Digital PDFs

AA-MES5A-TE

1988

46 pages

Original

2.1MB

view download

OCR Version

1.7MB

view download

Document:	ULTRIX-32 Guide to the Error Logger System
Order Number:	AA-MES5A-TE
Revision:	0
Pages:	46
Original Filename:

OCR Text

ULTRIX-32
Guide to the
Error Logger System

Order No. AA-MEOS5A-TE

ULTRIX-32 Operating System, Version 3.0

Digital Equipment Corporation

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 be used or
copied only 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.

The following are trademarks of Digital Equipment Corporation:

DEC
DECnet
DECUS

Q-bus
RT
-

VAX
VAXstation

ULTRIX

VMS

ULTRIX-11

MicroVAX

ULTRIX-32

ULTRIX Worksystem Software

PDP

UNIBUS

dlilg]iltla]]

MASSBUS

UNIX is a registered trademark of AT&T in the USA and other countries.
IBM is a registered trademark of International Business Machines Corporation.
MICOM is a registered trademark of Micom System, Inc.
This manual was written and produced by the ULTRIX Documentation Group in Nashua, New

Hampshire.

Contents

About This Manual
AUIBINCE

covviiiiiiieeii

e e eee

vii

...eoicuiieiiiiriiiii et
eee e e ett
e e s eeeeee e eesereeseaeeseeane e

vii

COMVENEIOIIS 1ovvutiieet
eeeeieeee
eeeaessee s reeeeseees s eeieeeeeee
et
e
ee

vil

OrANIZALION

Error Logger Overview

1.1 Error Logger Components

..........cccccevevviiiiiierioiieeereoiieeeesseiseessesiinnennns

1.1.1 The Kernel BUffer .........cccccoviiiiiiiiiiiiiiieee
e eeesaeneae
eeeeeee
e,

1-1

1-2

1.1.2 The elesd Daemon

........cccocvvvevimnioiiiieeeeeeeeeeeeeeeeeeenenns ceerrrieeann

1-2

1.1.3 The elcsd.conf File

.......cccoovviiiiiiiiiiiiiiieee
e eeeeeeeie e eeeee
e e e,

1-2

1.2 Error Logger OPeration ..........cccoccceeivvoeieeeieeceeeeeeeiieeseeereseeseissesssssrsssssons

1-2

1.2.1 Local System Crash ......ccccooviiiiiii
e
eeeeeeeiee
ieoeie
e eeains
1.2.2 Remote System Shutdown .........cccccovevveiiieoieeeeeeieiereseiineeeeesnnn,
1.2.3 Remote System Crash .........ccocoviviiiiiiiciiiiiiieeeeeeeeceeeeeeee
e,

1-4

1-3
1-4

2 Error Logger Management and Maintenance
2.1 Manual Control of Error Logger Functions

......ccccccoovovveeeeeeeevveevenn,

2-1

2.1.1 Disabling Error Logging ......cccccooovviiiiiiieeeeeeeeeeeeeeeeeeseneeeeeenn,
2.1.2 Enabling Error Logging ......cccccccovviivimiiieeeeeeceeeeeeeeeeeeeeiieeseenenn,

2-2

2.1.3 Enabling Error Logging in Single-User Mode

....ccocccevvvvvennn..

2-1

2-2

2.1.4 Clearing the Errorlog Buffer

..........ccccoooviiiiiiiiiiiiiiiiiiiiiiiiiiennenn,

2.1.5 Logging Messages to the Errorlog File

2-3

........cccccoovvieiiiiiiiiiiiiniiiirnriiciee
e,

2-3

......cccccooiviiiiiiiiiiiiiniiiiiiiiiiiee
et
eeeeaneer e s e e eanareenns

2-3

2.1.6 Restart the elcsd Daemon
2.2 ReconfiguratiOn

2.3 elcsd.conf File Parameters

2.3.1 StAtUS

2-3

........cccoooviiiiiiiieninnnnnn.n.

P TSP

2-5

.oovieeciiieeee
et e e e e e e e s e
aa

2-5

2.3.2 Errorlog File SiZe

..oocoooiiiiiiiiiiiiiiicii

2.3.3 Errorlog Directory Path

2-6

........vvvviiiiriiiiiiiiiencccee
e,

2-6

2.3.4 Backup Errorlog Directory Path

.......cccoviiiiiiiiiriiiiiiiiiieneeinnns

2.3.5 Single-User Errorlog Directory Path

........ccccoooviiiiiiiiiiiiiiiiennnnnn.

2.3.6 Remote Hosts Errorlog Directory Path
2.3.7 Remote Host Name To Log To

............ccccoviiiiiieeennnnnns

2-6

2-7
2-7

...ccooivrririiiriieiecciccccneeeeen,

2-7

2.3.9 Sample Server and Client elcsd.conf Files .......ccccovevviiviiienennnn.

2-7

2.3.8 HoStS T LOZ

coovviiiiiiiiiiieieiiiie

..........cccccccvvvevirivrreivnnieerennnnnnnns

2-8

2.3.9.2 Client system file entries ..........cccccviviriiinriiiiiiiiiiiieeeeennn.

2.3.9.1 Server system file entries

2-9

2.4 Error Logger File Maintenance ...........ccccccceeiiiiiviieeriririiiiieiicciiiisceneenenns

2-10

2.4.1 The elcsdlog File

...ccoooeveeiiiiiiiiiiiiieeeiie et

2.4.2 Backing up the Errorlog File
2.4.3 Clearing an Errorlog File

2-10

.....ccoviveiiiiiiiiiiiiiiiiiiieeen,

2-10

2.5 Crucial Error Logger System Messages

3 The

2-10

........c.cccoovmriiriimiiriiiiiiiciiiciir
e,

........cccccooceeevvrivvvereeeiiiiinennnnn.

2-12

....ccccoeeeeeiiiiiiiciciieeeeeeeeeeeeee
e

3-2

Error Report Formatter

3.1 Generating ReportS

3.1.1 Formatting Reports

....ccccoovviviiiiiiiiiiiiiiieeeciieecvierei

3.1.2 Generating Reports from Specific Files

......ccccccovvviiviiiiiiiiinnnnn.

3.1.3 Generating Reports from Specific Host Systems
3.1.4 Reporting Errors As They Occur

.................

3-6

........cccooovvvvvvrrviviiiiiiiieneennnen.

3-6

3.1.5 Generating Reports in Reverse Chronological Order

.............

3-6

.......cccoovevvvivvceeeennnnn.

3-6

........ccoooovevvviiiiiiivnvinnieeieieeeeennenn,

3-6

3.1.6 Generating Reports in Single-User Mode

3.1.7 Generating Summary Reports
3.2 Selecting Error TYPes

iv Contents

3-3

3-5

....ccccoiiiiiiiiiiiiiiii

3-7

......cccccccoiiiiiiiiiiiiiieeiiiiieiiieecereveivienen,

3-7

3.2.2 Selecting Class ETrrors

3.2.1 Selecting Adapter Errors

...ccccccovveeiiiiiiiiiiiiiiiieveeee

3-8

3.2.3 Selecting Disk EITOTrs

.....cooocooiiiiiiiiiiiiiiireeeeeeeceee
e ceescecnaans

3-8

3.2.4 Selecting Mainframe Errors .......ccccocceeeeviiiiiiiiiiiieeiiiiiiiieeeeeennann,

3-8

........ccccccccrvirrrirereevivvnnnnnnnn.

