Digital PDFs

AA-LY20B-TE

June 1990

439 pages

Original

15MB

Document:	ULTRIX Reference Pages Section 8: Maintenance
Order Number:	AA-LY20B-TE
Revision:	0
Pages:	439
Original Filename:

OCR Text

ULTRIX

Reference Pages Section 8: Maintenance

Order Number: AA-LY20B-TE
June 1990

Product Version:

ULTRIX Version 4.0 or higher

This manual describes commands for system operation and maintenance for system
administrators on both RISe and VAX platforms.

digital equipment corporation
maynard, massachusetts

Restricted Rights: Use, duplication, or disclosure by the U.s. Government is subject to restrictions as set forth in
subparagraph (c) (1) (ii) of the Rights in Technical Data and Computer Software clause of DFARS 252.227-7013.

Portions of the information herein are derived from copyrighted material as permitted under license agreements with
AT&T and the Regents of the University of California. © AT&T 1979, 1984. All Rights Reserved.
Portions of the information herein are derived from copyrighted material as permitted under a license agreement with
Sun MicroSystems, Inc. © Sun MicroSystems, Inc, 1985. All Rights Reserved.
Portions of this document © Massachusetts Institute of Technology, Cambridge, Massachusetts, 1984, 1985, 1986,
1988.
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:

Illmaala
CDA
DDIF
DDIS
DEC
DECnet
DEC station

DECUS
DECwindows

DTIF
MASSBUS
MicroVAX
Q-bus
ULTRIX
ULTRIX Mail Connection

ULTRIX Worksystem Software
UNIBUS
VAX
VAXstation
VMS
VMS/ULTRIX Connection
VT
XUI

Ethernet is a registered trademark of Xerox Corporation.
Network File System and NFS are trademarks of Sun Microsystems, Inc.
POSIX is a registered trademark of the Institute of Electrical and Electronics Engineers.
PostScript and Display PostScript are registered trademarks of Adobe Systems, Inc.
System V is a registered trademark of AT&T.
Tektronix is a trademark of Tektronix, Inc.
Teletype is a registered trademark of AT&T in the USA and other countries.
UNIX is a registered trademark of AT&T in the USA and other countries.

About Reference Pages

The ULTRIX Reference Pages describe commands, system calls, routines, file
formats, and special files for RISe and VAX platforms.

Sections
The reference pages are divided into eight sections according to topic. Within each
section, the reference pages are organized alphabetically by title, except Section 3,
which is divided into subsections. Each section and most subsections have an
introductory reference page called intra that describes the organization and
anything unique to that section.
Some reference pages carry a one- to three-letter suffix after the section number, for
example, scan(1mh). The suffix indicates that there is a "family" of reference
pages for that utility or feature. The Section 3 subsections all use suffixes and other
sections may also have suffixes.
Following are the sections that make up the ULTRIX Reference Pages.

Section 1: Commands
This section describes commands that are available to all ULTRIX users. Section 1 is
split between two binders. The first binder contains reference pages for titles that fall
between A and L. The second binder contains reference pages for titles that fall
between M and Z.

Section 2: System Calls
This section defines system calls (entries into the ULTRIX kernel) that are used by
all programmers. The introduction to Section 2, intra(2), lists error numbers with
brief descriptions of their meanings. The introduction also defines many of the terms
used in this section.

Section 3: Routines
This section describes the routines available in ULTRIX libraries. Routines are
sometimes referred to as subroutines or functions.

Section 4: Special Files
This section describes special files, related device driver functions, databases, and
network support.

Section 5: File Formats
This section describes the format of system files and how the files are used. The files
described include assembler and link editor output, system accounting, and file
system formats.

Section 6: Games
The reference pages in this section describe the games that are available in the.
unsupported software subset. The reference pages for games are in the document
Reference Pages for Unsupported Software.

Section 7: Macro Packages and Conventions
This section contains miscellaneous information, including ASCII character codes,
mail addressing formats, text formatting macros, and a description of the root file
system.

Section 8: Maintenance
This section describes commands for system operation and maintenance.

Platform Labels
The ULTRIX Reference Pages contain entries for both RISC and VAX platforms.
Pages that have no platform label beside the title apply to both platforms. Reference
pages that apply only to RISC platforms have a "RISC" label beside the title and the
VAX-only reference pages that apply only to VAX platforms are likewise labeled
with "VAX." If each platform has the same command, system call, routine, file
format, or special file, but functions differently on the different platforms, both
reference pages are included, with the RISC page first.

Reference Page Format
Each reference page follows the same general format. Common to all reference pages
is a title consisting of the name of a command or a descriptive title, followed by a
section number; for example, date(l). This title is used throughout the
documentation set.
The headings in each reference page provide specific information. The standard
headings are:
Name

Provides the name of the entry and gives a short description.

Syntax

Describes the command syntax or the routine definition. Section 5
reference pages do not use the Syntax heading.

Description

Provides a detailed description of the entry's features, usage, and
syntax variations.

Options

Describes the command-line options.

Restrictions

Describes limitations or restrictions on the use of a command or
routine.

Examples

Provides examples of how a command or routine is used.

iv About Reference Pages

Return Values

Describes the values returned by a system call or routine. Used in
Sections 2 and 3 only.

Diagnostics

Describes diagnostic and error messages that can appear.

Files

Lists related files that are either a part of the command or used
during execution.

Environment

Describes the operation of the system call or routine when
compiled in the POSIX and SYSTEM V environments. If the
environment has no effect on the operation, this heading is not
,used. Used in Sections 2 and 3 only .

. See Also

Lists related reference pages and documents in the ULTRIX
documentation set.

Conventions
The following documentation conventions are used in the reference pages.

The default user prompt is your system name followed by a right
angle bracket. In this manual, a percent sign (% ) is used to
represent this prompt.

A number sign is the default superuser prompt.

user input

This bold typeface is used in interactive examples to indicate
typed user input.

system output This typeface is used in text to indicate the exact name of a
command, routine, partition~ pathname, directory, or file. This
typeface is also used in interactive examples to indicate system
output and in code examples and other screen displays.
UPPERCASE
lowercase

The ULTRIX system differentiates between lowercase and
uppercase characters. Literal strings that appear in text,
examples, syntax descriptions, and function definitions must be
typed exactly as shown.

rlogin

This typeface is used for command names in the Syntax portion
of the reference page to indicate that the command is entered
exactly as shown. Options for commands are shown in bold
wherever they appear.

filename

In examples, syntax descriptions, and routine definitions, italics
are used to indicate variable values. In text, italics are used to
give references to other documents.

[ ]

In syntax descriptions and routine definitions, brackets indicate
items that are optional.

{ I}

In syntax descriptions and routine definitions, braces enclose lists
from which one item must be chosen. Vertical bars are used to
separate items.

About Reference Pages v

In syntax descriptions and routine definitions, a horizontal ellipsis
indicates that the preceding item can be repeated one or more
times.

A vertical ellipsis indicates that a portion of an example that
would normally be present is not shown.
cat(l)

Cross-references to the ULTRIX Reference Pages include the
appropriate section number in parentheses. For example, a
reference to cat(1) indicates that you can find the material on the
cat command in Section 1 of the reference pages.

