

>* 04.00 Is DU?-Q a FULL or HALf
>* 04.01 Line speed for ~UP-O (0

~uplex

line [D=FULL) (S):
0:9600.1:

~:50.-960Q.

>i The line owner is DECnet
>* 05.00 Is DUP-O multipoint [D=H)? [YIN]:

>* 01.00 Device Driver Process name [<RET>=Done] [S R:0-3]: UHA
>* 02.00 How many UNA controllers are there [0 P:1.-16. 0:1.1:

>* Ol.01 CSR address for UNA-a [0 R:160000-177777 0:174510):
>* 03.02 Yector address for UNA-O [0 R:0-774 0:1201:
>* 03.03 Device priority for UHA-O [0 R:4-6 0:5]:

>* 04.07 Set the state for UNA-O ON when loading the network [O=N]? [YIN): Y

>i The line owner is DECnet
>i

>* 01.00 Device Driver Process na.e «RET>=oone] (S R:O-l]: DLM
>* 02.00 How many DLM controllers are there (0 R:1.-16. 0:1.1:

>*
>i
>;
>i
>*
>*
>*
>*
>*
>*
>*
>*
>*
>*
>*
>*
>*
>*
>i
>*
>*
>*
>i
>*
>*
>i
>i

>;
>;

04.07 Set the state for OLM-O ON when loading the network [O=N]? [YIN):
The line owner is OECnet
05.07
05.08
05.09
05.10
05.12
05.09
05.10
05.11
05.09
05.10
05.12
05.09

How many tributaries should DLM-O support [0 R:1.-12. D:1.]: 4
Does OLM-O define only Switched Virtual Circuits (SVC) [O=N1? [YIN]: N
Ooes OLM-O.O define a Switched Yirtual Circuit (SYC) [D=N1? [YIN]: Y
will DLM-O.O be used for incoming calls (O=N)? (YIN]: Y
Address of OTE that may call OLM-O.O [O=None] [S R:0.-16.]:
Does OLM-0.1 define a Switched Virtual Circuit (SVC) [D=N1? [YIN]: Y
Will DLM-O.1 be used for ;nco.ing calls [D=N)? (YIN]:
DTE Address which DLM-O.1 calls [O=None) [S R:0.-16.): 31108a88888804
Does OLM-0.2 define a Switched Virtual Circuit (SVC) [D=N]? [YIN): Y
Will DLM-0.2 be used for incoming calls [O=N]? (Y/N1: Y
Address of OTE that may call OLM-0.2 [O=None] [S R:O.-16.]: 31108888838804
Does DLM-0.3 define a Switched Virtual Circuit (SYC) [D=N]? [YIH):

01.00 Device Driver Process name «RET>=Done) (S P:0-3]: SDP
02.00 How many SOP controllers are there [0 R:1.-16. 0:1.]:
03.01 CSR address for SDP-O [0 R:160000-177777 0:177777]: 160050
03.02 Vector address for SO?-O [0 R:O-774 0:0]: 310
03.03 Dev;ce priority for Sop-o [0 R:4-6 0:5]:
04.05 Will SDP-O use mode. signal .onitoring [O=N]? [YIN]: Y
04.07 Set the state for SDP-O ON when lOddinq the network (D=N1? (YIN]:
The line owner is PSI

>* 01.00 Device Driver Process name «RET>=Oone] [S R:0-3]:
>i
>i
>; <EOS> Do you want to:
)*
<RET>-Continue, R-PeDeat section, P-Pause, E-Exit (S]:

>;
>;

(continued on next page)

A-4

DECnet-RSX Network Generation and Installation Guide

>;
>;

=====================================================================
NeT - Section 4 - Define the efX System
====:================================================================

>; Creating build files for CEX, the co •• unications executive.

>* 01.00 Base address for partition CEXPAR [0 R:14000-113500 D:113000]:
>* 01.01 Length of partition CEXPAR [0 R:4300-5000 D:5000]:

>;
>; Network events will be logged.
>;

>; Creating build files for AUX, the CEX Auxillar, routine.

>; Creating build files for STCRC, the CRe16 calculation routine.

>* 01.00 Does the target syste. h.ve a K6-11 [D=N]! [YIN]: Y

>; creating build files for CETAB, the CEX configuration tables.

>;
>; Creating build files for the DO"s, the device driver modules.
OMC
DUP

>;
UtU
>;
OUI
>; Creating build files for-OlM, the data link mapping task

SOP
>; Creating build files for SOP, the OUP-11 SOle driver

>;
>;
>;

>; <EOS>
>*

>;
>;

Do you want to:
<RET>-Continue,

~-Aepeat

section, P-Pause, E-Exit [5]:

=============================================================:=======
NET - Section 5 - Define the Co •• Exec Support Components
>; =====================================================================
>;
>;
>;

);
>;
>;
>;
>;
>;
>;
>;

Creating bui l d files for

~TINIT,

the network init ial her.

Creating build files for NTl, the network loader.
Creating build files for PRD, the general microcoae loader.
creating build files for EVLlEVC, the event logger and collector.

>* 05.01 How lIIany Event Buffers sh ould

>;
>;
>; <EOS>

~VL

use (0 R:4.-SC. 0:6.):

00 you want to:

<RET>-Continue, P-Repeat section, P-Pause, E-Exit (S]:

=====================================================================
NET - Section ~ - Define the System ~anagement Utilities
=====================================================================

>;
>;
>;

>; Creating build files for NCP, the network control proqram.
>; Creating build files for NMVACP
>; Volatile Aep an~ driver.

and

N~ORV,

the

Network

"anage.ent

(continued on next page)

Sample NETGEN Dialog

A-5

>i
>i Creating build files for CFE, the configuration file editor.
>i

>; Creating build files for NO~, the network dUMp analyzer.
>;
>; Creating build files for VNP, the virtual network control prograM.
>i

>; Creating build files for LOO, the loop test sender task.
>i
>i Creating build files for EVF, the event file interpreter progra ••

>;
>;

>i <EOS>

00 you want to:

<RET>-Continue, R-Repeat section, P-Pause, E-E.it [5]:

>; ===============================================================azaz••

NET - Section

7 - Define the CEX Products

>; =====================================================:====_===_=az===

>;
>; Questions concerning the foLlowing CEX Products will now be asked.
>;
DECnet
PSI

>;
>;

>; =====================================================================
DEC - DEC net CEX Product Generation Procedure

>;
Starting questions at 15:58:27 on 01-MAY-85
>; =====================================================================

>; Copyright (C> 1981, 1982, 1983, 1985 by
>i Digital EquioMent Corporation, Maynard,

>; Generate the DECnet
>; 5ystetl.

C~X

~ass.

Product for your RSK-11M target

>; =====================================================================
>; DEC - Section 1 - Define the target and reMote nodes
>i =====================================================================

>*
>*
>*
>*
>*
>*

01.00 What is the tar~et node name (S R:0-6J: BOSTON
02.00 What is the target node address (5 R:O.-8.1: 11.16
03.0~ Target node ID (D=Nonel (5 ~:0.-32.l: DECNET/PSI
04.00 Do you want to generate a routing node (D=NJ? [YIN]: Y
04.01 Do you want to ~enerate a LeveL Z routing node (O=N]? (YIN]: N
05.00 Highest node nUMber in this area (D R:16.-1023. D:16.]: 300

>; Extended network support for network command terminaLs and products
layered on DECnet, will be included.

>i
>i

>* 07.00 Remote node natle «RET>=Done] (5 ~:O-6]: LONDON
>* 07.01 What ;s the retlote node address (0=1.1] (5 R:0.-8.]: 1.15
>i

>* 07.00 Remote node name «RET>=OoneJ (S

>; The DECnet MACRO user library and

R:O-~]:

FORTR~N/C080L/3A5IC+2

library

will

tar~et

disk

>i be ;ncluded.

>; The DECnet MACRO user library will
>i as "DLOO:[1,11NETLIB.MLB".

placec

your

(continued on next page)

A-6

DECnet-RSX Network Generation and Installation Guide

)i
)i The DECnet FORTRAH/COBOL/8ASIC+2 object librJry will be placed on your
)i target disk
as "DLOO:(1,1]NETFOR.OLe-.
)i

>i If necessary, please Move the. to

t~e

library

disk

(LS:)

>; target syste., if this is different fro. your target disk (DLOO:).

your

>i
)i

>i <EOS) Do you want to:
<RET>-Continue, R-Repeat section, P-Pause, E-Exit [S]:
>*
>i
>i
>i =====================================================================
)i
)i

DEC - Section

2 - Define the DECnet COMMunications COMponents

=====================================================================

>;
>; Creating build fi les for XPT, the routing process.