3-9

3.2.6 Selecting Tape Errors ....cccccceeieiiiiiiiiiiiiiiiiiiiiiieeeeeiieeneeen

3.2.5 Selecting Operating System Errors

3-9

3.3 Selecting Specific Errors .......cccooocciiiiiiiiiiiriiieeeeeereecececescnninnreereeeeeeneee,

3-9

3.3.1 Selecting Errors by Record Type

....ccccoevveeiiviiiiiiiieiiniiiirniinnnnn.

3.3.2 Selecting Errors by Sequence Number

3-10

.............cccccocevviiiieiinnnns

3-11

3.3.3 Selecting Errors Using a Time Range .....cccccccceiiivvivviinnnnnnnnn.

3-11

3.3.4 Excluding Error TYpes

3-12

.....cccccciiiiiiiiiiiieeeccccccceee
e

3-13

3.4 Using uerf OPtions

3.5 uerf Option Reference

...ccoovoeriiiiiiiiiiiiieeeeeeeeeeiee

........cccoooviiiiiiiiiiiiiiiie

3-13

......cceviiveiiiiiiiiiiiiiininnn.

Examples
2-1:

The elcsd.conf File with Default Entries

2-2:

elcsd.conf File Entries for a Server System

............ccccooomvririiiinnn.

2-8

2-3:

elcsd.conf File Entries for a Client System

..........cccccoccviviviviinnnnn.

2-9

3-1:

The Help DiSPlay .....couevviiieeiiiiiiiiiiiiiiiiiieieeeeeeeeeee
s

3-3

3-2:

Memory Error in Brief Format

.......ccccocooovviiiiiiiiiiiiiceiiieeeeeeeereeeeeneenens

3-4

The uerf Command Options ..........cccccoevviviiiiriiiiieiiiinieeeeneeeeeeesieieeenas

3-13

Tables
3-1:

Contents v

About This Manual

This guide describes how the ULTRIX Error Logger System records and
reports errors and other events occurring on your MicroVAX or VAX
system.

Audience
The audience for this manual is anyone who manages error information on
While this person is usually the system manager, others such

the system.

as any DIGITAL service representative can use the Error Logger System
for analyzing error information to help identify the cause of problems in
the system.
This guide assumes that the reader is familiar with basic
ULTRIX-32 system functions.

Organization
Chapter 1

Error Logger Overview
Describes the components and operation of the error logger
system.

Chapter 2

Error Logger Management and Maintenance
Describes how to control error logger functions and how to
reconfigure the error logger. It also describes file parameters
and file maintenance procedures.

Chapter 3

Error Report Formatter - uerf
Describes how to generate and format error reports.
It also
describes how to select specific error types for specific
purposes.

Conventions
The following conventions are used in this manual:

special

In text, each mention of a specific command, option,
partition, pathname, directory, or file is presented in this
type.

command(x)

In text, cross-references to the command documentation

include the section number in the reference manual where
the commands are documented. For example: See the cat(1)
command. This indicates that you can find the material on
the cat command in Section 1 of the reference pages.
UPPERCASE

The ULTRIX-32 system differentiates between lowercase
and uppercase characters.

Enter uppercase characters only

where specifically indicated by an example or a syntax line.
italics

In syntax descriptions, this type indicates terms that are
variable.

example

In examples, computer output text is printed in this type.

example

In examples, user input is printed in this bold type.

This is the default superuser prompt.

>>>

This is the console subsystem prompt.

In examples, a vertical ellipsis indicates that not all of the
lines of the example are shown.

<KEYNAME > In examples, a word or abbreviation in angle brackets

indicates that you must press the named key on the
terminal keyboard.

viii About This Manual

Error Logger Overview 1

The error logger is a system management utility designed to help you and
your DIGITAL field service representative analyze stored error information.
The error logger records and reports errors and other events that occur on
your MicroVAX or VAX system.

Together with the error report formatter,

you can produce error reports that will help you identify the cause of

problems with your system.

This chapter describes the error logger system.
°

Error logger components

Error logger operation

It covers:

Chapter 2 describes tools and files that you can use to control error logger
functions manually.

These functions include reconfiguring the errorlog

configuration file, enabling error logging in single-user mode, and clearing
the kernel buffer.

The error logger system also contains the ULTRIX error report formatter,
uerf, which allows the user to generate and format several specific types of
error reports.

Information on uerf is contained in Chapter 3.

In all discussions of the error logger, the following definitions apply:

server:

The system that receives error messages from other systems.

client:

Any system that logs messages to the server.

1.1

Error Logger Components

The error logger consists of:
o

A kernel buffer that automatically logs error messages in binary form

The elcsd daemon, which transfers error messages from the kernel
errorlog buffer to an errorlog file (local or remote)

The elcsd.conf file, which contains configuration parameters for the
elcsd daemon

The error logging system files reside in the /etc directory.

1.1.1

The

Kernel Buffer

When you start up the system, the error logger automatically logs all error
messages in binary form to an errorlog buffer in the kernel.

These

messages are stored in the kernel errorlog buffer temporarily until the
elcsd daemon transfers them to an errorlog file on disk, where they are
stored permanently.

The part of the error logger that sends error messages to the kernel
errorlog buffer cannot be disabled.

1.1.2

The

elcsd

Daemon

The elcsd daemon is invoked when the system starts up in multiuser
mode. The daemon transfers error messages from the kernel errorlog buffer
to the errorlog file specified in the elcsd.conf file until you shut the
system down.

You can start and stop the elcsd daemon by using the eli command.
However, the elcsd daemon does not run when the system is in single-user
mode.

During single-user mode you must use the eli command with the

—-S option to run the elcsd daemon and transfer error messages to the

errorlog file.

There are other occasions when you will want to enable or disable the
elcsd daemon manually.

See Chapter 2 for a discussion of the eli

command and its options.

1.1.3

The elcsd.conf File

The elcsd.conf file contains the default information used by the elcsd
daemon to configure error logging.

The parameters in the elcsd.conf file

determine the names and locations of the main, backup, and single-user
errorlog files, the size limit for the files, and provide information on how
messages are logged between systems.

The default errorlog file is /usr/adm/syserr/sysérr.hostname.
See Chapter 2 for a discussion of the elcsd.conf file and its parameters.

1.2

Error Logger Operation

When you start the system in multiuser mode, the elcsd daemon is
invoked by an entry in the /etc/rc file and begins logging error messages to
the errorlog file.

If you delete this entry, the elcsd daemon will not

transfer error messages from the kernel errorlog buffer to the errorlog file.

Here is the entry as it appears in the /etc/rc file:

1-2

Error Logger Overview

[

-f

/etc/elcsd

]

{

echo

“starti

errlog

daemon

elcsd’

>/dev/console

}
See the rc(8) command in the ULTRIX Reference Pages for more
information on the /etc/rc file.

The default errorlog directory path for multiuser error logging is
/usr/adm/syserr.

To change the default directory path, or any other

parameters in the elcsd.conf file, see Section 2.3.
The following sections describe how the error logger operates when:
o

The local system crashes

The remote system crashes

The remote system shuts down

1.2.1

Local System Crash

When a system crashes, the error messages in the kernel errorlog buffer
are saved by an entry in the /etc/rc.local file.

When the system comes

back up, the messages are appended to the errorlog file.

The default entry

in the rc.local file is:
/etc/savecore

/usr/adm/crash

>/dev/console

This entry saves the core dump and the messages in the kernel errorlog
buffer after a system crash.

The core dump (vmcore and vmunix) is stored

in the directory /usrfadm/crash.