Online Reference Pages
The ULTRIX reference pages are available online if installed by your system
administrator. The man command is used to display the reference pages as follows:
To display the Is(1) reference page:
% man

To display the passwd(1) reference page:
% man passwd

To display the passwd(5) reference page:
% man 5 passwd

To display the Name lines of all reference pages that contain the word "passwd":
% man -k passwd

To display the introductory reference page for the family of 3xti reference pages:
% man 3xti intro

Users on ULTRIX workstations can display the reference pages using the
unsupported xman utility if installed. See the xman(1X) reference page for details.

Reference Pages for Unsupported Software
The reference pages for the optionally installed, unsupported ULTRIX software are in
the document Reference Pages for Unsupported Software.

vi About Reference Pages

intro (8)
Name
intro - introduction to system maintenance and operation commands

Description
This section contains information related to system operation and maintenance. In
particular, commands used to create new file systems, newfs, mkfs, and verify
the integrity of the file systems, fsck, icheck, dcheck, and ncheck are
described here. The f ormat(8v) reference page should be consulted when
formatting disk packs. The crash(8v) reference page should be consulted in
understanding how to interpret system crash dumps.

Maintenance 8-1

VAX

27S0d(S)
Name
2780d - IBM 2780/3780 RJE emulator daemon

Syntax
2780d

Description
The 2780 d program is automatically executed when a user runs 2780 e(1) or
3780e(1). Once executed, 278 Od transmits all data files (df*) in
/usr/spool/rje which have associated control files (cf*), to the IBM system.
Two users executing 27 80 e (3780 e) do not cause two daemons to run. The first
one executed handles both transmissions. The daemon's process id is kept in
/usr/spool/rje/. rjed. Each time a daemon is run, it checks that file to see if
an active daemon is running and exits if one is.
If 2780 d has trouble accessing files, the system manager should clean up the spool
directory by removing any outdated received files or data files which are not
associated with a corresponding control file.

The 2780 d program may become disabled when: the call to the IBM system has
not been completed within one minute, the IBM system has gone down during
transmission, the /usr/spool/rje/. rjed file can not be read, there has been an
error in sending or receiving.
To re-establish the connection to the IBM system, restart the daemon by typing
2780 d on the command line.
When files are received from the IBM system, 278 Od places them in
/usr / spool/rje/rjetemp. out. If your usemame is included in the job
control statements, the files are given a file name with the form rjedaytime and placed
in your home directory. If your usemame is not included, the files are placed in
/ u s r / s p 0 01 / r j e. If there are problems in renaming the files, they remain in
/usr/spool/rje/rjetemp.out.
To obtain a log of all files that are sent and received the system manager must create
/usr/spool/rje/acctlog. Once created, this file will list each file
transmission. Errors are logged in /usr/adm/rjelog.

Files
/usr/spool/rje/.rjed
Process identification file
/usr/spool/rje/cf*
Control file
/usr/spool/rje/df*
Data file
/usr/spool/rje/zf*
Renamed control file
/usr/spool/rje/acctlog
File transmission log file
8-2 Maintenance

2780d (8)
/usr / adm/ rjelogError log file

See Also
Guide to System Exercisers

8-52 Maintenance

comsat(8c)
Name
comsat - biff server

Syntax
letc/comsat

Description
The comsat command is the server process which receives reports of incoming mail
and notifies users if they have requested this service. The comsat command is
invoked by inetd(8c), when it detects an incoming packet on the datagram port
associated with the' 'biff" service specification. For further information, see
services(5). The comsat command reads the packet, which is a one line
message of the form:

user@mailbox-offset
If the user specified is logged in to the system and the associated terminal has the
owner execute bit turned on (by a "biff y"), the offset is used as a seek offset into
the appropriate mailbox file and the first 7 lines or 560 characters of the message are
printed on the user's terminal. Lines which appear to be part of the message header
other than the From:, To:, Date:, or Subject: lines are not included in the displayed
message.

Restrictions
The message header filtering is prone to error.

Files
/etc/utmp

Information about who is logged on to which terminals

See Also
Guide to System Exercisers

Maintenance 8-81

dump(8)
Name
dump - create file system dump

Syntax
fete/dump [ key [ argument... ] filesystem ]

Description
The dump command copies all files changed after a certain date from a specified
filesystem to a file, a pipe, magnetic tapes, or disks. The key specifies the date and
other options to be used by dump.
Dumping a filesystem requires operator attention. An operator must intervene when
the end of a tape or disk is reached, when the end of the dump occurs, or when an
unrecoverable input disk read error occurs (if more than 32 read errors occur). In
addition to alerting all operators in the operator group, dump interacts with the
operator at the control terminal when dump can no longer proceed, or if something is
grossly wrong. All questions that dump poses must be answered by typing yes or
no.
Because a full dump uses considerable system time, dump checkpoints itself at the
start of each tape or disk volume. If writing that volume fails, dump asks the
operator to restart from the checkpoint after the present tape or disk has been
replaced.
The dump utility reports to the operator periodically, giving usually low estimates of
the number of blocks to write, the number of tapes or disks the dump will take, the
time to completion, and the time until the tape or disk must be changed. The output
is verbose, so that others know that the terminal controlling dump is busy, and will
be for some time.
This utility supports EOT handling which allows the use of multiple media. The
utility prompts for the next volume when it encounters the end of the current volume.

Options
With the dump command, you specify a string of one or more of the options
described below. If no options are specified, the key 9u is assumed.

0-9

Specifies the dump level. All files that were modified since the last date
stored in the file / etc/ dumpdates for the same filesystem at lesser
levels will be dumped. If no date is determined by the level, the beginning
of time is assumed. Thus, the level 0 causes the entire filesystem to be
dumped.

Indicates that the next argument is a number that specifies the size, in
1024-byte blocks, of a storage medium, such as a diskette or removable
disk cartridge. See the first example.

Indicates that the density of the tape, expressed in bits per inch, is taken
from the next argument. This density is used in calculating the amount of
tape used per reel. The default density is 1600 bpi.

Places the dump on the file or device specified by the next argument. If the
name of the file is -, dump writes to standard output. The default dump

8-82 Maintenance

dump(8)
device is / dev / rmt Oh.
n

Notifies, by means similar to a wall(1) command, all users in the group
operator when dump needs operator attention.

Prints output file size in bytes, or number of volumes for devices. See the
third example.

Indicates that the next argument specifies the size of the dump tape, in feet.
When the specified size is reached, dump waits for the reel to be changed.
The default tape size is 2300 feet.

Writes the date of the beginning of the dump on the file
/ etc/ dumpdates if the dump completes successfully. This file records
a separate date for each filesystem and each dump level. The format of
/ etc/ dumpdates consists of one free format record per line: filesystem
name, increment level and ctime(3) format dump date. The superuser can
carefully edit / etc/ dumpdates to change any of the fields.

Tells the operator which file systems need to be dumped. This information
is taken from the files /etc/dumpdates and /etc/fstab. The W
option causes dump to print out, for each file system in
/ etc/ dumpdates, the most recent dump date and level, and highlights
those file systems that should be dumped. If the W option is used, all other
options are ignored, and dump exits immediately.