>i
>i Creating build fi les for RCP, the routing event processing task.

>i Creating build fi les for ECl, the ne twork services process and driver.

); Creating build files for NETACP, the network services AtP.
>;
>* 05.01 Should HErACP be checkpointable (O=N]? (YlH] :
>i
>i Creating build fi les for OCp, the DDC"P line protocol process.

>;
>; Creating build files for EPM, the Ethernet protocol lIanager process.

>; <EOS>
>*

00 you want to:

<RET>-Continue, R-Reoeat section, P-Pause, E-Exit (S]:

=====================================================================
DEC - Section 3 - Define the OECnet Network ~anage.ent Coaponents
>i =====================================================================
>i
>;

>;
)i
)i

Creating bu i ld files for NI CE, the
server task.

ne t wo rk

inforllation

and

control

>;
>; Creating build fi les for EVR, thE- network event rece; ve r ta sk.
>;
>; Creating bu i l d files for NTO, the node st at e dhplay t dS k.
>;
>; Creating bui ld files for NTOEMO, the node state di splay serVl"r task.
>;
>i Creating build files for LIN, the l; nk watcher task.
>i
>i Creelting build files for PUR, the loop test ",irror ta sk.
>i
)i

creating build files for NYP, the network connect veri fi cat ion task.

>i
)i

>i <EOS>
>*
>i
>i

00 you want to:

<R:T>-Continue,

~-Repeat

section, P-Pause, E-fxit (S]:

(continued on next page)

Sample NETGEN Dialog

A-7

>; =====================================================================
>; DEC - Section 4 - Define the DECnet Satellite support COllponents
>; =====================================================================
>:
>* 02.00 00 you want the Satellite Support Components (D=t4] ?

(y IN]:

>; Creating build files for DLL, the down-line
>: cOllponent.
>:
>; Creating bu i ld fi les for DUM, the up-line

syste~

loader,

host

systetll

dUlllper,

host

>i cOllponent.

): Creating build files
): COllponent.

for

CCR,

the

console carrier requester, host

files

for

HLD,

the

down-l i ne

>;
>: Creating build
>; co.ponent.
>;

task

loader,

host

>* 06.01 Number of incoming connections to support [0 1l:1.-32. D:4.]:

>i To generate the HLD database which describes the tasks
>i load, you may execute the HLDDAT.CMD co~mand procedure

>; after your network generation is co.plete.
>: located as follows.
>:
>;
OLOO:(110,001]HLDOAT.CMD

to down-line
at any ti.e
This co •• and file will be

>;
>i <EOS>
>*

>;
>;

Do you want to:
<AET>-Continue, A-Repeat section, P-Pause, e-e.it (S]:

>; =====================================================================
>;

DEC - Section

S - Define the DECnet File Utilities

>; =====================================================================
>;
>: Creating build files for NFT, the network file access user task.

>; Creating

build

files

for

FTS,

the

network

file

access

spooler

>i user task.

>i Creating build files for FAL, the network file access server.

>. 04.01 Should FAL support RMS file access (O=Y]? (YIN]: Y
>. J4.0~ Should FAL be overlaid (O=N]? (YIN]: y
>* 04.03 00 you want to use the memory resident RMS library CD=N]? [YIN]: N
>* 04.04 Do you want the minimum size RMS [o=y]? [YIN]: Y
>. 04.05 NUIIIDer of incomin1 connections to support [0 R:1.-10. 0:4.]:
>* 04.06 User data ~uffer size [0 R:260.-2048. 0:1024.]:

>;
>; Creating buila files for

MC~,

the

network

comlllanj

file

subllission

>i server.
>i

>i <£OS>
>*

Do you want to:
<qeT>-Continue,

~-Repeat

section, P-Pause, E-exit [S):

(continued on next page)

A-a

DECnet-RSX Network Generation and Installation Guide

=~===================================================================

=====================================================================

DEC - Section

6 - Define the DECnet Terminal and Control Utilities

>; Creating build files for RMT and RMTACP, the
>; task and ACP.

re~ote

network

ter~inal

>* 02.01 "axilllu. nu.ber of si.ultaneous RMT users [D R:1.-15. 0:4.]:
>i
>i Creating build files for HT: and

P~HACP,

the reMote

>; driver and ACP.

network

ter.inal

>* 03.01 Number of inco.ing connections to support [D R:1.-16. D:4.]:
>* 03.02 Number of ~T: units to generate [0 R:4.-64. D:4.l: 6
>i
>i Creating build files for NCT, the network

>i Creating build fi les for RTH,
>i support.
>i
>i Creating build files for TLK, the
>i

the
re~ote

co~.and

network

ter.inal server
cOlllllland ter.inal host

talk user task.

>; Creating build files for LSN, the relllot e talk server task.

>i
>i Creating build files for TCl, the relllote task control server task.
>i

>i <EOS>

Do you want to:
<RET>-Continue, R-Repeat section, P-Pause, E-Exit [s]:

>i
>;

==========:==========================================================

>i
>i

DEC - DECnet CEX Product Generation Procedure
OECnet question/answer section completed at 16:02:09 on 01-MAY-85

>i
>i

>;
>i

=====================================================================
==============================================================
PSI - PSI CEX Product Generation Procedure
Starting questions at 16:02:16 on 01-MAY-85

==============================================================

~ass.

>i Generate the PSI CEX Product f?r your ~SX-11~ target
>; system.
>i
>i ==============================================================

>;
>;

PSI - Section' - Soecify target network characteristics

==============================================================

>* 0'.00 Packet Switching Oata ~etwork name? (SJ: TELENfT
>* 03.00 What is the transport suba1dress range (5 p:O.-9.]: 2-5

>; <£05>
>*

>;
>;

00 you vant to:

<RET>-continue, P-Pepeat

s~ct;on,

P-?ause, E-Exit (5]:

(continued on next page)

Sample NETGEN Dialog

A-9

==============================================================
PSt - S~ction 2 - Specify local OTcs
>; ==============================================================
>;
>;
>i

>;
>i
>*
>*
>*
>*
>i

01.00 What is th~ local OTE addr~ss (S R:O.-15.1: 31103d888888
02.00 Which lin~ 15 associated with this OTE (S R:1.-9.1: SOP-O
03.00 How .any virtual circuits can be active at once? [0 R:1.-255.1: 30
04.00 What chann~ls can be used for outgoing calls [S1: 30-20

>; Specify PVCs for this OTE.

>; Note that any DLM PVCs ~ust also be included.
>;
>; The following OLM circuits hav~ be~n generated:
>; OUI-O.O S
>; OLIII-0.1 S
>; OLf~-O. 2 S
>; OUI-0.3

>*
>*
>*
>*
>*

>;
>;

05.00 Enter
05.00 Enter
05.00 Enter
05.00 Enter
06.00 Enter

a PVC used by
a PVC used by
a PVC us~d by
a PVC used by
a closed user

this OTE (S]: PVC1,3
this Ole (5): PVC2,4
this OTE (S] : DL~-O.3,5
this OlE (S]:
group for ttd s OTE (5] :

>* 01.00 What is the local OTe address (S R:O.-15.1:
>; <EOS>
>*

>;
>;
>;
>;
>;
>;
>;

Do you want to:
<RET>-Continue,

R-R~peat

section, P-Pause, E-Exit [S1:

================================================= ============
?SI - Section 3 - Specify incoming call d~stinat ons
================================================= ============

>; Specify the X25 destinations

>;
>;

the
the
the
the
07.00 What is the
07.01 What is the

inco~;ng call destination (S ~:0.-6.]: EO
l)riorHy (0 R:O.-255. 0:127.]:
oDj ec t-i d [5 R:1.-o.]: ED
re",ot~ OTE
address [$ R:G.-15.1: 311::166666666
subaddress range (5 R:O.-9.1: 20
closed user ;roup nallle (5 P:0.-6.]:
call ",ask (5 P:O.-32.]: 1101
c 11 II value (5 R:1.-32.J: 1101

>* 01.00 Identify an
~* o Z .00 What is the
>* 03.00 Wtut is the
>* 04.00 What is the
>tr OS.OO What ;s the
>* 06.::10 What is the
>* 07.00 "'hat is the
>tr 07.01 What i c.; the

incoming call destination [5 R:O.-~.]: DICK
odority (0 ~:0.-25). 0:127.]:
object-id (S R:1.-6.]: OICK
r e"'o t e OTE address (5 ;):'J.-15.]: 311066666660
suoaddr~ss range [ S Q:O.-~.]: 21
closed user group nallle (5 R:O.-~.]:
call :'IIask (S Q:O.-32.): 1001
cdll vdlue (5 R:1.-32.J: 1001