The kernel errorlog buffer data will be

stored in the file /usr/adm/syserr/elbuffer temporarily.

When the elcsd

daemon is started, it appends the error messages to the errorlog file and
removes the elbuffer file.

See the Guide to System Crash Recovery for

more information on savecore and system crashes.

If you do not have adequate disk space to save the system core dump,
add the —e option to the savecore entry in rc.local.

This entry saves only

the data in the kernel errorlog buffer:
/etc/savecore

—e

/usr/adm/crash

>/dev/console

Note
You must have a savecore entry in the rc.local file to save data
in the kernel errorlog buffer after a crash. If you do not, you
may not be able to determine why the system crashed.

Error Logger Overview 1-3

1.2.2

Remote System Shutdown

If client systems are logging errors to a server and the server system is
shut down for maintenance or repairs, the client system logging to the host
1s automatically informed of the shutdown.

The client system then begins

logging errors to its own errorlog file. When the server system comes back
up, the client is informed that service has been restored and resumes

logging error messages to the server.
After a shutdown occurs, there are two files containing client messages:
®

The errorlog file on the server system

The errorlog file on the client system

1.2.3

Remote System Crash

If client systems are logging errors to a server and the server system
crashes, the client system is not informed of the crash and continues
trying to log errors to the server.
messages are not saved.

Because the server is down, these

When the server system reboots, the error

messages from the client are again logged in the errorlog file on the server
system.

If the server system is going to be down for some time, you can
reconfigure the client remote system by changing its elcsd.conf file to log

locally by changing /etc/elcsd.conf.

Then use the eli command with the —r

option to reconfigure error logging on the server.

If a client system goes down while logging errors to the server system, the
server system
1s not affected.

When the client system comes back up, it

automatically begins sending error messages to the server again.

1-4 Error Logger Overview

Error Logger Management and Maintenance 2

This chapter describes tools and procedures used to manage and maintain
It contains information on:

the error logger system.
°

Manual control of error logger functions

Reconfiguration

Error logger file maintenance

Error messages

2.1

Manual Control of Error Logger Functions

The eli command lets you manually control these error logger functions:
o

Enable and disable error logging

Clear the errorlog buffer

Log messages to the errorlog file

Restart the elcsd daemon

The following sections describe how to use eli to perform these functions.

Refer to eli(8) in the ULTRIX Reference Pages for more information on
the command options.
2.1.1

Disabling Error Logging

To disable error logging to an errorlog file, type:
#

/etc/eli

-d

You can use this option when your file system is full while you free up
more disk space.

However, when you use this option, the elcsd daemon

stops logging error messages.

Messages will queue up in the kernel buffer

until the buffer becomes full.

These messages will be logged to disk when

the error log daemon is restarted.

2.1.2

Enabling Error Logging

To re-enable error logging in multiuser mode, type:
#

/etc/eli

-e

/etc/eli

or:

—e

When you use the —f option, the system suppresses the prompts you
usually see when you select the — e option.
The elcsd daemon will create
a new errorlog file.

The default pathname will be

/usr/fadm/syserr/syserr.hostname.
2.1.3

Enabling Error Logging in Single-User Mode

Error messages can be logged during single-user mode and multiuser mode.
In single-user mode, however, you must start up the elcsd daemon
manually to log error messages to the errorlog file.
Before starting the elcsd daemon in single-user mode, run fsck on the file

system you will be logging to check for file system inconsistencies. The
root file system is the default file system for error logging in single-user
mode.
Then invoke the eli command by typing:
#

eli

-s

The —s option enables error logging in single-user mode.

Error messages

are transferred from the kernel errorlog buffer to the file specified in the
elcsd.conf file for messages generated during single-user mode.

The root (/)

directory is the default location for the single-user errorlog
When the system makes the transition to multiuser mode, by default,
it appends the messages in /syserr.hostname to the end of the multiuser

file.

errorlog file, syserr.hostname. in /usr/adm/syserr.
save

error messages

If you do not want to

generated during single-user mode,

remove the

user errorlog file before you make the transition to multiuser mode.

single-

change the single-user directory path or any other parameters in the
elcsd.conf file, see Section 2.3.

2-2

Error Logger Management and

Maintenance

2.1.4

Clearing the Errorlog Buffer

To clear the kernel errorlog buffer, type:
#

/etc/eli

-i

This command line initializes the kernel errorlog buffer.

The messages in

the kernel errorlog buffer are cleared without being logged to disk.
If a problem occurs anywhere in the system that stops the elcsd daemon

from running, an error message informing you that error messages are not

being logged to an error log file appears on the console every 15 minutes.
To stop this message, type:
#

/etc/eli

-q

To re-enable logging this missed-error message to the console, type:
#

/etc/eli

2.1.5

—-w

Logging Messages to the Errorlog File

You can log a one-line message to the errorlog file with this command line:
#

/etc/eli

The system prompts you to enter the message.

message that is contained in a file.

You can also log a

For example, using the file myfi le,

the command line entry is;
#

eli

-f

-1

myfile

/dev/null

This example logs a message, up to and including the first newline, from

the file named

myfi le.

The example also directs the eli prompt to

/dev/null.

2.1.6

Restart the elcsd

Daemon

If you change any entries in the elcsd.conf file, you must restart the elcsd
daemon so that the new parameters will take effect.
#

2.2

eli

Type:

—r

Reconfiguration

Reconfiguration is the process of disabling the error logger, modifying the
errorlog configuration file, and then restarting the error logger.

This

section describes the error logger configuration file, elcsd.conf.

The

elcsd.conf file contains the information used by the elcsd daemon to
configure error logging.

The entries in the elcsd.conf file direct local error

logging as well as error logging between systems.

Error Logger Management and

Maintenance 2-3

Anyone can view the elcsd.conf file, but you must be superuser to change
the entries in the file. Whenever you change the entries in the elcsd.conf
file, use the eli command with the —r option to reconfigure the elcsd

daemon.
You can log error messages between systems by setting up an errorlog file
on a server system to receive error messages for a client system. For

example, if you have a MicroVAX system with limited disk space, you can

log its error messages to a larger system.

However, you must specify in

the elcsd.conf files on both systems how you want to log errors.

The

client elcsd.conf file must specify where messages are to be sent, and the
server elcsd.conf file must have an entry naming the client system it will
accept messages from. Entries in the server elcsd.conf file must also

specify the pathname for the client messages file.

To enable error logging between systems, you must have the following
entry in the /etc/services file for both server and client systems:
elcsd

704/udp

#error

log

daemon

Note
You can only log errors to another system when the systems

share the same network.

See the Guide to Networking for

information on networks.

Example 2-1 shows an elcsd.conf file with default entries.

These entries

allow local logging only.

Example 2-1: The elcsd.conf File with Default Entries
#static

char

*sccsid

»@(#)elcsd.conf

elcsd

errlog

configuration

file
;

{
1

delimiter

l-local,2-togrem,4-remlog,5-remlog+priloglocal

errilog

DON'T

file

/usr/adm/syserr
/

backup

single

}
hosts

#remotel:S

remote

delimiter

log

:S
-

limit

erriog

errlog
user
#

remove

size

#
#

/usr/adm/syserr

(ULTRIX)
=«

DON'T

separate

(example)

comment

num.

dir.

path
dir.

remote
to

remove

file

2-4 Error Logger Management and

errors

blocks

path

hosts

log

out!

path

errlog

log

hostname

errlog

comment

from

remotes
remotel

Maintenance

dir.

path

out!
file
into

(together)
separate

file

2.3

elcsd.conf File

Parameters

Here are the parameters for the elcsd.conf file:
o

Status

Errorlog file size

Errorlog directory path

Backup errorlog directory path

Single-user errorlog directory path

Remote hosts errorlog directory path

Remote host name to log to

Hosts to log

The entries in Example 2-1 appear at the left-hand margin, and the
parameters are defined to the right.