Lists only those filesystems that need to be dumped.

Examples
This example dumps the file system / dev / r a 0 a to RX50 diskettes. The B option is
needed when running restore(8) to read this dump.
dump 9Bf 400 /dev/rra2a /dev/raOa

This example dumps the file system / u s r / use r s to a 6250 bpi tape on a TU78 tape
drive:
dump Oundf 6250 Idev/rmt?h lusr/users

This example reports number of bytes to be output for a level 0 dump of the root file
system. Please note: the file test is not created.
dump OSf test I

Restrictions
The dump programs returns a 1 on successful completion.
Sizes are based on 1600-bpi blocked tape.
Anything fewer than 32 read errors on the filesystem are ignored.
Each reel requires a new process, so parent processes for reels already written remain
until the entire tape is written.

Maintenance 8-83

dump(8)
Files
/ dev / r rp 1 9

Default filesystem to dump from

/ dev / rmt 0 h

Default tape unit to dump to

/etc/dumpdates
Dump date record
/etc/fstab

Dump table: file systems and frequency

/ etc/ group

Operator group definition

/ dev / tty

Required for user interface

See Also
dump(5), fstab(5), opser(8), restore(8), rrestore(8c)
Guide to Backup and Restore

8-84 Maintenance

dumpfs(8)
Name
dumpfs - dump file system information

Syntax
dumpfs fi/esys Idevice

Description
The dumpfs command prints out the super block and cylinder group information for
the file system or special device specified. The listing is very long and detailed.
This command is useful mostly for finding out certain file system information such as
the file system block size and minimum free space percentage.

See Also
disktab(5), fs(5), fsck(8), newfs(8), tunefs(8)

Maintenance 8-85

edauth (8)
Name
edauth - edit user auth entry

Syntax
edauth username

Description
The edauth command is an authorization editor. edauth creates a temporary file
with an ASCII representation of the current aut h database entry for the user
specified by username and then invokes an editor on the file. You can then modify
the user's auth fields. Upon leaving the editor, edauth reads the temporary file
and modifies the binary database to reflect the changes made. If there are errors in
the temporary file e da u t h will allow the user to resume editing the file to fix them.
The editor invoked is ed(l), unless the environment variable EDITOR specifies
otherwise. Here is an example of the temporary file produced by edauth:
uid = 268
password = MXP3BnKLEWW960BEJc9DbHb6
passlifemin = 1 hour
passlifemax = 60 days
passmod = 12/20/89 - 10:24:38
authmask = login,change-password,enter-password
fail_count = 0
audit id = 268
audit_control ~ or
audit_syscalls = creat,unlink
audit_tevents = login:0:1

Each field of the aut h entry is represented as a keyword followed by an equals sign.
The value part of the field may be an integer, a string, a time specification, a date, or
a comma-separated list of value keywords. The effect of the field is described in
auth(5).
The uid, fail_count, and audi t_id fields expect integer values.
The password field is a string containing the encrypted password. One way of
disabling an account is to set this to a non-empty string less than 24 characters in
length such as 'nologin'.
The passlifemin and passlifemax fields specify the password expiration
information. They may contain an integer specifying seconds, or a combination of
scaled values. The units recognized for scaling are seconds, minutes, hours,
and days. Only the first letter of the unit need be supplied. A passlifemax of
one day, one hour and five minutes could be specified as any of:
passlifemax =
passlifemax =
passlifemax =
passlifemax =

1 day 1 hour 5 minutes
25 h 5 m
90300 seconds
90300

in addition to other combinations.
The passmod field is a date. It is specified in the same format as the default output
of the ULTRIX date ( 1) command. The time portion is optional and defaults to
the beginning of the day.

8-86 Maintenance

edauth(8)
The authmask, audit syscalls, and audit tevents fields expect a
comma-separated list of vaiue tokens. For authmask this is zero or more of
login, change_password, and enteryassword. For the audit
information this corresponds to the name of the audit event. See the
audi tmask (8) manpage for more information on audit events.
The audit control field may be one of or, and, or off. See the
audcntl (2) manpage for more information on the affect of these values.

Restrictions
Only the superuser can edit auth entries.
Changing the auth entry will not affect the uid and audit information of exisiting
login sessions.
If the uid field of the entry is changed the mapping to the / etc/passwd file will be
affected. Changes to the passwd file will probably be necessary.

Diagnostics
Various messages about incorrect input. All are self-explanatory.

Files
/etc/auth. [dir,pag]
Contains all authorization information
/etc/passwd Maps usernames to UIDs

See Also
audcntl(2), auth(5), auditmask(8), getauth(8), vipw(8)
Security Guide for Administrators

Maintenance 8-87

edquota(8)
Name
edquota - edit user quotas

Syntax
edquota [ -p proto-user] users ...

Description
The edquota command is a quota editor. You can specify one or more users on
the command line. For each user, edquota creates a temporary file with an ASCII
representation of the current disc quotas for that user and an editor is then invoked on
the file. You can then modify the quotas, add new quotas, and so forth. Upon
leaving the editor, edquota reads the temporary file and modifies the binary quota
files to reflect the changes made.
If a valid quota file does not exist, edquota refers the user to quotacheck(8).
The user then has the option of either supplying the command quotacheck -f
filesystem, which automatically creates the necessary quota file, or creating an
empty quota file by various other means.
If the -p option is specified, edquota duplicates for each user, the quotas of the
prototypical user specified. This is the usual mechanism used to initialize quotas for
groups of users.

The editor invoked is vi(1), unless the environment variable EDITOR specifies
otherwise.
Only the superuser can edit quotas.

Diagnostics
Various messages about inaccessible files. Other messages are self-explanatory.

Files
quotas

At the root of each file system with quotas

/ etc/ fstab

Used to find file system names and locations

See Also
quota(l), quota(2), quotacheck(8), quotaon(8), repquota(8)
"Disk Quotas in a UNIX Environment," ULTRIX Supplementary Documents,
Volume 3: System Manager

8-88 Maintenance

elcsd (8)
Name
elcsd - error logging daemon

Syntax
letc/elcsd

Description
The elcsd daemon logs hardware and system-related error packets~ or error
messages, from the kernel errorlog buffer to the file syserr. <hostname>.
Hostname is the name of the local system. At system startup, the e 1 c s d daemon is
invoked out of / etc/ rc. For further information, see rc(8). The elcsd daemon
must be running whenever the system is in multiuser mode.
To invoke the elcsd daemon in single-user mode, use the command with the-s
option.
You can configure the elcsd daemon to log error packets locally from a remote
ULTRIX host, or you can configure the e 1 c s d daemon to send local error packets to
a remote system. Do this by changing the entries in the e 1 c s d . con f file. For
further information, see elcsd. conf(5). The daemon uses an internet datagram
socket to log information across systems.
Parameters related to e 1 c s d, such as the desired location for the errorlog file or
files, as well as instructions for local and remote logging, can be specified in the error
logging configuration file, / etc/ elcsd. conf. For further information, see
elcsd. conf(5).

Restrictions
You must have superuser privileges to invoke the elcsd daemon.