>*
>*
>*
>*
>*
>*
>*
>*

>;
>;

01.00 Identify an
02.00 Wha t is
03.00 IIIhat is
04.00 What is
05.00 What is
06.00 .hat is

>;
>;

th~

>* 01.JO Identify an incoming call destin~t;on (5 R:O.-6.]:
>* 02.00 What is the oriority ~D R:O.-25S. O:1?7.]:

L~PRY

(continued on next page)

A-10

DECnet-RSX Network Generation and Installation Guide

>* 03.00 What
>* 04.00 ~hat
>* 05.00 What
>* 06.00 What
>* 07.00 What
>* 07.01 What

is
is
is
is
is
is

the object-id (S R:1.-~.J: LARRY
the ~e.ote OTE address [5 P:0.-15.]: 31106666~666
the subaddress range (S ~:0.-9.]: 20
the closed user Q~OUP name [S R:0.-6.]:
the call mask (S"R:0.-32.]: 1001
the call value (5 ~:1.-32.]: 1001

>*
>*
>*
>*
>*
>*
>*
>*

01.00 Identify an
02.00 What is the
03.00 What h the
04.00 What is the
05.00 What is tht>
06.00 What h the
07.00 What is tht>
01.01 What is the

incoming call destination (5 R:0.-6.]: ElLEE~
priority (0 R:0.-255. 0:127.]:
object-id (S R:1.-6.]: EILEEN
~ e",o te
OTE add~ess (S R: o. -1 5 ~ ] : 31106')666666
subaddress range (S R:O.-9.J: 20
closed user q~oup name (S R:O.-6.]:
call ",ask [S R:0.-32.]: 1111
call value [5 R:1.-32.]: 1111

>;
>* 01.00 Identify an incoming call destination

R:O.-6.l:

>; Specify the X29 destinations

>;
>;

>* 01.00 Identify an
>* 02.00 What is the
>* 03.00 What is the
>* 04.00 What h the
>* 05.00 What is the
>* 06.00 What h the
>* 07.00 What is the
>* 07.01 What is the

incoming call destination !:S R:O.-6.]: "'IKE
(0 ~:0.-255. 0:127.]:
ooject-id CS q:1.-6.J: MIKE
(S R:0.-15.]: 311066666666
~ emo t e on address
suoaddress range (S R:O.-9.]: i 24
closed user qroup name (5 ~:O.-6.]:
ca II ",ask (S R:0.-32.J: 1111
call value [S ~:1.-32.]: 1111
p~ior;ty

>* 01.00 Identify an incoming call 1estination CS Q:O.-6.]:
>; <EOS>
>*

>i
>i
>i
>i
>i

Do you want to:
<RET>-Continue, R-Repeat section, P-Pause, E-Exit (S]:

==============================================================
PSI - Section 4 - Specify logical names fo~ remote OT:s
==============================================================

>* 01.00 Identify a re~~te OTE (5~:
>* 01.00 Idt>ntify a ~emote OTe (5]:
>* 01.00 Identify a remote OTE (SJ:

ONE""",11

flfJ'3,22222222

>; <EOS>

00 you want

<~eT>-Conti~ue,

>;
>;
>;

to:
~-Repeat

section, P-Pause, E-Exit [S]:

==============================================================
?SI - Secti~n 5 - 5pecify reQuired comoonents
==============================================================
>;

>i
>i

>i
>i Creating

~uild

f;l~s

for

~~,

build files

fo~

PLI, the X.2S packet level interface

C~~ating

the PSI

~etwork

device

>i Creating build files for LAB, the X.25 level 2 protocol
>i
>i Creating build files fo~ X25ACP, the X.25 serve~ task
>i

>; <EOS>
>*
>i

handle~

00 you want to:
<PET>-Continue, R-Reoeat section, P-Pause, e-Exit (5]:

(continued on next page)

Sample NETGEN Dialog

A-11

==============================================================
PSI - Section 6 - Specify optional components
>; ==============================================================
>;
>;

>* 03.00 How .any re.ote X.29 terMinals dre supported CO R:1.-40.1: 4
>; Creatinq build files for

X29~CP,

the reMote terminal support task

>; Creating build files for TR_, the trace caoture task
>; Creating build files for TRI, the trace interpreter task

>; <£OS>
>*

>;
>;
>;
>;
>;
>;
>;
>;
>;
>;
>;
>;

00 you want to:
<RET>-Continue, R-Repeat section, P-Pause, E-EKit [SJ:

==============================================================
PSI - PSI CEX Product Generation Procedure
_ll questions answered at 16:u3:51 on

01-~AY-85

==============================================================
=====================================================================
NET - Section ~ - Complete the CEX System Definitions
=====================================================================

>; Creatin; buiLd files for OLK, the Direct Line
>* 02.0J What ;s the Large Data
>; <E05>
>*
>i

00 you want to:
<qET>-Continue,

~uffer

R-~eoeat

Acc~ss

process.

(LOB) size CD R:192.-1484. D:518.J:

section, P-Pause, E-EKit [5):

>;
>;

=====================================================================
NET - Section 9 - ~uiL~ the CEX System at 1b:04:1S on )1-~AY-~5
>i =====================================================================
>;
>;

>; All Questions have now been asked and the seLectee components wiLL now
>; be buiLt. This may take fr~m one to three hours, oeoendinq on the
>; selection of comoonents and the system you are running on.

>;
>;

>SET IlJIC=C1,2)
>PIP OU:[1,2J~ETHLP.ULo=IN:(1,2JNETHLPe.HLP
>PIP OU:C1,?)NET~LP.UL~/PU/NM
>5£T IUIC=[11Q,Q24J

>TK9
>T(8

iOU:C110,24)C~XBLO.C~D

>~AC
>TK~

iOU:[110,24JCETAdASM.CMO

~OU:(110,24JSTCqC~LD.C~D
~OU:(110,24JC~T_S'LD.C~D

>TKB iOU:C110,24J~TINIT3LD.C~O
>TK3 iOU:[110,24)NTL3LO.CMD
>TK8 iOU:C1'O,24)MLOBLO.C~D
>~AC

iO~:C11C,24)EVLASM.C~D

>TKB iOU:[1'O,241EVLaLD.C~0
>PIP OU:(110,54)/NV=IN:[131,24)EVL.OAT
>TKB aOu:(110,24)EVC8LO.C~0
>PIP OU:[110,54)/NV/CO=IN:C135,54JNCP.TSK
>~AC

iOU:C110,24)N"VACPAS~.CMD

>TKB iOU:[110,24J~"VACP~LO.CMD
>PIP OU C110,54)/NV/CO=IN C132,54JCFE.TSK
>PIP OU C110,54)/NV/CO=IN C141,54)NOA.TSK
>PIP OU C110,54J/NV/CO=IN (132,54JVNP.TSK

(continued on next page)

A-12

DECnet-RSX Network Generation and Installation Guide

>TK9 iOU:(110,24JLooaLO.C~O
>PIP OU:(110,54]/NV/CO=I~:(135,54]EVF.TS(
>MAC iOU:(11Q,24]DLXASM.CMO
>TKB iOU:(110,24]OLX9LO.C~D
>PIP OU:(110,54]OLX.OAT/~V=!N:(131,24]DlX.~AT

>TK9
>PIP
>TK3
>TKB
>PI 0
>TKa
>PIP
>PIP
>SET
>PIP
>PIP
>SET
>OIP
>SET
>OIP
>SET

~OU:(110,24]OMC8L).CMO

>TK~

iOU:(110,24]XPT3LO.C~)