The following sections discuss each of

the elcsd.conf file parameters.

2.3.1

Status

The first parameter in the elcsd.conf file is the status line, Whlch indicates

where error messages are logged.
The possible status conditions and their meanings are:
1-local

Log local messages here.

2-logrem

Log messages from remote systems here.

Log local and remote messages here.

4-remlog

Log messages from this system to a remote
system.

5-remlog+ priloglocal

= Log messages from this system to a remote

system, and log server as well as high priority
messages here.

When you log error messages between systems, you must decide how to
log the different priority levels of errors, based on the amount of disk
space you have available.
There are three priority levels: severe, high, and low.

A severe error

means the system is going down. High-priority errors are errors such as
recoverable machine checks and hard errors.

Low-priority errors include

soft errors, restarts and CRDs (corrected read data).

The default status condition is 1 (local).

Errors that occur on the local

system are logged on the local system.

Error Logger Management and

Maintenance 2-5

Enter 2 (logrem)

on the status line if you are a server system and you

want to log messages from a client system.

Enter 3 on the status line to log server (local)

and client (remote)

messages.

Enter 4 (remlog)

on the status line if you are a client system and you

want to log messages to a server.

Enter 5 (remlog+ priloglocal)

on the status line if you have sufficient disk

space and, in addition to logging all messages to a server, want to log

severe and high-priority messages on your client system.
No matter what status entry you specify, severe kernel messages are sent
to the local system console.

2.3.2

Errorlog File Size

The errorlog file size parameter defines the maximum size of every errorlog
file on the system. You can leave the errorlog file size parameter blank,
and the system will notify you when the file system is 98% full.
Otherwise, specify the maximum number of blocks (for example, 512 bytes)
that you want for each errorlog file.

When you do not specify the size of the errorlog file,

a message is written

to the errorlog status file, /usr/adm/eicsdlog, when the file system is 98%
full.

At this point, you should compress or back up and remove the

current errorlog file before the system stops logging messages to disk.

message is also written to the system log.

See the syslog(8)

command in

the ULTRIX Reference Pages.

2.3.3

Errorlog Directory Path

To start an errorlog file, you must specify a directory path in the
elcsd.conf file.

The errorlog directory path specifies the main errorlog file. The default

path is /usr/adm/syserr.
different directory.

You can, however, direct error messages to a

If you do, first verify that the alternate directory

exists and then specify this alternate when you invoke the uerf command.

2.3.4

Backup Errorlog Directory Path

The backup errorlog directory path is the file the system writes to when
the main errorlog file becomes full or when there is an error and the
system cannot access the main file.

There is no default path.

Consequently, you should specify a backup errorlog directory that is
different from the main errorlog directory.

2-6 Error Logger Management and

Maintenance

2.3.5

Single-User Errorlog Directory Path

The single-user errorlog file is /syserr.hostname. The root (/)

directory is

the default location for this file. You can direct single-user error messages
to a different directory, but you must be sure that the directory you
specify is mounted during single-user mode.
When the system makes the
transition from single-user mode to multiuser mode, errors logged in

Isyserr.hostname are appended to syserr.hostname in the multiuser errorlog
directory path and the single-user errorlog file is removed.

2.3.6

Remote Hosts Errorlog Directory Path

The remote hosts errorlog directory path is the directory for the errorlog
file containing messages logged from remote, or client systems. The default
path is /usr/adm/syserr.
It is a good idea to use the same path name that
you used for the main errorlog directory.

2.3.7

Remote

Host Name To Log To

This parameter specifies the name of the remote (server) system that you
are logging to.
This entry belongs in the client elcsd.conf file.
There is
no default remote host name.

2.3.8

Hosts To Log

The hosts to log parameter specifies the name of the remote (client)
system that
you are logging to, and specifies how to log the client’s
messages. The parameter takes two arguments:
°

The argument, S, sets up a separate errorlog file, syserr.hostname, for

each client system
o

2.3.9

The default argument, R, logs error messages from all client systems
to the syserr.remotes file.

Sample Server and Client elcsd.conf Files

The following examples show and describe the contents of a server
elcsd.conf file (see Example 2-2) and a client elcsd.conf file (see Example
2-3).
These examples are based on the following case:
Assume that you want to log error messages to a VAX server, piano, from
two MicroVAX clients, flute and violin.
You have decided to set up
individual errorlog files on the server for each remote client.
Plus, since
your client systems have adequate disk space, you decide to log severe and
high-priority error messages on each client system.

Error Logger Management and Maintenance 2-7

2.3.9.1

Server system file entries - To set up the required server files,

you edit the server’s elcsd.conf file as shown below.

Example 2-2: elcsd.conf File Entries for a Server System
##istatic

char

*sccsid

"@(#)elcsd.conf

(ULRIX)

»;

#
#

elcsd

errlog

configuration

file

{

delimiter

status

l-local,2-logrem,4-remlog,5-remlog+priloglocal

errlog

file

/usr/adm/syserr

DON'T
size
#

remote

}

delimiter

hosts

num.

dir.

out!

blocks

path

backup

errlog

dir.

path

single

user

errlog

dir.

path

log

remote

hosts

errlog

dir.

path

DON'T
:S

comment

hostname
log

|limit

errlog

/usr/adm/syserr

remove

log

remove

separate

comment
file

out!
R

remotes

file

(together)

flute:S
violin:S

If you examine the contents of this file, you will notice the following
entries:

;

The error logging status, 3, indicates that both server and client
messages are to be logged to the server, piano.

The errorlog directory path default, /usr/adm/syserr, is specified as the
main errorlog directory.

The remote hosts errorlog directory path default, /usr/adm/syserr, is
specified as the directory for error messages logged from remote

(client)
o

systems.

Both flute and violin are listed as the hosts to log.

Consequently,

the VAX server will log messages for each of these client systems.
Including the S argument with each host name (flute:S, for example)
specifies that each client will have its own errorlog file on the server.

These individual files will be subordinate to the remote hosts errorlog
directory, /usr/adm/syserr.

Therefore, error messages for flute will

reside on the VAX server at /usr/adm/syserr/syserr.flute.

Similarly,

error messages for violin will reside on the VAX server at

/usr/adm/syserr/syserr.violin.

2-8 Error Logger Management and

Maintenance

2.3.9.2

Client system file entries -

To set up the required client files,

you edit the elcsd.conf file for both flute and violin.

The following example

shows the client file entries.

Example

2-3: elcsd.conf File

fistatic

char

*sccsid

Entries for a

Client System

»@(#)elcsd.conf

(ULTRIX)

#
#

elcsd

errlog

configuration

file

{

delimiter

status

1-local,2-logrem,4-remlog,5-remlog+priloglocal

DON'T

errlog

file

/usr/adm/syserr

remove

size
#

(imit

errlog

dir.

path

user

errlog

dir.

path

remote

hosts

errlog

dir.

path

remote

delimiter

log

blocks

backup

piano

path

single

log

hosts

num.

dir.

out!