Diagnostics
The e 1 c s d daemon logs general status and error messages to
/usr / adrn/ elcsdlog in multiuser mode. This file is purged whenever the
daemon is restarted.

Files
/etc/elcsd.conf
/usr/adrn/elcsdlog

See Also
elcsd.conf(5), eli(8), rc(8), uerf(8)
Guide to the Error Logger

Maintenance 8-89

eli (8)
Name
eli - error log initialization program

Syntax
eli [ options ]

Description
The eli command initializes error logging. This command enables error logging of
hardware and system-related error packets, or error messages, from the kernel
errorlog buffer. It can also disable error logging, reconfigure error logging
parameters, or initialize the kernel errorlog buffer.

Options
-d

Disables error logging.

-,-e

Enables error logging in multiuser mode.

-f

Forces the subsequent option; the system will not prompt. This is the only
eli command option you can use with another option.

-h

Prints information about the eli command.

-i

Initializes the kernel errorlog buffer. The previous contents of the errorlog
buffer are lost.

-I

Logs a one-line status message to the kernel errorlog buffer.

-0

Only supported for local error logging. Disables logging error packets to
disk by the e 1 c s d daemon. High priority error messages continue to be
printed at the console. Note that error log packets can be viewed by using
the uerf-n option in real time, but are never written to the disk. For this
reason, this option is rarely used.

-q

Suppresses the periodic display, on the console, of the missed error
message that results from a" full kernel errorlog buffer.

-r

Reconfigures error logging. Use this option after changing the
/ etc/ elcsd. conf file.

-s

Enables error logging in single-user mode.

-w

Enables the missed error message to appear on the console every 15
minutes. This option is the opposite of -q.

Examples
This example logs a one-line message into the errorlog file.
eli -f -1 "This is a test message"

This example logs a message, up to and including the first new line, from the file
myfile:
eli -f -1 < myfile > /dev/null

8-90 Maintenance

eli (8)
Restrictions
You must have superuser privileges to use the eli command.
Only the -f option can be used with other eli options. You must use the syntax
shown in the examples above.

Diagnostics
eli: Request Aborted.
The requested action was aborted. The reason for the aborted command is included
with the output.

Files
/etc/elcsd.conf

See Also
elcsd.conf(5), elcsd(8), uerf(8)
Guide to the Error Logger

Mai ntenance 8-91

ext_srvtab (Skrb)
Name
ext_srvtab - extracts service key files.

Syntax
/var/dss/kerberos/bio/ext_srvtab [-n] [hostname ... ]

Arguments
hostname Name of the service key file.

Description
The ext _ s rvt ab utility extracts service key files from the Kerberos database.
When you invoke the utility, it prompts you to enter the master key string for the
database. If the -n option is specified, ext srvtab fetches the master key from
the master key cache file, produced by kstash (8krb) .
For each hostname that you specify in the command line, ext srvtab creates the
service key file hostname-new-srvtab. This file contains all the entries in the Kerberos
database with an instance field that matches hostname. In addition, the hostnamenew-srvtab file contains all the keys registered for Kerberos services that can run on
the host specified in hostname.

Options
-0

Causes ext_srvtab utility to fetch the master key from the master key
cache file.

Files
/var/dss/kerberos/dbase/principal.pag
/var/dss/kerberos/dbase/principal.dir
/var/dss/kerberos/dbase/principal.ok

See Also
kdb_edit(8krb), kdb_init(8krb), kdb_util(8krb), kdb_destroy(8krb)

8-92 Maintenance

fingerd (Bc)
Name
fingerd - remote user information server

Syntax
/etc/fingerd

Description
The f ingerd program is a protocol that provides an interface to the Name and
Finger programs at several network sites. The program returns an in depth status
report on either the system currently in use or on a particular user whether that user is
logged in or not.
The fingerd program is never executed directly by a user, but rather by the
ULTRIX Internet listener daemon / etc/ inetd. When / etc/ inetd is listening
on port 79, it passes any requests that it receives to /etc/fingerd. Once
fingerd receives control, it reads a single command line terminated by a
<CTRL/F> which is passed to finger(1). The fingerd command closes its
connections as soon as the output is finished.
If the line is null, for example a <CTRL/F> is sent, then finger returns a default
report that lists all the users logged into the system at that moment.
If a user name is specified as e ric A F, for example, more extended information is
listed for only that particular user, whether he is logged in or not. Allowable names
in the command line include both login names and user names. If a name is
ambiguous, however, all possible derivations are returned.

See Also
finger(l), services(5), inetd(8c)

Maintenance 8-93

fitset(8)
Name
fitset - determine if subset fits on a system

Syntax
fitset [ -d ] [ root-path]

Description
The fi tset utility is used to determine if the files in a software subset will fit on a
system.
Subset inventory records are read from the standard input. For each record, the space
required to install the file described in that record is deducted from the available free
space on the file system to which it would be installed. Only currently mounted UFS
file systems are used in the computations. Space requirements for files already on the
disk will be modified to account for the size of the resident copy. After all records
have been read, the free space computed for all file systems is checked. If the space
required to install the files would cause any file system to be more than 90% full,
fitset returns an exit status of-1.
The setld utility uses fi tset to size all subsets before attempting to install them.
The root-path argument is the pathname of the top directory for the hierarchy into
which the files are going to be installed. If no root-path is specified, the directory '/'
is assumed.

Options
-d

Enable debugging. This will make fit set print voluminous status
information on standard output. This information is the initial file system
statistics, the file system location of the file from each input record and the
statistics for the file system after the space required to install the file has
been deducted.

Restrictions
NFS mounts are ignored. If software would be installed to an NFS mounted
directory, it is sized against the file system containing the NFS mount point.
The program does not detect the use of symbolic links in paths to root-path or any of
the mount points. This can cause fit set to size a subset incorrectly if root-path is
a symbolic link or a symbolic link exists in the path of any of the pathnames used
with the mount command to mount local file systems.

Examples
To determine if a particular subset will fit on the system, redirect the contents of the
subset inventory file into fitset. For example:
fitset < /usr/etc/subsets/ULTUUCP400.inv

To determine if the same subset will fit in a hierarchy rooted at /var /tmp/ root,
the command would be:
fitset /var/tmp/root < /usr/etc/subsets/ULTUUCP400.inv

8-94 Maintenance

fitset (8)
Diagnostics
fitset: root path must be absolute
A relative pathname was specified for root-path. This path must be absolute.

fitset: cannot stat root-path (error message)
The root-path cannot be accessed. The error message provides more information.
fitset: root-path is not a directory.
Either root-path is not a directory or it is a symbolic link to something which is not a
directory.

Files
/usr/etc/subsets/*.inv
Subset inventory files

See Also
stl_inv(5}, setld(8}
Guide to Preparing Software for Distribution on ULTRIX Systems

Maintenance 8-95

VAX

flcopy(8v)
Name
flcopy - copier for floppy

Syntax
/etc/f1copy [ -h ] [ -tn ]

Description
The flcopy command copies the console floppy disk (opened as /dev/floppy)
to a file created in the current directory, named floppy, then prints the message
Change Floppy, hit return when done. Then flcopy copies the local
file back out to the floppy disk.
The -h option to flcopy causes it to open a file named floppy in the current
directory and copy it to / dev / floppy. The -t option causes only the first n
tracks to participate in a copy.