>PIP

OU:(11J,54]XPT.OAT/NV:I~:(131,!4JXPTDL~.DAT
iOU:(110,24JqCP1~LO.CMO

>TK~

>MAC
>TK?
>PIP
>TK6
>TK3
>PlO
>TKa
>PIP
>TK2
>TK~
>TK~

>TK3

OU:(110,54]/~V=IN:(131,24]O~C.DAT

iOU:C110,24]OOHARBlO.CMO
iOU:(110,24JOUP8LO.C~O
OU:(110,54)/NV=IN:(131,24]OU~.OAT

iOU:(110,24]UNAaLO.CMO
OU:(110,54]/~V=IN:C131,241UNA.OAT
OU:C110,54]/NV/CO=IN:(131,54]UNA~C.TSK

IUIC=(1,1J
OU:(1,1]NETLI9.~L3/NV=IN:(1,10]NETlI5U.~La
OU:(1,1)/NV=I~:(134,24JNETFOR.OL~

IUIC=C110,024]
OU:(11a,024]/~V=IN:(137,24)~TEST.CMD

IUIC=C110,054]
OU:C11C,054]/NV/CO=IN:C133,54JOTS.TSK,OTR.TSK

IUIC=C110,024)

iOU:(110,24]ECLA5M.CM~
iJU:(110,24]:CLdLD.C~D
OU:[110,54]~Cl.DAT/NV=IN:(131,2'JECL.DAT
~OU:(110,24]N£T~co~LO.C~D

iOU:(110,24JOCPdLD.C~O
Ou:(110,i4]/NV=I~:C131,24)DCP.OAT,DC~ST.OAr
~ou:(110,24]~P~3LD.CMD
OU:C110,54]/NV=I~:C131,24)~~~.OAT,EP~ST.DAT
@OU:C110,24J~ICE3LD.CMD

~Ou:(110,24]:V~~LO.C~O
~OU:(110,24]~TO~LD.C~O
;OU:(110,24]~T~~~~3LO.CMO

>TK~

~ou:(110,24]LIN~LD.C~O

>TK3
>TK9
>TK3
>PIP

~OU:(110,24]~I~aLO.C~D

>TK~

~JU:(110,24JNV~~LO.CMO

~ou:(11Q,24JDLL3LD.C~D

Ou:L110,54]/~V/CO=I~:C136,54JSEC*.SYS,TE~*.SYS
~OU:(110,24JOUM8LD.C~~

>TK~

@Ou:(110,24]CC~9LD.CMO

>TK3
>PIP

OU:(110,1J/~V=IN::132,24)~LOOAT.C~D

>S~T

IUIC=(110,1)]

~OU:(11J,24]HLOQLD.CMD

>LBR ou:(110,10]HLOLlq/C~:1:0:64.:~.C=IN:(130,10)HLOTa.~AC
>TK3 iOU:C113,24]~FT~lD.C~O
>SET IUIC=(110,24]
>MAC iOU:(110,24]FTSAS~.CMO
>TK9 .OU:(110,24]FTS2l0.CMO
>SET IUIC=(110,24)
>MAC iOU:(110,24JF.L.S~.CMO
>T(3 iOU:(110,24)fAL3LO.C~D
>TK3 iOU:(110,24JMC~3LD.C~D
>TK3

>~AC

~Ju:(110,24]~~TAS~.C~D
iOU:(110,24JRMTalD.C~D

>~AC

iOU:(110,24J~MHACP.)M.CMD

>TKa
>TK2

iOU:(110,24]~~H.CP3LD.CMO
iOU:(110,24)NCT8LD.C~D

(continued on next page)

Sample NETGEN Dialog

A-13

>PIP

OU:[110,54]~CT.OAT/NV=IN:(140,24)NCT.OAT

>~AC

~OU~(110,24]RTHASM.CMD

>T(B iOU:C110,24]RTHBLO.CMD
>PIP OU:(110,54]RTH.OAT/~V=IN:(140,24JRTH.DAT
>TK3 iOU:(110,24JTlKBLD.CMD
>TK9 ~OU:(110,24JLSNBlD.CMO
>~AC

;OU:[110,24)TCLAS~.CMO

>TKS

~OU:[110,24]TCL9l0.CMO

>i
>TK~

iOU:(110,24JOl~aLO.C~O

>PIP OU:(110,54]/NV=IN:(152,24]OLM.OAT,OL~STA.OAT
>T(8 iOU:(110,24)SDPBLO.CMO
>PIP OU:(110,54]SOP.OAT/~V=IN:(1S2,24]SOP.OAT
>PIP OU:(110,001]/NV=IN:(151,a10]~.pPF
>PIP OU:(110,001]~.CMO/~E/NV=OU:(110,001J*.PRF;*
>~AC

iOU:(110,24]~WTAgASM.CMO

>TK3 iOU:(110,24]NW~LO.C~O
>PIP OU:(110,54J/NV=IN:(152,24]N~.OAT
>TK3 iOU:(110,24]PLtSLO.CMO
>T(3 ~OU:(11a,24JLA~aLO.CM~
>TK3 iOU:(110,24JX25BLO.CMD
>PIP OU:(110,54]/NV/CJ=IN:(153,54JSCPXTS.TS(
>T(9 @OU:(110,24JX29~LO.CMO
>TK? ~OU:(110,24]TRAaLO.CMO
>TK~

~OU:(110,24)TRI3LO.C~O

>SET IUIC=(1,1J
>PIP OlOO:(1,1]/NV=Iti:(1S5,24]PSIFCR.OL~
>SET IUIC=(110,24]
>sn IUIC=[1,1J
>~IP

OLC~:(1,1J/NV=IN:(1,1]PSI.ML~

>S~T

IUIC=(11),24]

>i
>i =====================================================================
>i
NET - Section 1) - Generat;on Clean Up

>; =====================================================================
>i

Creatin~

~qCF'j.

TXT, the net work confi;urat;on de5cr;ot;on f i l !.'.

>i Creating NETCFE.C~!), the network confi-:;uration c ol7'ma nC file.
>i C"t'at;n~ NETINS.CMO, the network installation cOlllman d f; l to.
>i Creat;n~ NET~OI.C""O, the network remcvill command f i l e.
>i

Th~

network HELP

fil~

liorary

~as

~een

placed on your

tar~et

~isk,

as:

OL)O:(1,2]~ETHLP.UL~

);

>i Th;s file 5hould Oe moved to your tarcet systems libr3ry device
>i

(L~:).

>i =====================================================================
>i
NeT - RSX-11"" CEX System Generation Procedure
>i
Stoopea at 17:03:47 on 81-~.y-j5

>; =====================================================================
>;
)i

<EOF>

A-14

DECnet-RSX Network Generation and Installation Guide

B
How to Include NETPAN (Up-line Panic Dump) in an
RSX-11 S System

The NET PAN component of DECnet-RSX is activated when a detectable abnormal system crash occurs. Registers are saved automatically in a predefined place in memory
and a copy of the entire system memory is sent up-line. CDA or NDA can then be used
to analyze the problem.
The remainder of this appendix tells you how to include NETPAN in an RSX-ll S system when you use a distribution kit consisting of an RLO 1 or larger disk pack.

B.1

How to Include NETPAN in an RSX-11S System

The steps that you must take to include NETPAN in a system occur at the following
points in the system generation process:
•

In set-up operations prior to invoking SYSGEN

•

In Phase I of SYSGEN

•

In Phase II of SYSGEN

The operations that you must perform at these points in the system generation process
are broken out in the example that follows.

8-1

8.1.1

Preparing for SYSGEN

Issue the following set-up instructions before you initiate the SYSGEN process:
>SET IUIC=[11,10]
>PIP xx: PANIC. SAV/RE=XX:PANIC. MAC
!SAVE THE NON-NETWORK PANIC DUMP
SOURCE
>PIP xx:PANIC.MAC=yy: [136, 10]NETPAN.MAC !GET THE NETWORK PANIC DUMP SOURCE
>SET IUIC=[1,20]
>LBR xx:ddd.OBJ=yy: [136, 24]BOOTS/EX:ddd
where
ddd

is the name of the physical line used for the up-line dump. The final three
letters of ddd will always be DRV. For example, the physical line name
for a DUP device will be DUPDRV.

is the RSX distribution device

is the DECnet distribution device
!DISMOUNT THE NETWORK DISTRIBUTION DISK

>DMO yy:

8.1.2

SYSGEN Phase I Considerations

In the Executive Options section of SYSGEN, answer the following questions as indicated:
>* 50. DUMP ANALYSIS: A- PANIC B- CRASH (CDA) [S]: A
>* 53. Enter PANIC memory dump device CSR [0 R:160000-177700 D:177564]:nnn~nn
where

nnnnnn

is the CSR address of the physical line used for the up-line dump. This is
the CSR address for the device you specified in ddd.

In the Assemble the Executive section of SYSGEN, you are asked whether you want to
edit any of the Executive files. If you are using an Ethernet device for the up-line dump
you must answer Y to the editing question and edit the Executive source file to add a
line defining the N$$NI symbol. If you have an M9301 ROM set for bootstrapping over
a DECnet device, Unit Number 0, or an M9312 ROM set for bootstrapping over a
DECnet device, Unit Number 0 or 1, answer Y to the editing question and edit the Executive source file to add a line defining the R$$BOT symbol. These two symbols are
mutually exclusive; you should never have both symbols defined in the same Executive
source file. If you are not using an Ethernet device and any of the following conditions
apply, answer N to the edit question:
•

You are not using an M9301 or M9312 ROM.