#
#

}

comment

#
/usr/adm/syserr

hostiname

DON'T

separate

remove
file

log

or
:R

comment
-

out!

remotes

file

(together)

If you examine the contents of this file, you will notice the following
entries:

The error logging status, 5, specifies that all error messages are

logged to the server, and that severe and high-priority error messages
are logged to the client as well
e

The errorlog directory path default, /usr/adm/syserr, is specified as the
main errorlog directory.

The remote hosts errorlog directory path default, /usr/adm/syserr, is
specified as the directory for error messages logged from remote

(server)
o

systems.

The server, piano is listed as the remote hostname to log to.

Consequently, the client will log messages to this server system.

Error Logger Management and

Maintenance 2-9

2.4

Error Logger File

Maintenance

This section discusses the files related to the error logger and explains how
to

maintain them.

It describes the elcsdlog file and explains how to

backup and restore an errorlog file.

2.4.1

The elcsdlog File

The elcsdlog file contains status messages about the elscd daemon.
resides in the /usr/adm directory.

Critical status messages in this file are

also logged by the syslog utility to the system log and to the terminal of
any user entered in the syslog.conf file.

See syslog(8)

in the ULTRIX

Reference Pages for information about this program.

Critical messages should be acted upon as soon as possible or the system
may become unable to log error messages.

The elcsdlog file is cleared

when you reboot the system.

2.4.2

Backing up the

Erroriog File

You should periodically back up or delete the errorlog files to free up disk
space.

How often you do this depends on the amount of disk space

available and the size of your file system.

However, it is good practice to

back up the errorlog file at least once a month for a MicroVAX system

and at least once every two months for a VAX system. See Guide to
System Backup and Restore for information on backing up files as part of
normal system maintenance.

Note

When you do backups of any kind, be sure that the elcsd
daemon is running.
#

-ax

Do this with the command:

grep

elcsd

If the elcsd daemon is not running, start it manually with the
- S option of the eli command for single-user mode, or the - e
option for multi-user mode.

2.4.3

Clearing an Errorlog File

In addition to backing up your errorlog files, you can clear individual files

2-10 Error Logger Management and

Maintenance

periodically.

The following steps describe the procedure:

Note

You do not need superuser privileges for every command shown
below.

It makes the task easier, however, if you simply log on

and work as root until you complete the entire procedure.
1.

Change your working directory to the main errorlog directory and get
a listing of the contents.
#

¢cd

For example, type:

/usr/adm/syserr;ls

Locate the errorlog file you intend to clear.

For example, if the

hostname is piano, you are looking for the file named syserr.piano.
3.

Copy the errorlog file to a new location, and rename the file. For
example, type:
#

syserr.piano

/usr/adm/copies/syserr.old.piano

You now have two copies of the errlog file.

The error logger

program will continue to log error messages to the original file,

/usr/adm/syserr/syserr.piano but not to the copy,

/usr/adm/copies/syserr.old.piano.
Note
Do not use the mv command to save the errorlog file,
/usr/adm/syserr.hostname, or the system will continue to log
error messages to that file in its new location.

You can compress the saved file to save disk space.

the compact(1)

See

command in the ULTRIX Reference Pages

for details.

If you want the data in the saved file to remain accessible, enter the
uerf command with the —f option and the full pathname of the saved

file.

For example, type:
#

uerf

—f

/usr/adm/copies/syserr.old.piano

In response, the error logger displays the contents of the saved file.
4,

Remove the original errorlog file, /usr/adm/syserr/syserr.piano with the
command:

Error Logger Management and

Maintenance 2-11

/usr/adm/syserr/syserr.piano

To re-enable error logging, type:
#

eli

-f

-e

When you use the - f option, the system suppresses the prompts you
usually see when you select the — e option. The elcsd daemon creates

a new errorlog file, at /usr/adm/syserr/syserr.hostname.

2.5

Crucial Error Logger System Messages

There are three crucial messages that indicate a problem in the error
logging procedure:
1.

When the kernel identifies an overflow of its errorlog buffer, it
generates the following error message which is sent to the system
console every 15 minutes:
Errlog

Buffer

Full:

<n>

Missed

Error(s)

If you receive this message from the kernel, you can respond by
typing the ps command to see if the elcsd daemon is running.

If it is not running, restart it using the eli command with the - e
option.

If you are in single-user mode, use the eli command with the

- S option.

If the elcsd daemon is running, then the current system tasks may

be causing a high volume of errors, which the elcsd daemon cannot
move from the kernel errorlog buffer to the errorlog file quickly
enough.

You may want to suppress the printing of this message

using the eli command with the — q option until these tasks have
finished.
When the space needed for error messages is about to exceed the
limit set in the elcsd.conf file, the system writes the following
message to the /etc/syslog.conf and /etc/elcsdlog files:
exceeded

errlog

file

Iimit

size,

error

not

logged

disk!

You can respond to this problem as follows:

Save and then remove the current errorlog file.

See the previous

section, Clearing an Errorlog File, for instructions on how to do this.
Increase the size limit in the elcsd.conf file. You should set the limit
based on the disk device and the size of the file system. See the
previous section, Errorlog File Size, for details.

2-12 Error Logger Management and

Maintenance

Reconfigure the elcsd.conf file by using the eli command with the - r

option.
When the system is about to run out of file system space and is no
longer logging error messages to disk, it writes the following message
to the /etc/syslog.conf and /etc/elcsdlog files:
file

system

almost

full,

CAN'T

log

errors

/usr/adm/syserr/syserr.hostname

This message can occur when the errorlog file size limit in the
elcsd.conf is not specified.

You can respond to this problem as

follows:
Find out why the file system is full

It may be necessary to remove

or compress other files on the disk or to save and then remove the
current errorlog file.

See the previous section, Clearing an Errorlog

File, for more information on how to perform these tasks.
Consider setting the errorlog file size limit in the elcsd.conf file. You
should set the limit based on the disk device and the size of the file
system. See the previous section, Errorlog File Size, for details.
If you change the elcsd.conf file, you must reconfigure it by using

the eli command with the —r option.

Error Logger Management and Maintenance 2-13

The Error Report Formatter 3

The ULTRIX error report formatter, uerf, reports errors and events that
occur on your MicroVAX or VAX system.

The uerf program accesses the

error messages and events stored in the errorlog file, translates them from

binary code to ASCII, and sends them to the output device you specify.
The uerf program uses the data files, uerf.bin, uerf.hlp and uerf.err.

The

uerf.bin file is the event information data base, uerf.hlp is the help file,

and uerf.err is the error message file.

The uerf program searches sequentially for the uerf data files. It looks:
1.

In the directory specified when using a full pathname with the uerf

command.
2.

In the /etc directory.

In the directories specified in your shell PATH environment variable.

The uerf command has options that let you generate various types of
reports on specified errors.

For example, you can produce a report

containing all the error messages in the errorlog file, or you can use
options to produce a report with only a few error types. In relation to

these options, this chapter provides information on:
°

Generating reports

Selecting error types

Selecting specific errors

For your convenience, this chapter also contains a quick reference for uerf
options.
Note

You do not need superuser privileges to use the uerf command.

3.1

Generating Reports

The uerf program produces reports based on the error messages stored in
an errorlog file.

Using the uerf command with its options, you can

produce error reports to a terminal screen, to a file, or to a printer.