Files
/dev/floppy
/dev/rrx??
floppy

In the current directory

See Also
Guide to System Exercisers

Maintenance 8-105

ftpd(8c)
Name
ftpd - DARPA Internet File Transfer Protocol server

Syntax
/usr/etc/ftpd [ -d ] [ -I ] [ -ttimeout ]

Description
The ftpd server is the DARPA Internet File Transfer Protocol server process. The
server uses the TCP protocol and is invoked by inetd (Bc) when it receives a
connection on the port specified in the ftp service specification. For further
information, see services (5) .
The ftp server currently supports the following ftp requests. Case is not
distinguished.

Request

Description

ABOR
ACCT
ALLO
APPE
CDUP
CWD
DELE
HELP
LIST
MKD
MODE
NLST
NOOP
PASS
PASV
PORT
PWD
QUIT
RETR
RMD
RNFR
RNTO
STOR

Abort previous command

8-106 Maintenance

Specify account
Allocate storage
Append to a file
Change to parent of current working directory
Change working directory
Delete a file
Give help information
Gi ve list of files in a directory (1 s -lg)
Make a directory
Specify data transfer mode
Give name list of files in directory (Is)
Do nothing
Specify passworq
Prepare for server-to-server transfer
Specify data connection port
Print the current working directory
Terminate session
Retrieve a file
Remove a directory
Specify rename-from file name
Specify rename-to file name
Store a file

ftpd(8c)
STOU

Store a file with a unique name

STRU

Specify data transfer struCture

TYPE

Specify data transfer type

USER

Specify user name

XCUP

Change to parent of current working directory

XCWD

Change working directory

XMKD

Make a directory

XPWD

Print the current working directory

XRMD

Remove a directory

The remaining ftp requests specified in Internet RFC 959 are recognized, but not
implemented.
The ft pd server interprets file names according to the globbing conventions used by
csh (1). This allows users to utilize the metacharacters *?[] {} .....
The ftpd server authenticates users according to three rules:
1.

The user name must be in the password database, / etc/passwd, and not
have a null password. In this case a password must be provided by the client
before any file operations may be performed.

The user name must not appear in the file /etc/ftpusers.

If the user name is anonymous or ft p, an anonymous ft p account must be
present in the password file (user ftp). In this case the user is allowed to log
in by specifying any password (by convention this is given as the client host's
name),

In the last case, ftpd takes special measures to restrict the client's access privileges.
The server performs a chroot (2) command to the home directory of the ftp user.
To preverit system security from being breached, it is recommended that the ftp
subtree be constructed with care. Thus the following rules are recommended:
""rtp)

Make the home directory owned by ftp and unwritable by anyone.

~ftp/bin)

Make this directory owned by the superuser and unwritable by anyone.
The program 1 s (1) must be present to support the list commands. This
program should have mode 111.

""ftp/etc)

Make this directory owned by the superuser and unwritable by anyone.
The files passwd(5) and group(5) must be present for the Is
command to work properly. These files should be mode 444.

""ftp/pub)

Make this directory mode 777 and owned by ftp. Place the files,
which are to be accessible by the anonymous account, in this directory.

Options
-d

Enables certain debugging messages that are printed by ftpd.

-I

Logs each ftp session to the s y slog.

-t

Sends the inactivity timeout period to timeout; otherwise, the ftp server will
timeout an inactive session after 15 minutes.
Maintenance 8-107

ftpd (Bc)

Restrictions
Support does not exist for aborting commands.
The use of an anonymous account is inherently dangerous and should be avoided
when possible.
The server must run as the superuser to create sockets with privileged port numbers.
The server maintains an effective user id of the logged in user, reverting to the
superuser only when binding addresses to sockets. The possible security holes have
been extensively scrutinized, but are possibly incomplete.

Files
/etc/ftpusers
Contains the list of unauthorized users

See Also
ftp(1c), services(5), inetd(8c), syslog(8)

8-108 Maintenance

fverify (8)
Narne
fverify - verify software subsets

Syntax
fverify [-yo]

Description
The fverify command reads subset inventory records from standard input and
verifies that the attributes for the files on the system match the attributes listed in the
corresponding records. Invoking f ve r if y with no arguments causes it to report
errors to the standard output and prompt the user for confinnation before making
corrections.
Missing files and inconsistencies in file size, checksum, user id, group id, mode, and
file type are reported. If inconsistencies in user id, group id or mode are detected the
user is asked if they are to be corrected. If a missing file is a directory, it will be
created. All errors and infonnational messages are logged to the file
/var/adm/fverifylog.
The setld utility uses fverify when installing subsets to guarantee correct
installation.

Options
Specifying both options will cause the -y option to be ignored.
-0

Error reporting only. When specified, fverify reports errors and does
not perfonn any fixes.

-y

Fixes only. When specified, fverify fixes modes, DIDs and GIDs on
any files which have incorrect values for these attributes. No user input is
required.

Restrictions
Problems with file size, file type, and file checksum cannot be corrected.
Because the subset inventories give file names as relative paths, run fver i fy in the
root directory to which the software is installed.
Many of the files on the system cannot be read or modified without appropriate
privilege. Attempting to run fverify without appropriate privilege may result in an
excess of access errors being reported.

Examples
To use fverify to report verification problems with the BASE subset on DLTRIX,
Version 4.0 (VAX), use the following command:
% cd I
% letc/stl/fverify -0 </usr/etclsubsetsIULTBASE400.iov

Maintenance 8-109

fverify (8)
To correct all problems with the same subset without being presented any prompts or
diagnostic output, type:
% cd I
% letc/stl/fverify -y < lusr/etc/subsets/ULTBASE400.inv

To interactively repair verification problems with the COMM subset on ULTRIX,
Version 4.0 (RISC), use the fverify command with no arguments:
% cd I
% letc/stllfverify < lusr/etclsubsetslUDTCOMM400.inv

Diagnostics
fverify: out of memory
Not enough memory is available from the system.
pathname: cannot stat (error-message)
The file pathname is listed in the inventory but is not on the system. The errormessage explains why.
pathname: file type 'x' should be 'y'
The file is listed in the inventory as being of type 'y', but the copy of the file on the
disk is type 'x'. The file type codes are:

b block device files
c character device files
d directory files
f regular files
I hard links
p fifos (named pipes)
s symbolic links
pathname: checksum n should be m
There is a checksum error on pathname. The values nand m are the actual and
expected values.
pathname,' size n should be m
The size of file pathname is not as expected. The numbers nand m are the actual and
expected values.
pathname,' gid n should be m
The group id for pathname is not as expected. The group ids nand m are the actual
and expected values.
pathname,' uid n should be m
The user id for pathname is not as expected. The user ids n anq m are the actual and
expected values.
pathname,' permissions string1 should be string2
The permissions for pathname are not as expected. String 1 and string2 are the actual
and expected permissions. The format of string1 and string2 is the same as that used
by the 1 s command.