•

You want to boot the system over a unit number greater than 0 using an M930 1 or
over a unit number greater than 1 using an M9312.

•

You do not want to use the hardware bootstrap.

8-2

DECnet-RSX Network Generation and Installation Guide

The following sequence shows an example of the edit question and the edit sequence.

>* 1. Do you wish to edit any of the Executive fi les? [Y/N]:Y
>;

SGN - Pausing to al low editing of Executive source

AT.

PAUSING. TO CONTINUE TYPE "RES ... AT. "

>EDI RSXMC.MAC
[OOONN LINES READ IN]
[PAGE 0]
*I
Now enter one of the two lines shown below:

N$$NI=O

;UP-LINE DUMP DEVICE IS AN ETHERNET DEVICE

R$$BOT=O

; AUTOMATIC REBOOT THROUGH THE M9312 OR M9301 ROM

Now exit the editor:

*ED [EXIT]
>RES ... AT.
NOTE
If you do not include the symbol R$$BOT = 0 in RSXMC.MAC, the NET PAN

routine attempts a software automatic reboot of the system through the device
associated with the up-line dump CSR address you specified in the Executive
Options section of SYSGEN. If you do include R$$BOT = 0, NET PAN jumps to
the M930 1 or M9312 ROM and allows the ROM to perform the automatic bootstrap operation.

8.1.3

SYSGEN Phase II Considerations

At the beginning of Phase II, before building the Executive, you are asked whether you
want to edit the Executive build file. Answer YES to the question and then edit the file
as shown. If however, you are planning to reboot the system through an M9312 ROM,
see the discussion below.
>* DO YOU WISH TO EDIT THE EXECUTIVE BUILD FILE? [Y/N]:Y
>EDI RSXBLD.CMD
[OOONN LINES READ IN]
[PAGE 1]
*PL NLDRV
[1,20]RSX11S/LB:NLDRV
*1
[1,20] ddd

where
ddd

is the name of the physical line used for the up-line dump. The final three
letters of ddd will always be DRV. For example, the physical line name
for a DUP device will be DUPDRV.

How to Include NETPAN (Up-line Panic Dump) in an RSX-11S System

8-3

If you are planning to reboot the system through an M9312 ROM, you must do a more

extensive edit of the Executive build file as shown:
>* DO YOU WISH TO EDIT THE EXECUTIVE BUILD FILE? [YIN]: Y
>EDI RSXBLD.CMD
[OOONN LINES READ IN]
[PAGE 1]
*PL NLDRV
[1,20]RSXIIS/LB:NLDRV
*1
.
[1,20]ddd

*PL GBLDEF
GBLDEF=$USRTB:O
*I
GBLDEF=REBOOT:xxxxxx
where
xxxxx:x

is one of the reboot addresses listed in Table B-1.

Table 8-1:

Reboot Addresses for the M9312 ROM

Option

Unit Number

CPU Diagnostics

Reboot Address

173004

173030

173204

173230

Notes:
1.

The reboot addresses contained in Table B-1 signify boot entry points that are independent
of the device type used.

The M9312 has four ROM slots. Only three of these slots are needed for a DECnet bootstrap
ROM. Therefore, the M9312 ROM can be inserted into either Slots 1,2, and 3 or Slots 2,3,
and 4. (Option A in Table B-1 means that the ROM is inserted in Slots 1, 2, and 3. Option B
means that the ROM is inserted in Slots 2, 3, and 4.)

When you define a reboot address using GBLDEF, the following TKB diagnostic message
will appear during the Executive build:

*DIAG* - MODULE ddd MULTIPLY DEFINES SYMBOL REBOOT
where
ddd is the name of the physical line that you previously specified for the up-line dump.

Regardless of the type of ROM you use, exit from editing mode in the following manner:
*ED
[EXIT]
RSX-I1S SYSGEN will then proceed to build the Executive with the up-line panic
dump routine included.

8-4

DECnet-RSX Network Generation and Installation Guide

C
Rebuilding CETAB, CEX, and DDMS

This appendix tells you how to rebuild CETAB, CEX, and DDMs. These components
cannot be rebuilt with a component mode generation. Rebuilding these components
may be necessary because of accidental deletion, or because of patches to the objects.

C.1

Logical Device Assignments

Before executing any of the commands described below, you must first make the following logical device assignments:
ASN ddu:=OU:

!The target device (where tasks are to go)

ASN ddu:=LB:

!The system device (where system I ibraries are)

ASN ddu:=MP:

!The device where maps are to go (may be NL:)

ASN ddu:=LS:

!The device where I istings are to go
! (needed for CETAB.MAC assembly only)

ASN ddu:=IN:

!The Network, DECnet or PSI Object
!Disk, created by PREGEN

C.2

Rebuilding Commands

Command lines used to rebuild CETAB, CEX, and DDMs are listed below. These command lines are used in various combinations to rebuild specific components. For example, rebuilding the PCL DDM requires commands b,c, and d. The xxx shown in the command is the name of the individual component (CETAB, CEX, and peL.) The x in the
VIC (such as [x, 24] indicates the group code chosen during NETGEN.

C-1

The sections that follow make reference to the command lines to use to rebuild NET,
DEC, and PSI components.
a)
b)
c)
d)
e)

MAC @[x,24]xxxASM.CMD
TKB @[x,24]xxxBLD.CMD
PIP [x,54]/NV=IN: [131,24]xxx.DAT
PIP [x,54]/NV=IN: [131, 24]xxxST.DAT
PIP [x,54]/NV=IN: [131,24]xxxLN.DAT
f) PIP
[x,54]/NV=IN: [131,54]xxxMOD.TSK
g) PIP [x,54]/NV=IN: [131,54]xxxMC.TSK
h) PIP [x,54]/NV=IN: [152,24]xxx.DAT
i) PIP
[x,54]/NV=IN: [152, 24]xxxLN.DAT
j) PIP
[x,54]/NV=IN: [152,54]xxxMC.TSK
k) PIP [x,54]/NV=IN: [152, 54]xxxTA.DAT
I) PIP
[x,54]/NV=IN: [152, 54]xxxDG.TSK

C.2.1

NET Components

To rebuild the following processes. IN: must be assigned to the Network object disk.
Process Name
(as in xxx above)

Commands to Use
(see list of commands above)

CETAB

a,b

CEX (and AUX)

C.2.2

DEC Components

To rebuild the following device driver modules (DDMs), IN: must be assigned to the
DECnet object disk.
DDMName
(as in xxx above)

Commands to Use
(see list of commands above)

DL,DLV,DMC,QNA,
DPV,DU,DUP,DUV

b,C

DMP ,DMV ,PCL

b,c,d

DZ,DZV,DHU,DHV

b,c,e

UNA

b,c,g

b,c,e,f

KDP

b,c,e,g

KDZ

b,c,e,f,g

C-2

DECnet-RSX Network Generation and Installation Guide

C.2.3 PSI Components
To rebuild the following device driver modules (DDMs), IN: must be assigned to the PSI
object disk.
DDMName
(as in xxx above)

Commands to Use
(see list of commands above)

DLM

b,h,k

SDP,SDV

b,h

KMX,KMY

b,h,i,j,l

Rebuilding CETAB, CEX, and OOMS

C-3

D
DTS/DTR Test Programs

DTS and DTR are the DECnet-llM/M-PLUS transmitter and receiver test programs.

0.1

Types of Tests

Four basic tests are provided by DTS and DTR:
•

Connect test

•

Data test

•

Disconnect test

•

Interrupt test

Each test is divided into a set of subtests. The tests and subtests are described in the following sections.

0.1.1

Connect Tests

Connect tests verify the ability of the network software to process connect, connect
accept, and connect reject requests with and without optional user data. Connect tests
that the user can perform are:
•

Connect reject without user data

•

Connect accept without user data

•

Connect reject with 16 bytes of standard user data

•

Connect accept with 16 bytes of standard user data

0-1

•

Connect reject with received user data used as reject user data

•

Connect accept with received user data used as accept user data

D.1.2 DataTests
Data tests provide a full range of test capabilities from the very simple data sink operation through data integrity checking. Data tests the user can perform are:

•

Sink test. DTR ignores all data received. No sequence or content validation is performed.

•

Sequence test. Data messages transmitted by DTS to DTR include a 4-byte
sequence number. If a message is received out of sequence, DTR aborts the logical
link and the test.

•

Pattern test. Data messages transmitted to DTR have both a sequence number and
a standard data pattern. If either the sequence number or the received data does not
match the expected data, DTR aborts the logical link and the test.