For

example, to generate an error report and display it on your terminal one
screenful at a time, type:
#

/etc/uerf

By looking at the types and number of errors and events, you can tell how
reliable the system is. For example, if a report shows a large number of
errors for a particular device, you can call
representative before the device fails.

a DIGITAL service

Furthermore, when a failure does

occur, the error report provides information on the events that led up to

the failure.
The error logger system records and reports the following errors and
events:

Errors —

Device errors, device timeouts, machine checks, bus errors,

and memory errors, such as hard or soft error correcting code (ECC)
errors

Events —

System startup and shutdown messages, system failures or

panics, and informational messages

To see what options are available with the uerf command, use the help
option:
#

/etc/uerf

-h

This command displays a list of the available options and their meanings.
Example 3-1 shows the uerf help display.

The rest of this section explains the uerf options. These options enable you
to:

Format reports

Generate reports from specific files

Generate reports from specific host systems

Report on errors as they occur

Generate reports in reverse chronological order

Generate reports in single-user mode

Generate summary reports

For a detailed list of options in quick reference format, refer to the uerf
Option Reference in Section 3.5 or see uerf(8) in the ULTRIX Reference

Pages.

3-2 The

Error Report Formatter

Example

3-1: The

Help Display

UERF

uerf<cr>
PROCESSING

input_filename

-n

-H

all

events

help

ascii_host_name

-R

default

input

file

specified

input

file.

display

this

process

events

from

screen.
the

process

events

from

kernel

process

events

for

the

process

events

from

-t

time_range

process

events

-0

output_format

format

the

output

display

summary

- X

exclude

selected

-Z

display

the

entry

selected

OPTIONS

Events

may

errorlog

specified

the

-S

SELECTION

the

OPTIONS

-h
-f

process

HELP

brief,

selected

events

reverse

terse

from

format.

output.

format

follows.

-A

adapter_list

-D

disk_devices

-M

mainframe_type

-0

os_specific

-T

tape_devices

-¢c

class_of_events

-e

error_types

-r-record

-S

sequence

see

uerf(8)

***¥E*

3.1.1

For

detailed

type

list

order.

range.

events.

(below)

hex

file
time

full

immediately.

hosts.

erroriog

specified

buffer

options

number

*#*x*=*

Formatting Reports

Use the -0 option to format uerf error reports.

This option enables you

to format reports in three ways:

brief

Reports event information in a short format

full

Reports all available information for each error

terse

Reports binary event information and displays register values

and other ASCII messages in a condensed format
This option requires a parameter. If you do not select the —o option, the
default output format is brief.
Some errors, such as ASCII messages, produce the same information
whether you use the full format or the brief format.
provide more information in the full format are:

Errors that do

machine checks, panics,

memory errors, MSCP disk errors, and TMSCP tape errors.

The

Error

Report Formatter 3-3

The brief output format is shown in Example 3-2 for this command line:
#

/etc/uerf

-M

mem

Example 3-2: Memory Error in Brief Format
2

—————

EVENT

CLASS

EVENT

SEQUENCE

INFORMATION

SYSTEM

-----

EVENT

MEMORY

ERROR

ULTRIX

Thu

SYSTEM

May

18:14:46

1987

EDT

guitar

PROCESSOR

ok e ko

SYSTEM

OCCURRED/LOGGED

e sk ok ok ok ok ok

-----

101.

NUMBER

e d ok sk s ol ok ok sk sde

ERROR

TYPE

OPERATING
OCCURRED

SEGMENT

s o

x01845106
TYPE

UNIT

KA780

INFORMATION

-----

UNIT

CLASS

MEMORY

UNIT

TYPE

MS780E

ERROR

SYNDROME

-----

780-E

CSR

MEMORY

REGS

CRD

ERROR

-----

CSRA

x00101E6C

INTERLEAVE:

RAM:

ADAPTER

CSRB

x00001000

SIZE:

ERROR

SUMMARY

DIAG

ECC

HAS

START

CSRC

x148F020E

CODE:

MEMORY

MEM
ERR

INTERNAL

2-WAY

64K
x3

15.

BITS:

«xO

VALID

DATA

ADDR:

SYNDR/CHK

xO
BITS:

«xE

CRD

CSRD

x080AD002

ERROR

ADDR:

ERROR

LOG

ERR

SYNDR/CHK

ERROR

3-4 The Error Report Formatter

x91EO

REQUEST

ADDR:

BITS:

x1015A

To show the full output format for this report, which displays all memoryrelated mainframe errors, type:
#

/etc/uerf

—-o

full

-M

mem

In addition to the information shown in Example 3-2, the full output
format produces this information:
—————

CNTRLR
NO.

ADDITIONAL

MEMORY

INFO

ERRS

-----

1.
ON

THIS

ADD

To produce the terse output format for all memory-related mainframe
errors, type:
#

/etc/uerf

-o

terse

-M

mem

This report gives you binary event information, for the events you specify.
The terse memory-related mainframe errors report looks like this:
MEMORY

ERROR:

MEMORY

CRD

KA780

ERROR

MS780E

CSRA

x00101E6C

CSRB

x00001000

CSRC

x00000000

CSRD

‘

CNTRLR
NO.

3.1.2

x1A0AD275

ERRS

1.
ON

THIS

ADD

Generating Reports from Specific Files

The file name (-f) option of uerf selects errors from the errorlog file you
specify. For example,
#

/etc/uerf

-f

/usr/adm/syserr/syserr.guitar.old

Use this option to look at old or backup errorlog files rather than at the
default file /usr/adm/syserr/syserr.hostname. You can also use this option to
access the single-user errorlog file, /syserr.hostname, in the root directory.
Do not use the -n

Be sure to specify the full path name for the file.
option with the —f option.

You must specify a parameter with the —f option or you will get an invalid
parameter message.

The

Error Report Formatter 3-5

3.1.3

Generating Reports from Specific Host Systems

The host (-H) option of uerf selects errors for a specific host system from
the /usr/adm/syserr/syserr.hostname or /usr/adm/syserr/syserr.remotes file.
Consequently, you must specify a host name as a parameter.
This option
i1s useful when you are logging errors from several remote systems to the
errorlog file /syserr.remotes on the host.
See Chapter 2 for information on
remote logging.

3.1.4

Reporting Errors As They Occur

The now (-n) option of uerf reports errors to the terminal as they occur,
before they are logged to the errorlog buffer. Here is the command line:
#

/etc/uerf

-n

You can use this option when you run disk or tape exercisers.
use the —f option with this option.

3.1.5

Do not

Generating Reports in Reverse Chronological Order

To generate reports in reverse chronological order, use the ~R option of
As shown in the following example, you can select the
starting point of the report by combining -R with other uerf options.

the uerf command.

/etc/uerf

-R

300

In response to this command, the uerf program produces a report that lists
all startup messages, beginning with the most recent.
3.1.6

Generating Reports in Single-User Mode

You can run uerf in single-user mode.
However, if the errorlog file and
the uerf data files, uerf.bin, uerf.hlp and uerf.err, are not located in the
root directory, you must first mount the file systems on which they reside.
In addition, the line printer spooler is not operational during single-user
mode.
Therefore, to print an error report to a line printer while in singleuser mode, you must direct the output to a printer special file, such as:
# /etc/uerf

3.1.7

-f

/syserr.hostname

/dev/Ilp

Generating Summary Reports

The summary (-S)

option of uerf produces a summary of the errorlog file,

or a summary of those records that you select.
support summaries.

All uerf selection options

The default format for summary output is terse.

However, you can generate summary output in full or brief format by
including the output (-0) option and specifying the desired format.