8-110 Maintenance

fverify (8)
Creating directory path name
The directory pathname is listed in the input inventory but does not exist on the
system. The fverify command will attempt to create the directory.
cannot correct pathname (error-message)
A problem reported with pathname could not be corrected. The error-message
explains why.
pathname corrected
A problem with pathname was successfully corrected.
Cannot create dir pathname (error-message)
The fverify program could not create a directory it was attempting to create. The
accompanying error-message explains why.
n verification errors encountered.

m corrections performed.
Before exiting, the fverify program prints these statistics describing what had
been done.
The exit status from fverify is the total number of errors detected minus the total
number of successful fixes.

Files
/usr/etc/subsets/*.inv
Subset inventory files
/var/acim/fverifylog

Log File

See Also
Is(1), stl_inv(5), setld(8)
Guide to Preparing Software for Distribution on ULTRIX Systems

Maintenance 8-111

genra(8)
Name
genra - produce distribution RA60 media

Syntax
/etc/genra I -{wv}] [hostname:]product_code special

Description
The genra utility is used to produce RA60 distribution media. The utility searches
/etc/kitcap for the product_code (as provided by the user on the command line),
creates a new file system on the partition defined in the kit description, mounts the
disk, .transfers files and subsets as described in the kitcap kit description, uses s um( 1)
to verify files for accuracy in transfer, and then unmounts the file system. The
genra utility can create multiple· directories on the RA60 for separate products, and
can also combine multiple products into one product, all controlled by
/ etc/ki tcap.
The optional hostname: argument is the name of a remote TCP/IP network machine
that contains the kitcap file. The utility will search /etc/kitcap on the remote
machine for the product_code and use it for creating the media. The colon (:) is a
required delimiter for TCP/IP networks, and there is no space between the colon and
the product_code. For example, if the product code was ULT-2.2-V-BW, and the
kitcap file to be used was on node "mynode", the proper syntax for this option would
be
mynode:ULT-2.2-V-BW

The product code is a user defined code that is located in / etc/ki tcap, and
describes the partition, directories, and subsets, that make up a kitcap description. It
can be any set of numbers or letters, and is usually formed in a way to have some
meaning as to the product that it describes. For example, if the product name is
MYPRODUCT and it is version 1, a proper product_code for that product might be
MYPOIO.
The special argument indicates that the device is a special device such as
/ dev / r r a 1 a. The actual partition that the utility will write the files on is defined
in the kitcap description for the product_code that is being used. Therefore, it makes
no difference what partition is appended to the special device file used on the
command line.

Options
-w

Write only.

-v

Verify only.
If neither option appears on the command line, then by default, the utility
will write, then verify, the files in the kit descriptor.

8-112 Maintenance

genra(8)
Restrictions
You must be a superuser to run this program. If the optional hostname is used, you
must be able to access files on the remote host as superuser.

Files
/dev/ra?a

ra device special files

/ etc/ki tcap Kit descriptor database

See Also
Guide to System Exercisers

Maintenance 8-201

makedbm (8yp)
Name
makedbm - make a yellow pages dbm file

Syntax
makedbm [ -i yp _inputJtle ] [ -0 yp_output_name ] [ -d yp_domain_name ] [ -m
yp master name ] infile outfile
makedbm-[ -u dbmfilename ]

Description
The makedbm command takes the file specified by the argument infile and converts
it to a pair of files in dbm(3x) format, namely outfile .pag and outfile. dir.
Each line of the input file is converted to a single dbm record. All characters up to
the first tab or space form the key, and the rest of the line is defined as the key's
associated data. If a line ends with a backslash (\), the data for that record is
continued onto the next line. It is left for the clients of the yellow pages to interpret
the number sign (#); makedbm does not treat it as a comment character. The infile
parameter can be a hyphen (-), in which case makedbm reads the standard input.
The makedbm command is meant to be used in generating dbm files for the yellow
pages service. The makedbm command generates a special entry with the key
yp_Iast_modified, which is the date of infile.

Options
-i

Create a special entry with the key yp_input_file.

-0

Create a special entry with the key yp_output_name.

-d

Create a special entry with the key yp_domain_name.

-m

Create a special entry with the key yp_master_name. If no master host
name is specified, yp_master_name will be set to the local host name.

-u

Undo a dbm file. That is, print out a dbm file one entry per line, with a
single space separating keys from values.

Examples
The following example shows how a combination of commands can be used to make
the yellow pages dbm files passwd. byname. pag and passwd. byname. dir
from the / etc/passwd file. The percent sign (%) signifies the system prompt.
% awk 'BEGIN { FS =

":"; OFS = ""; }
{ print $1, $0 }' /etc/passwd > ptmp
% makedbm ptmp passwd.byname
% rm ptmp

The awk command creates the file ptmp which is in a form usable by makedbm.
The makedbm command uses the ptmp file to create the yellow pages dbm files
passwd.byname. dir and passwd.byname .pag. The rm command removes
the ptmp file.

8-202 Maintenance

makedbm (8yp )
See Also
yppasswd(1yp), dbm(3x), ypmake(8yp)

Maintenance 8-203

Rise

MAKEDEV(8)
Name
MAKEDEV - makes system special files

Syntax
/dev/MAKEDEV device-name? ..

Description
TheMAKEDEV shell script is normally used to install special files. It resides in the
/dev directory, the normal location of special files. Arguments to MAKEDEV are
usually of the form device-name? where device-name is one of the supported devices
listed in Section 4 of the ULTRIX Reference Pages and? is a logical unit number. A
few special arguments create assorted collections of devices and are listed below.

DECstation* Creates a DEC station specific setup.
std

Creates all standard devices for all systems.

local

Creates those devices specific to the local site. This request causes
the shell file / dev /MAKEDEV . local to be executed. Site specific
commands, such as those used to setup dialup lines as ttyd?, should
be included in this file.

Because all devices are created using rnknod(8), this shell script is useful only to the
superuser.

Diagnostics
Either self-explanatory, or generated by one of the programs called from the script.
Use sh -x MAKEDEV in case of trouble.

See Also
Guide to System Exercisers

Maintenance 8-209

miscd(8c)
Name
miscd - miscellaneous services daemon

Syntax
letc/miscd

Description
Themiscd daemon is the service daemon for some utility internet services. It is
started by inetd(8c) when it receives a packet or a connection on an appropriate
socket. Once invoked, rtli s cd detennines the type of request and the protocol and
then attempts to service it.
When performing services for datagram functions, the daemon will stay active until
approximately two minutes pass without a service request. It will then terminate
until reinvoked by inetd(8c.)
Invocations of mi s cd serving for connection-based sockets will terminate when the
connection is broken.

Restrictions
The mi s cd daemon supports only a subset of the internet utility functions. The
supported services are echo, discard, systat, daytime, quote, chargen, and time.

See Also
Guide to System Exercisers

Maintenance 8-343

showmount (8n1s)
Name
showmount - show remotely-mounted file systems

Syntax
lusr/etc/showmount [ -a ] [ -d ] [ --e ] [ host]

Description
The showmount command lists all of the NFS client machines that have remotely
mounted a filesystem from the NFS server host . This information is maintained by
the mountd(8nfs) daemon on host, and is saved across crashes in the / etc/ rmtab
file. The default value for host is the value returned by hostname(1) or
gethostname(2).