•

Echo test. Data messages received by DTR are transmitted back to DTS. There is
no sequence or data validity checking done by either DTR or DTS.

D.1.3 Disconnect Tests
Disconnect tests are designed to determine whether DTS can detect the difference
between disconnect and abort sequences generated by DTR as well as receive the
proper optional user data. Disconnect tests that can be performed by the user are:
•

Disconnect without data

•

Abort without user data

•

Disconnect with 16 bytes of standard user data

•

Abort with 16 bytes of standard user data

•

Disconnect with received connect user data used as disconnect user data

•

Abort with received connect user data used as abort user data

D.1.4 Interrupt Tests
Interrupt tests provide a full range of test capabilities from very simple data sink operations through data integrity checking. Interrupt tests that the user can perform are:

•

Sink test. DTR ignores all interrupt data received. No sequence or content validation is performed.

•

Sequence test. Interrupt messages transmitted by DTS to DTR contain a 4-byte
sequence number. If a message is received out of sequence, DTR aborts the logical
link and the test.

0-2

OECnet-RSX Network Generation and Installation Guide

•

Pattern test. Interrupt messages transmitted to DTR have both a sequence number
and a standard data pattern. If either the sequence number or the data pattern is not
received with the expected data, DTR aborts the logical link and the test.

•

Echo test. Interrupt messages received by DTR are transmitted back to DTS. There
is no sequence or data validity checking done by either DTR or DTS.

D.2

Operational Characteristics

DTR functions as a slave to DTS. DTS initiates each test by issuing a connect request to
DTR. Parameter information pertinent to the type of test requested is passed by DTS to
DTR in the optional data of the connect request. DTS has a user interface that enables
the user to specify the test to be performed. Sufficient parameters are available to allow
for a variety of tests, including test duration, buffer size, and buffering level.
The version of DTR supplied on the kits supports a maximum of three logical links,
thus allowing three concurrent tests. DTS can handle only a single test and logical link,
but the user can invoke multiple copies ofDTS.
The DTS command syntax allows for two types of buffer level options: BUFS and
FLOW. BUFS is used by DTS only and FLOW is used by DTR only. The BUFS parameter
specifies the transmit buffering level for DTS. This is the number of transmit requests
that DTS attempts to keep outstanding to the network. In an echo test, the BUFS parameter also specifies the receive buffering level for DTS. The parameter of the FLOW
option specifies the receive buffering level for DTR.
Both DTS and DTR have 512 bytes of buffer space built into each task. The buffer space
is at the end of each task image. To increase the buffer space of either task, the user
need only employ the IINC option at installation time. DTS requires at least (BUFSMSG)
bytes of buffer space of a data test and «BUFS + I)MSG) bytes for an echo test. DTR
requires (FLOWMSG) bytes of buffer space for each data test.

OTS/OTR Test Programs

0-3

0.3

DTS Command Syntax

Formats for DTS commands are described below. The following conventions apply to
the presentation and use of the DTS commands:
Syntax Conventions:
Brackets [ ]

Designate an optional parameter.

UPPERCASE

Designates the actual code that appears in the call.

lowercase italic

Designates a parameter that is replaced with an actual value when assembled.

No space

Do not use a space after the prompt > in a command string.

Default values

Values specified as defaults are valid only for the first test. Before the
conclusion of each test, DTS updates the default values with the parameters of the next test, which become the default values for the next test.
Thus, it is necessary to specify only those parameters that are to be
changed in the next test.

Input

DTS accepts an indirect command file as input.

Command Format:

DTS> [nodename[aci]]::[/TEST = type](/PRINT = choice]
Arguments:
nodename

Name of the node where DTS resides; the default is the local node name.
The name must be terminated by a double colon [::] or underscore for
Version 2 compatibility.

aci

Access control information, in the following format:
[luser-id[lpassword[laccount]]]

For more information, refer to the DECnet-RSX Network Management
Concepts and Procedures manual.
type

Test type:
CON
DIS
DATA
INT

choice

Connect test
Disconnect test
Data test (default)
Interrupt test

YES (default) or NO to indicate whether DTR messages are to be printed
at the remote node as well as at the local node.
NOTE

The print option is used by DTR to either display or suppress message
output to the local console device. Messages include both test results
and error messages. See appendix section D.3.5

D-4

DECnet-RSX Network Generation and Installation Guide

0.3.1

Connect Test

Format:

CONNECT TEST> [type][/DATA = data]
Arguments:

Type of connect test:

type

ACC
REJ

Type of user data:

data

NONE
STD
RCVD

0.3.2

Connect accept test
Connect reject test (default)

No optional user data (default)
Standard user data
Return connect user data

Disconnect Test

Format:

DISCONNECT TEST> [type][/DATA = data]
Arguments:

Type of disconnect test:

type

DSC
ABT

Type of user data:

data

NONE
STD
RCVD

0.3.3

Disconnect test (default)
Abort test

No optional user data (default)
Standard user data
Return connect user data

Data Test

Format:

DATA TEST> [type][/MSG = mmm][/BUFS = bbb][/TIME = time][/BAUD = nnnnnn]
[/FLOW = flow]
Arguments:
type

Type of data test:
SINK
SEQ
PAT
ECHO

DTS/DTR Test Programs

Sink test (default)
Sequence test
Pattern test
Echo test

0-5

mmm

Data message length in bytes: mmm must be greater than 0 for a sink or
echo test, greater than 4 for a sequence test, and greater than 5 for a pattern test. The maximum value is 1024 bytes. The default is the NSP segmentsize.

bbb

DTS transmit buffering level, in the range of 1 to 16. The default is 1.

time

Test duration:
nnnS
seconds
nnnM
minutes
nnH
hours (maximum is 18)
nnn
defaults to seconds
The default is 2M.

nnnnnn

Line baud rate in bits per second (default is 0). If nnnnnn is specified,
the percentage of line bandwidth used by data transfers is calculated.
This percentage is inaccurate for baud rates exceeding 64K bps.

flow

Flow control type or DTR receive buffering level, in the range of 1 to
16:
NONE
No flow control
SEG:nn
Segment flow control (invalid for RSX)
MES:nn
Message flow control
The default is MES: 1. SEG:nn is invalid for RSX DTR because the network interface does not provide that capability.

0.3.4

Interrupt Test

Format:

INTERRUPT TEST> [type][/MSG = mm][/TIME = time]
Arguments:
type

Type of interrupt test:
SINK
Sink test (default)
SEQ
Sequence test
PAT
Pattern test
ECHO
Echo test

Interrupt message length in bytes: mm must be greater than 0 for a sink
or echo test, greater than 4 for a sequence test, and greater than 5 for a
pattern test. The maximum value (and the default) is 16 bytes.

time

Test duration:
nnnS
seconds
nnnM
minutes
nnH
hours (maximum is 18)
nnn
defaults to seconds
The default is 2M.

0-6

DECnet-RSX Network Generation and Installation Guide

D.3.5

DTS/DTR Error Messages

Error codes generated by the DTS and DTR test programs, and the meaning of each, are
listed in this section.
Code

Word

Description

IE.BAD

177777

ABT$-The optional user data exceeds 16 bytes.
ACC$-Either the temporary link address in the connect block is
invalid or the optional user data buffer length exceeds 16 bytes.
CON$-Either the optional user data buffer length exceeds 16 bytes
or the field length count (access control) in the connect block is
too large.
REJ$-Either the temporary link address in the connect block is not
valid or the optional user data buffer exceeds 16 bytes.
XMI$-The interrupt message exceeds 16 bytes.

IE.IFC

177776

ABT$-LUN not assigned to NS:.
ACC$-LUN not assigned to NS:.
CLS$-LUN not assigned to NS:.
CON$-LUN not assigned to NS:.
GLN$-LUN not assigned to NS:.
GND$-LUN not assigned to NS:.
REC$-LUN not assigned to NS:.
REJ$-LUN not assigned to NS:.
OPN$-LUN not assigned to NS:.
SND$-LUN not assigned to NS:.
SPA$-LUN not assigned to NS:.
XMI$-LUN not assigned to NS:.