3-6 The

Error Report Formatter

The following example shows the command that you would type to generate
a terse summary of all errors recorded for the day in the errorlog file:
#

/etc/uerf

-t

s:00

-S

Or, to generate a full summary of the same information, you would type:
#

3.2

/etc/uerf

-t

s:00

-S

-0

full

Selecting Error Types

The uerf command allows you to generate error reports for the following:
°

Adapter errors

Class errors

Disk errors

Mainframe processor errors

Operating system errors

Tape errors

For each of these error types, the uerf command has a corresponding
option.
Each of the options has corresponding report parameters that you
can specify to generate various types of reports.
The followmg sections explain how to select each of these error types.

3.2.1

Selecting Adapter Errors

The -A option selects errors for the adapter parameters that you specify.
The adapter parameters the error report formatter supports are:
aie

BVP-type controller

aio

BVP-type controller

bla

BI LESI adapter

bua

BI UNIBUS adapter

nmi

Nautilus memory interconnect

uba

VAX UNIBUS adapter

The following example shows how to generate a report for uba (VAX
errors that have been logged in the errorlog file:

UNIBUS)

The Error Report Formatter 3-7

/etc/uerf

—-A

uba

To generate a report for more than one parameter, use the comma (,)

a delimiter.

For example, the following command generates a report for all

uba and nmi adapter errors:
#

/etc/uerf

3.2.2

—-A

uba,nmi

Selecting Class Errors

The -c option selects errors for the class event that you specify.

The

class parameters that the error report formatter supports are:
err

Hardware-detected and software-detected error conditions

oper

Operational events, such as device status or error messages,
time changes, and system startup and shutdown messages

maint

Events that occur during system maintenance, such as running

the on-line functional exercisers

3.2.3

Selecting Disk Errors

The —-D option selects errors for the MSCP disk parameters you specify.

See the ra(4)

command in the ULTRIX Reference Pages for the disk

types that the error report formatter supports.

3.2.4

Selecting Mainframe Errors

The —M option selects these types of processor errors for the systems for
which you are logging errors:
cpu
mem

CPU-related errors, such as machine checks
memory-related errors, such as single-bit CRD (corrected read

data)

and double-bit uncorrectable errors

When you do not specify any parameters, all mainframe errors are
reported.

The following command produces a report containing CPU-related mainframe
errors for the system:
#

3-8 The

/etc/uerf

-M

cpu

Error Report Formatter

3.2.5

Selecting Operating System Errors

The -O option selects errors for the operating system parameters you
specify.
When you do not specify any parameters, all operating system
events are reported.
The operating system parameters that the error
report formatter supports are:

3.2.6

aef

Arithmetic exception faults

ast

Asynchronous trap exception faults

bpt

Breakpoint instruction faults

cmp

Compatibility mode faults

pag

Page faults

pif

Privileged instruction faults

pro

Protection faults

ptf

Page table faults

raf

Reserved address faults

rof

Reserved operand faults

scf

System call exception faults

seg

Segmentation faults

tra

Trace exception faults

xfc

Xfc instruction faults

Selecting Tape Errors

The -T option selects errors for the TMSCP tape parameters that you
specify. See the tms(4)

command in the ULTRIX Reference Pages for the

tape types that the error report formatter supports.

3.3

‘

Selecting Specific Errors

Some uerf options let you select the following errors specifically:
o

Errors specified by record number

Errors specified by sequence number

Errors within a specified time range

Errors except the ones that you specifically exclude

The following subsections describe the uerf options that let you select or

exclude specific errors.

The Error Report Formatter 3-9

3.3.1

Selecting Errors by Record Type

The —r option selects errors for specific record types.

Consequently, you

must specify a parameter that defines which record type you want the
program to select. Use this option to access errors, such as ASCII

messages, which are not accessed by other uerf options. The -r option also
offers an alternate way to specify some error events, such as disk and tape
errors.

Devices, such as UNIBUS communications devices, log to the errorlog file
in an ASCII rather than binary message format.
Note

It is possible that the text of an ASCII error message, when
reported in the brief or the full output format, will generate more
than one error message. If another error occurs and is reported

at approximately the same time, these ASCII text messages may
not print out consecutively.

You can use the terse output format

to see such messages as one unit.

The record parameters supported by the error report formatter are:
Hardware-Detected Errors
100

machine check

101
102

memory corrected read data/read data substitute (crd/rds)
disk errors

103

tape errors

104

device controller errors

105

adapter errors

106

bus errors

107

stray interrupts

108

asynchronous write errors

109

exceptions/faults

112

stack dump

113

ka650 error and status registers

114

6200 vector 60

115

6200 vector 54

Software-Detected Errors
200

panics (bug checks)

201

ci ppd info

3-10 The

Error Report Formatter

Informational ASCII Messages
250

informational

251

8600/8650 snapshot taken

Operational Messages
300

startup

301

shutdown

310

time change

350

diagnostic information

351

repair information

The following example produces all system startup messages, including
hardware devices configured and their CSR (control status register)
addresses:
#

/etc/uerf

-r

300

The following example outputs all of the operational messages - startup,
shutdown, time change, and diagnostic - within the record option:
#

3.3.2

/etc/uerf

-r

300-350

Selecting Errors by Sequence

Number

The -s option selects specific errors from the errorlog file.

The selection

corresponds to the sequence number given as a parameter to the -s
option.

Consequently, you must specify the sequence number. A sequence

number is assigned to an event when it is logged to the errorlog file. You
can use this option to output specific errors after viewing the errorlog file
at your terminal.

Note

Sequence numbers are restarted when the system is rebooted.

the errorlog file contains error messages from before and after a

reboot, there may be duplicate sequence numbers in the file.

3.3.3

Selecting Errors Using a Time Range

The -t option selects errors for the time period you specify. The format
for the -t option is:

/etc/uerf

-t

s:ddd-mmm-yyyy,hh:mm:ss e:dd-mmm-yyyy hh:mm:ss

The Error Report Formatter 3-11

You must specify a start date or time or an end date or time when you
use this option. However, you can abbreviate the format using the following
default parameters:
®

The current date is the default date when none is specified.

The default start time is 00:00:00.

The default end time is 23:59:59.

When you do not use this option, the uerf command processes the entire
errorlog file.
The parameters for the -t option are:
s

Specifies the start date and time

Specifies the end date and time

Day

mmm

Month

yyyy

Year

Hour

Minute

Second

The following example shows how to produce an error report containing all
errors for the 24-hour period of October 23, 1988:
#

/etc/uerf

-t

s:23-0ct-1988,00:00:00

e:23-0ct-1988,23:59:59

The following command produces errors from the beginning of the errorlog
file until February 29 of the current year; the default end time is 23:59:59:
#

3.3.4

/etc/uerf

-t

e:29-feb

Excluding Error Types

The -x option excludes the error types that you specify from the error
report.

Although this option works with most other options, you cannot

exclude the —f option, the —h option, the —H option, the —n option, the -0
option, the -t option, or the —R option. The following command reports all
errors except disk errors and operating system events:
#

/etc/uerf

-0

-x

3-12 The Error Report Formatter

full

-D.

3.4

Using uerf Options

This section provides some notes and examples of uerf use.

See Table 3-1

for a quick reference of the command options.
°

Be sure to type options in the correct case since the uerf command
has both uppercase and lowercase options.
output format, while

Options can be used together.
#

/etc/uerf

-A

For example, -0 specifies

— O specifies operating system events.

uba,nmi

For example:
-f

syserr.guitar.old

-TH

guitar

The preceding command produces an error report from the file
syserr.guitar.old for the system guitar.