Options
-d

List directories that have been remotely mounted by clients.

-a

Print all remote mounts in the format:
hostname:directory

In this format, hostname is the name of the client, and directory is the
directory that has been mounted.

-e

Print the list of file systems exported by the NFS server, host.

Restrictions
If an NFS client crashes, showmount will incorrectly report that the client still has a
file system mounted, until the crashed client reboots.

See Also
hostname(1), exports (5nfs), rmtab( 5nfs), mountd(8nfs)

8-344 Maintenance

shutdown (8)
Name
shutdown - close down the system at a given time

Syntax
letelshutdown [ -k ] [ -r ] [ -h ] [ -0 ] time (warning-message ... ]

Description
The shutdown command provides an automated shutdown procedure that a
superuser can use to notify users when the system is shutting down.
The time is the time at which shutdown will bring the system down. It may be the
word 'now', indicating an immediate shutdown, or specify a future time in one of
two formats: + number or hour: min. The first form brings the system down in
number minutes. The second brings the system down at the time of day indicated,
using a 24-hour clock format.
At intervals which get shorter as shutdown nears, warning messages are displayed at
the terminals of all users on the system. Warning messages are also sent to users who
are logged in to a remote system that has mounted a file system or directory from the
local system using NFS. Five minutes before shutdown, or immediately if shutdown
is timed for less than five minutes, logins are disabled by creating /etc/nologin
and writing a message there. If this file exists when a user attempts to log in,
login(l) prints its contents and exits. The file is removed just before shutdown
exits.
At shutdown time, a message is written in the file /usr / adm/ shutdownlog.
This message contains the time of shutdown, who ran shutdown, and the reason.
Then, a terminate signal is sent at ini t to bring the system to single-user state.
If the - r, - h, or - k options are used, then shutdown executes reboot(8),
hal t(8), or avoids shutting the system down (respectively). The -0 option is for
use by opser only. It indicates to shutdown that it is being called by opser and
not to return to the user.

You should place the time of the shutdown and the warning message in
/ etc/nologin. Use the message to inform the users about when the system will
be back up and why it is going down.

Restrictions
You can kill the system only between now and 23:59, if you use the absolute time
for shutdown.

Files
/ etc/nologin Tells login not to let anyone log in
/usr/adm/shutdownlog
Log file for successful shutdowns

Maintenance 8-345

shutdown (8)
See Also
login( 1), wall( 1), halt(8), opser(8), reboot(8), rwalld(8c)

8-346 Maintenance

sizer(8)

Name
sizer - a program that sizes system hardware

Syntax
letelsizer [ options ]

Description
The s i z e r program reports various items of information about a kernel file and the
running system. It is also used to create a system configuration file in the doconfig
command.

Options
-b

Create a boot command file for the current system disk. The sizer program
creates the appropriate defboo and askboo command files in
/usr / sys / * . cons. These files can then be used to update the systems
console media. This only applies for VAX 8600, VAX 780, VAX 6200, and
VAX 3400 cpus. This also applies for the VAX 750 and VAX 8200 only if the
system disk is connected to an HSC.

-e

Return the CPU type of the running cpu.

-t timezone
Use timezone in the config file.
-wt

Returns the workstation display type.

-wu Returns the workstation display units.
-k image
Use the indicated kernel file instead of /vrnunix. This is useful when the
running system is not /vrnunix or you wish information about a system that
is not running.

-nfilename
Creates a configuration file. The system must be running the /vrnunix kernel
unless the -k option is used. The -n option creates a configuration file in
/trnp/filenarne and also a file required for MAKEDEV called
/trnp/filenarne.devs.

NOTE: The user should run doconfig to build a new kernel.
-r

Return only the name of the root device and partition.

-s

Return only the CPU sUbtype of the running cpu. This only applies to VAX
cpus. It returns following values for the following cpus:

Returned
Value
CPU
11
10
9

8
6

VAX8800
VAX8700
VAX8550
VAX8500
VAX8300
Maintenance 8-347

sizer(8)
5
4

3
1

VAX8200
VAXstation/MVAX 2000/3100
VAX60
MVAXIl, VAX3xOO, VAX62xx

The value 0 is returned for all other cpu types.

Files
/usr/sys/780cons/defboo.cmd
/usr/sys/8200cons/defboo.com
/usr/sys/8600cons/defboo.com
/usr/sys/780cons/askboo.cmd
/usr/sys/8200cons/askboo.com
/usr/sys/8600cons/askboo.com

See Also
config(8), doconfig(8)
ULTRIX Advanced Installation Guide

8-348 Maintenance

snapcopy(8)
Name
snapcopy - copy VAX 8600/8650 snapshot files

Syntax
/etc/snapcopy [-d] directory

Description
When the system crashes on a VAX 8600 or VAX 8650 system, the console
subsystem creates snapshot files containing binary information regarding the state of
the hardware at the time of the crash. Snapshot files are typically used by hardware
maintenance personnel to analyze and repair the cause of the system crash.
The snapcopy command copies any valid snapshot files (called snapl. dat and
snap2 . dat) from the console RL02 disk into the directory specified. The files are
renamed to show the time of the copy:

hr:min:sec-snapl. dat
hr:min:sec-snap2 . dat
You should delete the comment character for the snapcopy command in the
/etc/rc .local file only for VAX 8600 and VAX 8650 systems.
The directory is the name of the directory to which the snapshot files are to be
copied.

Options
-d

Invalidates the files on the console RL02 disk, meaning that they can be
rewritten by the console subsystem in the event of another system crash.

Diagnostics
snapcopy: not a directory
snapcopy: no write permission for directory
snapcopy: could not chdir to directory
The directory must exist and be writable.

Files
/dev/ttyc3

Special file by which snapcopy communicates with the system
console terminal

/dev/crl

Special file by which snapcopy communicates with the
console RL02 disk

/ etc/ rc .local File that contains site-specific commands to be executed when
the ULTRIX system is brought to multi-user mode

Maintenance 8-349

VAX

snmpd(8n)
Name
snmpd - Simple Network Management Protocol (SNMP) Agent for ULTRIX
gateways and hosts

Syntax
letc/snmpd [ -d debug/eve//ogfi/e ]

Description
The SNMP Agent, snmpd, performs SNMP operations on an ULTRIX gateway or
host. The daemon, which is started up by an entry in the / etc/ services file, sits
in the background and listens on SNMP port 161. When the snmpd daemon
receives an SNMP packet from a Network Management Station (NMS), the daemon
performs SNMP operations on the packet and returns a valid response to the NMS.
The s nmpd daemon extracts much of its information from kernel memory. Static
variables whose values are not available in the kernel take values from the SNMP
configuration file, / etc/ snmpd. conf.