IE.SPC

177772

ACC$-Invalid buffer argument; either the pending connect block or
the optional user data buffer is not word aligned, or one of them
is outside the user task address space.
CON$-Invalid buffer argument; either the connect block is not
word aligned, or the optional user data buffer (in or out)
are outside the user task address space.
GLN$-The buffer specified to receive the network data is outside
the user task address space.
GND$-The buffer assigned to receive network data is not word
aligned or is outside the user task address spice.
REC$-Either the data buffer is outside the user task address space or
the buffer length exceeds 8192 bytes.
REJ$-Either the connect block or the optional user data buffer (out)
is outside the user task address space or the connect block is not
word aligned.
(continued on next page)

OTS/OTR Test Programs

0-7

Code

Word

Description

SND$-Either the data buffer is outside the user task address space or
the buffer length exceeds 8192 bytes.
XMI$-The interrupt message buffer is outside the user task address
space.
IE.WLK

177764

XMI$-An interrupt message was transmitted before a previous interrupt message had been received by the target task.

IE.DAO

177763

CON$-Connection rejected by user and some optional user data was
lost.
GLN$-The network data was longer than the specified buffer.
GND$-The network data was longer than the mail buffer.
REC$-More message data was transmitted than requested.

IE.ABO

177761

ABT$-The specified logical link has been aborted or disconnected.
ACC$-The task that requested the connection has aborted or has
requested a disconnect before the connection could complete.
REC$-The logical link was disconnected during I/O operations.
REJ $- The task that requested the connection has aborted or has
requested a disconnect before the rejection could complete.
SND$-The logical link was disconnected during I/O operations.
XMI$-The logical link was disconnected during I/O operations. .

IE.PRI

177760

CLS$-The network is not accessed on the specified LUN.
CON$-The local node is shutting down. No logical link can be established.
GLN$-The network is not accessed on the specified LUN.
GND$-The network is not accessed on the specified LUN.
OPN$-The network is being dismounted or the user task has already
accessed the network.
REJ$-The network is not accessed on the specified LUN.
SPA$-The network is not accessed on the specified LUN.

IE.RSU

177757

ACC$-System resources needed for the logical link are not available.
CON$-System resources needed for the logical link are not available.
GND$-System resources needed for the network data queue are not
available.

IE.ALN

177736

ACC$-A logical link has already been established on the specified
LUN.
CON$-A logical link has already been established on the specified
LUN.
(continued on next page)

0-8

DECnet-RSX Network Generation and Installation Guide

Code

Word

Description

IE.NLN

177733

ABTS-No logical link has been established on the specified LUN.
SNDS-No logical link has been established on the specified LUN.
XMI$-No logical link has been established on the specified LUN.

IE.UR)

177667

CONS-Connection rejected by remote user task.

IE.NR)

177666

CONS-Connection rejected by Network. (see reasons below)

IE.NDA

177662

GNDS-There is no data in the network data queue.

IE.NNT

177642

ABTS-The issuing task is not a network task
ACCS-The issuing task is not a network task
CLSS-The issuing task is not a network task
CONS-The issuing task is not a network task
GLNS-The issuing task is not a network task
GNDS-The issuing task is not a network task
REC$-The issuing task is not a network task
RE)S-The issuing task is not a network task
SNDS-The issuing task is not a network task
SPAS-The issuing task is not a network task
XMI$-The issuing task is not a network task

Disconnect or Reject Reason Codes
Code
NESRES
NESNOD
NESNSR
NESUOB
NESFMT
NESMLB
NESABM
NESNNF
NESNSL
NESACC
NESABO
NESCOM

Value
1
2
3
4

6
10

12
13
42
46
47

Description
Insufficient network resources
Unrecognized node name
Remote node shutting down
Unrecognized object
Invalid object name format
Object too busy
Abort by network management
Invalid node name format
Local node shutting down
Access control rejected
No response from object
Remote node or object failed
Node unreachable

DTS/DTR Test Programs

(connect reject)
(connect reject)
(connect reject)
(connect reject)
(connect reject)
(connect reject)
(network abort)
(connect reject)
(connect reject)
(connect reject)
(connect reject)
(network abort)
(connect reject)
(network abort)

D-9

Index

A
Areas,
configuration guidelines, 2-7
defined, 2-7

c
CCR,2-17
CDA, B-1
CETAB,
rebuilding, C-1
CETAB.MAC, 1-4,6-33
CEX,
rebuilding, C-1
CFE, 1-4, 1-5,2-5,2-11,5-36,
5-37
Checkpointing, 3-13, 3-14, 3-16,
3-18
Circuit,
errors, 10-19
ON-STARTING errors, 10-17
Communications Executive, 4-3
Components,
always built, 2-11
never used, 2-11
optional, 2-11
required, 2-11
Configuration File Editor
seeCFE

Configuration Guidelines, 2-3

Configuration Guidelines,
Area, 2-7
Ethernet, 2-5
General, 2-4
Counters,
showing with NCP, 10-10
using for error detection,
10-10
CSR, 10-18
CSR address, 5-21

o
Data link mapping, 11-1
data link mapping (DLM), 1-5
DCP, 2-14, 2-15
DDM,
rebuilding, C-1
DL11,10-25
DLL,2-16
DLM,2-19
DLM (data link mapping), 1-5
DLM,
verifying installation, 11-4
DLX, 2-14, 2-15, 2-16,3-8, 5-38,
5-41,6-9,6-16,6-20
DMP/DMV,10-25

DMR,10-25
Down-line loading over DMP IDMV,
3-22·
DSRpool, 3-11, 3-18, 3-21

Index-1

DTE, 11-2
DTS,
command formats, D-4
DTS/DTR, 10-4, 10-22
DTS/DTR tests, D-l
DTS/DTR,

connect test, D-l
data test, D-l
disconnect test, D-l
interrupt test, D-l
DUK, 2-19
DUM, 2-17

Installing tasks,
using VMR commands, 9-8

K
KDA, 2-19
Kit,
end node, 1-1
full function, 1-1
pregenned end node, 1-1

ECL, 6-9
EPM, 2-14, 2-15
Ethernet,
error handling, 10-26
EVC, 2-15, 5-33
Event logging, 10-1
EVF, 2-15,5-36
EVL, 2-15, 5-33
EVLlEVC, 5-32
EVR, 2-16, 6-14
Executive address space, 3-18,
3-20
EXELIB.OLB, 5-3, 5-4
EXEMC.MLB, 5-3, 5-4
Explanatory text, 8-2
Extended network support, 6-6,
6-9,6-33

LAB process, 7-16
LAT, 6-28
Library,
memory resident, 1-2
supervisor mode, 1-2
LIN, 2-16, 6-14
Loading the network,
calling sequence, 3-12
using NCP, 3-12
using VNP, 3-12
LOO, 5-36
Loopback node,
clearing with NCP, 10-16
Loopback test,
preparing to run, 10-13
running, 10-14
Loopback,
controller, 10-13, 10-17
node level hardware test, 10-13
LSN,2-18,6-28, 10-4, 10-23

FAL, 2-17,6-23,6-24, 10-23
FTS, 2-17, 3-4,3-18,6-23
FTSDEQ, 2-17

MCM, 2-18, 6-23, 6-27
MIR, 6-14
MLD, 2-14, 5-32

HLD, 2-17
Host system,
requirements, 2-1

NCP, 1-5,2-11,2-15,3-9,3-13,
5-24,5-36
NCP,
using the LOOP command, 10-24
NCT, 2-18, 6-28
NDA, 5-36, 5-37, B-1
NETCFE.CMD, 1-4
NETCFG. TXT, 1-4

Installing DECnet, 1-4
Installing DECnet,
on RSX-IIM/M-PLUS, 9-3

Index-2

NETFOR.OLB, 2-16
NETGEN, 1-1, 1-2
NETGEN Question responses,
Break, 4-10
End-of-section, 4-9, 8-3
NETGEN Response format,
character string, 4-8
numeric, 4-8
YES/NO,4-7
NETGEN response formats, 4-7
NETGEN,
abnormal termination, 4-10
component mode, 4-5
dialog format, 4-7
dry run mode, 4-5
duplicating, 4-6
explanatory text, 4-7
operating modes, 4-5
organization, 4-3
parameter defaults, 4-1
question and answer mode, 4-5
response formats, 4-7
restore mode, 4-5
running on RSX-IIMI11M-PLUS,
4-2
running on VAX-II RSX, 4-2
saved response files, 4-6
starting, 4-1
NETINS.CMD, 1-3, 1-4,3-14,5-7,
6-25
NETINS.CMD,
executing, 9-3
modifying to add a new
component, 9-6
modifying to change the way a
task is installed, 9-6
modifying to install a task in
a different partition, 9-6
modifying to install and start
network automatically, 9-7
questions, 9-5
NETLIB.MLB, 2-16
NETPAN,3-23,6-20,B-l
NETREM.CMD, 1-3
Network command terminal support,
3-15
Network Control Program