The report contains adapter

errors for the VAX UNIBUS adapter and the Nautilus memory
interconnect, as well as TMSCP (Tape Mass Storage Control
Protocol)

device errors of all types.

In the following example, the time and output options produce error

messages for the current day in a terse format at your terminal:
#

3.5

/etc/uerf

-t

s:00

-0

terse

uerf Option Reference

Table 3-1 lists each uerf option alphabetically and provides brief
descriptions and examples.

Table 3-1: The uerf Command Options

Option
—~ A adapters

Usage

Examples

Selects adapter and device

/etc/uerf — A bua,nmi

controller errors.
- ¢ classes

Selects classes of errors.

/etc/uerf —c err

.
- D disks

.
)
Selects mscp disk devices.

/etc/uerf - D rx
Jetc/uerf — D ra60 rd31

Specifies the file from which

/etc/uerf - f

error messages are read.

/usr/adm/syserr/syserr.old

- f filename

Use full pathname.

Do not

use with the - n option.

The

Error Report Formatter 3-13

Option
-h

Usage

Examples |

Displays help message.

/etc/uerf — h

not use with another option.

- H hostname

Selects errors for specified

/etc/uerf - f

system.

/usr/adm/syserr/syserr.remotes

Use for remote

logging to syserr.remotes.
-~ M mainframe

Specifies processor error

/etc/uerf - M mem

types.

/etc/uerf

Displays errors as they

-1

- H guitar

occur.

— M cpu,mem

/ete/uerf —n - D

Do not use with - f

option.

— 0 output

Selects output format for

/etc/uerf — o brief

report.

/ete/uerf — o full

Default is brief

format.

- O operating
system

-~ R reverse

chronological

order

Selects operating system

/etc/uerf - O

errors.

/etc/uerf — O raf,ptf,ast

Selects records in reverse
chronological order.

/etc/uerf — A bua - R

/etc/uerf — R
letc/uerf - R - O

Selects errors for specified

/etc/uerf —r 300,301

record codes.

/etc/uerf —r 100-109

sequence

Selects errors for specified

lete/uerf — s 750-800

numbers

sequence numbers.

—r records
- §

/etc/uerf — o terse

S
-t time

- T tapes

Produces summary report.

/etc/uerf -S -0 brief

Selects errors within

/etc/uerf -t s:13-apr-1986

specified time range.

/etc/uerf -t s:13-apr,18:30

Selects errors for tmscp

/ete/uerf — T

tape types.

/etc/uerf - T tu8l

Excludes specified options.

/etc/uerf - A -x - D

You cannot exclude - f,

-H, -h, -n, -0, -t, or
— R options.
Displays entry in hex
format.

3-14 The Error Report Formatter

/etc/uerf — 7

Index

error logger (cont.)

defined, 1-1
client

defined,

description, 2-1 to 2-8

1-1

disabling, 2-1

configuration file (error logger)
See elcsd.conf file

enabling in multi-user mode, 2-2
enabling in single-user mode, 2-2
error message types, 2-5

error messages, 2-12

errors reported, 3-1

events reported, 3-1

elcsd daemon
disabling,

file maintenance, 2-10

1-2

local system crash and, 1-3

enabling, 1-2

remote system crash and, 1-4

rc file entry, 1-3

remote system shutdown and,

elesd.conf file
changing entries, 2-3
contents,

1-2

for client system, 2-9e
for server system, 2-3, 2-8e
parameters, 2-5 to 2-8

reconfiguring, 2-3 to 2-8
specifying client system, 2-7
specifying server system, 2-7
with default entries, 2-4e

elcsdlog file
contents, 2-10

directing to terminal, 3-1

displaying as errors occur, 3-6
excluding error types, 3-12
for adapter errors, 3-7
for class errors, 3-8
for disk errors, 3-8

for error type, 3-7 to 3-9
for operating system, 3-9
for record type, 3-10

options, 2-1 to 2-3

for sequence number, 3-11

error logger
1-1

ASCII messages and, 3-10n

brief format, 3-4e

for mainframe errors, 3-8

eli command

components,

error report

1-2

controlling manually, 2-1 to 2-3

for specific errors, 3-9 to 3-12
for tape errors, 3-9
for time period, 3-11

1-4

error report (cont.)

uerf command (cont.)

formatting, 3-3 to 3-5

help option, 3-2

from specific file, 3-5

option list, 3-14t

from specific systems, 3-6

options, 3-13 to 3-14

full format, 3-5e¢

record parameters supported, 3-10

generating, 3-1 to 3-6

using, 3-1 to 3-14

generating in reverse order, 3-6
generating in single-user mode, 3-6
specifying brief format, 3-4

specifying full format, 3-4
specifying terse format, 3-5

terse format, 3-3
error report formatter

See uerf command
errorlog buffer

clearing, 2-3, 1-2
errorlog file

backing up, 2-10
clearing, 2-10
logging message to, 2-3

move command and, 2-11n
remote system directory path, 2-7

remote systems and, 2-5
single-user directory path, 2-7
specifying backup, 2-6

specifying directory path, 2-6
specifying size, 2-6
event file

See errorlog file

S
server

defined,

1-1

U
uerf command
data files, 3-1
help display, 3-3e

Index-2

HOW TO ORDER ADDITIONAL DOCUMENTATION

DIRECT TELEPHONE ORDERS
In Continental USA

in Canada

and New Hampshire,
Alaska or Hawaii
call 800-DIGITAL

call 800-267-6215

DIRECT MAIL ORDERS (U.S. and Puerto Rico*)
DIGITAL EQUIPMENT CORPORATION

P.O. Box CS2008
Nashua, New Hampshire 03061

DIRECT MAIL ORDERS (Canada)
DIGITAL EQUIPMENT OF CANADA LTD.

100 Herzberg Road
Kanata, Ontario K2K 2A6
Attn: Direct Order Desk

INTERNATIONAL
DIGITAL EQUIPMENT CORPORATION

PSG Business Manager
c¢/o Digital’s local subsidiary
or approved distributor

Internal orders should be placed through the Software Distribution Center (SDC), Digital
Equipment Corporation, Westminster, Massachusetts 01473
*Any prepaid order from Puerto Rico must be placed
with the Local Digital Subsidiary:
809-754-7575

ULTRIX - 32
Guide to the Error
Logger System

AA-MEQ5A-TE

Reader’'s Comments

Note: This form is for document comments only. DIGITAL will use comments
submitted on this form at the company’s discretion. If you require a written reply and are eligible to receive one under Software Performance
Report (SPR) service, submit your comments on an SPR form.

Did you find this manual understandable, usable, and well-organized? Please

make suggestions for improvement.

Did you find errors in this manual? If so, specify the error and the page number.

Please indicate the type of user/reader that you most nearly represent.

Assembly language programmer

Higher-level language programmer

Occasional programmer (experienced)

User with little programming experience

(]

Student programmer

Other (please specify)

Name

Date

Organization

Street

City

State

lemq ode
Country

United States

BUSINESS REPLY MAIL
FIRST CLASS PERMIT NO.33 MAYNARD MASS.

POSTAGE WILL BE PAID BY ADDRESSEE

Digital Equipment Corporation
Documentation Manager
ULTRIX Documentation Group
ZKO3-3/X18

Spit Brook Road
Nashua, N.H.

s
|
|
|
1

H
|
|
|

|
|

|
|
|

|
i
|
|
I
|
|
l

:
x
l

[
|

r
|
|
l
|
a
|
|

-]
o

b ]

[ d

03063

Necessary
if Mailed in the

Cut Along Dotted Line

No Postage