Network Installation Command File
see NETINS. CMD
NFT,2-17,6-23,10-23
NICE,2-16,3-13,6-14
NMDRV, 2-14, 2-15, 3-12,3-13,
5-36
NMVACP, 5-36
Node,
Addresses, 2-4
Levell routing node, 2-9
Level 2 routing node, 2-9
Names, 2-5
Phase III, 2-9
NTD, 2-16, 6-14, 6-15
NTDEMO, 2-16, 6-14, 6-15
NTEST. CMD, 10-4
NTEST.CMD,
dialog, 10-6
failure handling, 10-10
preparing to use, 10-4
running, 10-5, 10-21
NTINIT, 3-9,3-12,5-9,5-32
NTL, 3-12, 5-32
NVP, 2-16, 6-14, 6-16
NW process, 7-16

o
Open user group, 11-1
Outgoing circuit,
subaddress, 11-5

p
Packet Switching Data Network
(PSDN), 2-14, 5-26,7-1
Partition layout,
after SYSGEN, 3-6
definition, 3-4
for adding CEXPAR, 3-5, 3-19
for large RSX-I1M systems, 3-10
for medium RSX-I1M systems, 3-9
for other partitions, 3-21
for RSX-llS, 3-22
for small RSX-I1M systems, 3-8
for user applications, 3-6
to avoid loading problems, 3-11
to include NETPAN, 3-23

seeNCP

Index-3

Partitions,
forGEN,3-5
system-controlled, 3-16
Passwords, 10-24, 10-27
PCL,5-18,5-25,8-7,10-13
Permanent virtual circuit (PVC),
7-6,7-7
PHO, 2-18,6-28
PLI process, 7-16
PREGEN, 1-1, 1-2
PREGEN,
copying media, 3-26
description, 3-24
dialog, 3-29
disk distribution media, 3-27
for a small dual-disk system,
3-25,3-27,3-30,3-34
for a standard system, 3-25,
3-26,3-31
initializing the disk, 3-27
magnetic tape distribution
media, 3-28
notes, 3-25
sample of a standard system,
3-36
sample small dual-disk system,
3-38
Pregenned end node,
dialogue, 8-5
installation procedure, 8-1
preliminary steps, 8-1
PSI, 4-3, 4-9
PSI,
error counters, 11-3
fatal error, 11-3
status counters, 11-3
warning message, 11-3
PSI.MLB, 2-20, 7-19
PSIFOR.OLB, 2-19, 7-19
PVC,
channel numbers, 11-4
maximum packet size, 11-4
maximum window size, 11-4

Q
QMG.OLB, 5-4
Question responses,
Break,8-3

Index-4

Question responses (Cont.)
Break questions, 8-3
Queue Manager (QMG), 3-4, 3-18, 5-12

R
RCP, 2-14, 6-9
Response format,
character string, 8-3
numeric, 8-2
YES/NO, 8-2
RMHACP, 6-28, 6-29
RMHACP/HTDRV, 2-18
RMS FAL, 6-24, 6-25
RMS11X.ODL, 5-4,6-24,6-25
RMS12X.ODL, 5-4,6-24
RMSLIB.OLB, 5-4, 6-24
RMSRES.STB, 6-24
RMSRES.TSK, 6-24
RMSRLX.ODL,6-24
RMT, 2-15, 3-17,6-28
RMTIRMTACP, 2-18
Routing parameters,
area maximum hop count, 2-11
highest node address, 2-10
line/circuit cost, 2-11
maximum area value, 2-11
maximum hop count, 2-11
RSX-11 PSI, 1-5
RSX-11M,
standard function system
restrictions, 3-15
system requirements, 3-4
terminal driver support, 3-17
RSX-11M-PLUS,
pregenned systems, 3-1
SYSGEN considerations, 3-3
system requirements, 3-1
RSX11M.STB, 5-3
RSXl1S.STB, 5-3
RSXMAC.SML, 5-3, 5-4
RSXMC.MAC, 3-22, 3-30, 5-3
RTH, 2-18, 6-28

s
Saved response flIes, 4-6
SCPXTS, 11-1
SCPXTS,
error messages, 11-2

SCPXTS (Cont.)
preparing to run, 11-1
running, 11-2
SLD, 2-17, 3-22
SVC Circuit,
incoming, 11-4
outgoing, 11-4
SVC circuits,
testing, 11-5
SVC,
maximum packet size, 11-5
maximum window size, 11-5
usage parameters, 11-4
Switched virtual circuit, 11-1
Switched virtual circuit (SVC),
5-26,7-7
SYSGEN, 1-1,3-3,3-4,3-5,
3-12,3-14,3-15
SYSGEN,
assemble the executive section,
B-2
dynamic checkpoint support,
3-16
editing files, 3-22
editing SYSVMR.CMD, 3-19
Executive options, 3-16
executive options section, B-2
FTS support, 3-4, 3-18
standard function system, 3-15
support for DECnet, 3-3, 3-15
system-controlled partitions,
3-16
terminal driver support, 3-17
SYSLIB.OLB, 5-3, 5-4
System controlled partitions,
3-15
System memory size,
for CEX, 3-20
for FCPPAR, 3-9
for Files 11 ACP, 3-7
for GEN, 3-5
for NETACP, 3-8
for NMVACP, 3-8
for other partitions, 3-21
System-controlled partitions,
3-16,3-22
SYSVMR.CMD, 3-18, 3-19

T
Target system,
definition, 1-2
disk capacity, 1-2, 2-1
requirements, 2-2
Target,
defining network
characteristics, 7-1
defining remote nodes, 6-1
defining the system, 5-11
TCL, 2-19,6-28,6-32
Test,
local node software, 10-4
local node test failure
conditions, 10-10
node level hardware loopback, 10-13
Remote Node, 10-21
Testing the network, 1-4
TLK, 1-5,2-18,3-17,6-28,
6-31, 10-4, 10-23
TRA, 2-19,7-19
Transport subaddress, 7-3
TRI, 2-19, 7-19
Tuning the network, 1-5

u
UIC group code,
DECnet, 8-6
for NETGEN output, 5-6
network, 4-6, 9-6, 9-7
Upline dump, B-2
UserID, 5-7, 5-9
User responses, 4-7, 8-2

v
VAX-11 RSX, 2-1, 4-2
VAX-II RSX,
running NETGEN on, 4-2
VAXNMS,
Installing DECnet-I1S on, 9-11
Privileges needed for
installation of DECnet-11S,
9-11
Vector, 10-18
Vector,
highest interrupt address, 3-4, 3-15

Index-5

Virtual Monitor Console Routine
seeVMR

Virtual Monitor Console Routine
(VMR),5-29
Virtual Network Processor
seeVNP

Virtual Network Processor (VNP),
8-7,8-8,8-9
VMLIB.OLB,6-24
VMR, 3-19, 3-20, 3-21, 9-7, 9-9,
9-10,9-13
VMR,
using to install NCP and NICE,
10-1
VNP, 1-5,2-11,5-36,5-37,9-10,
9-13

x
X.29, 1-5, 11-1
X.29 support, 7-3, 7-4, 7-11, 7-19
X.29,
verifying installation, 11-4
X25ACP,7-11,7-15,7-16
X29ACP,2-19,3-18,6-30,7-19
XTR,11-1

11SLIB.OLB, 5-3

Index-6

DECnet-RSX
Network Generation and Installation Guide
AA-H225D-TC
READER'S COMMENTS
What do you think of this manual? Your comments and suggestions will help us to improve
the quality and usefulness of our publications.

Please rate this manual:
Accuracy
Readability
Examples
Organization
Completeness

Poor
1
1

2
2
2
2
2

3
3
3
3
3

4
4
4
4
4

Excellent
5
5
5
5
5

Did you find errors in this manual? If so, please specify the error(s) and page number(s).

General comments:

Suggestions for improvement:

Name _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Date _ _ _ _ _ _ _ _ _ __
Title
Company
City

Department
Street _ _ _ _ _ _ _ _ _ _ _ __
State/Country _ _ _ _ _ _ Zip Code _ _ _ _ __

DO NOT CUT

FOLD HERE AND TAPE

111111

NO POSTAGE
NECESSARY
IF MAILED
IN THE
UNITED STATES

BUSINESS REPLY LABEL
FIRST CLASS PERMIT NO 33 MAYNARD MASS
POSTAGE WILL BE PAID BY ADDRESSEE

SOFTWARE DOCUMENTATION
1925 ANDOVER STREET TWO/E07
TEWKSBURY, MASSACHUSETTS 01876

DO NOT CUT

FOLD HERE

o
o

-4
-4