SNMP Trap Support
The cold start and authentication failure trap types are supported by snmpd.
The cold start trap type is generated by snmpd when snmpd is restarted. The
authentication failure trap type is generated when an attempt at using a community
fails. The attempt fails when an unauthorized client tries to use s nmpd or the
community is used in a way that the community type does not allow.
The snmpd daemon sends traps to all communities specified in the configuration file
with a community type traps.
The default is for the snmpd daemon to generate authentication failure traps.
However, if the following clause is specified somewhere in the / etc/ snmpd. conf
file, authentication failure traps are not generated:
no_authen_traps

SNMP Sets
When the snmpd daemon receives a set-request packet, it processes the variables in
the packet and verifies that they are valid read-write variables. While performing this
verification, the s nmpd daemon constructs a linked list of the set requests. After it
has completed the verification, it performs the actual set operations on the variables,
as if they were being performed simultaneously. If any actual set operation fails, all
of the previous set variables from the set-request packet are restored to their old
values.

SNMP Supported Variables
For a complete listing of the SNMP Management Information Base (MIB) variables
that are supported, see the Guide to Networking.

8-350 Maintenance

snmpd{Sn)
Options
By default, the s nmpd daemon uses the s y slag command to record its error
messages. However, you can obtain certain debugging and trace information by
specifying the -d flag, the appropriate debug level, and a log file for the output on
the snmpd command line.

-d debuglevel logfile
Outputs debugging and trace information.
You can specify anyone of the following debug levels with the -d flag:
1

Print the version number, start time, and exit time of snmpd. Also print out
when an SNMP packet is received, the address of the sender, and the packet
size in bytes.

Print out what the snmpd daemon has read from the / etc/ snmp. canf file.

Dump the SNMP packet that the snmpd daemon has just received and is about
to process. Also print out the route and interface address that the s nmpd
daemon is currently looking up. This debug level also dumps the SNMP
packet that the server is sending back in response to a received SNMP
message.

Dump the snmp variable tree. Also print out the static bootstrap array of tree
information.

The output for a debug level includes the information for all levels including and
below the level that you specify. For example, if you specify a debug level of 3,
your output includes debug information for levels 3, 2, and 1.
If no debug levels are set, s nmpd detaches itself from the controlling terminal and
executes in the background.

Restrictions
Not all of the MIB variables are supported.
Only the_mgmt_mib_interfaces_if Table_if Entry_if AdminStat us
variable is settable.

Files
/etc/snmpd.canf

SNMP configuration file

See Also
snmpext(3n), snmpd.conf(5n), snmpsetup(8n)
RFC 1066-Management Information Base for Network Management of TCPIIPbased Internets
RFC 1067-A Simple Network Management Protocol
Guide to Networking

Maintenance 8-351

snmpsetup (8n)
Name
snmpsetup - set up the Simple Network Management Protocol (SNMP) Agent

Syntax
lusrletc/snmpsetup

Description
The snrnpsetup command is an interactive facility for configuring the ULTRIX
SNMP Agent. When your system is configured as an SNMP Agent certain network
parameters can be monitored or managed by a Network Management Station (NMS).
The snrnpsetup command also allows you to configure Extended ULTRIX SNMP
Agents. Extended Agents are user-defined daemons that allow you to specify private
network parameters to be monitored or managed by the NMS.
Before you invoke snrnpsetup, the network must be up and running, your host
must have a name, and you must know the Internet Protocol address of any NMSs
that you want to have monitor or manage your system.
You can run snrnpsetup while in multiuser mode, but you must be root or
superuser.
To invoke snrnpsetup, type:
# snmpsetup

Diagnostics
The following error messages are issued by snrnpset up:

snmpsetup must be run by root
snmpsetup: ne~d letc/rc
snmpsetup: need letc/snmpd.conf
snmpsetup: non-interactive mode not currently supported
snmpsetup: network must be setup before running snmpsetup
snmpsetup: terminated with no installations made

Files
/ etc/ snmpd. conf

SNMP configuration file

/ etc/ rc

File that controls automatic reboot

See Also
snmpext(3n), snmpd.conf(5n), snmpd(Sn)
Guide to Networking

8-352 Maintenance

startcpu (8)
Name
startcpu - start one or all CPUs

Syntax
/etc/startcpu [cpunumber]

Description
The startcpu program is used to start the non-boot CPUs in a system. Normally,
it will be executed from the / et c / rc startup file as the system comes up in multiuser mode.
If no argument is used, startcpu will try to start all the non-boot CPUs in the
system. Otherwise, it starts the specified CPU.

See Also
startcpu(2), stopcpu(2), stopcpu(8)

Maintenance 8-353

statd (8c)
Name
statd - network status monitor daemon

Syntax
lusr/etclstatd

Description
The s tat d daemon monitors the status of the client and server sites in response to a
request made by the localloekd daemon. When a site failure is detected, statd
notifies the localloekd daemon, which then processes the recovery of the locked
files or file regions.

Restrictions
The crash of a site is only detected on its recovery.

Files
/ete/sm
jete/sm.baek
jete/state

See Also
Guide to System Exercisers

Maintenance 8-361

syslog (8)
Name
syslog - log systems messages

Syntax
/etc/syslog [ -mN ] [ -fname ] [ -d ]

Description
The s y slog command reads a datagram socket and logs each line it reads into a set
of files described by the configuration file /ete/ syslog. eonf. The syslog
command configures when it starts up and whenever it receives a hangup signal.
Each message is one line. A message can contain a priority code, marked by a digit
in angle braces at the beginning of the line. Priorities are defined in < syslog.h >, as
follows:

LOG ALERT

This priority should essentially never be used. It applies only to
messages that are so important that every user should be aware
of them, for example, a serious hardware failure.

LOG SALERT

Messages of this priority should be issued only when immediate
attention is needed by a qualified system person, for example,
when some valuable system resource disappears. These
messages are sent to a list of system people.

LOG EMERG

Emergency messages are not sent to users, but represent major
conditions. An example might be hard disk failures. These
could be logged in a separate file so that critical conditions could
be easily scanned.

LOG ERR

These messages represent error conditions, such as soft disk
failures, etc.

LOG CRIT

Such messages contain critical information, but which can not be
classed as errors, for example, 'su' attempts. Messages of this
priority and higher are typically logged on the system console.

LOG WARNING These messages are issued when an abnormal condition has been
detected, but recovery can take place.

LOG NOTICE

These messages fall into the class of "important information";
this class is informational but important enough that you don't
want to throw it away casually. Messages without any priority
assigned to them are typically mapped into this priority.

LOG INFO

These are information level messages. These messages could be
thrown away without problems, but should be included if you
want to keep a close watch on your system.

LOG DEBUG

These messages may be useful to log certain debugging
information. Normally this information is thrown away.

It is expected that the kernel will not log anything below LOG_ERR priority.

The configuration file is in two sections separated by a blank line. The first section
defines files that syslog will log into. Each line contains a single digit which
defines the lowest priority (highest numbered priority) that this file will receive, an
8-362 Maintenance

syslog (8)
optional asterisk which guarantees that something gets output at least every 20
minutes, and a pathname. The second part of the file contains a list of users that will
be informed on SALERT level messages. For example, the following logs all
messages of priority 5 or higher onto the system console, including timing marks
every 20 minutes:
5*/dev/console
8/usr/spool/adm/syslog
3/usr/adm/critical
eric
kridle
kalash

This example logs all messages of priority 8 or higher into the file
/usr / spool/ adm/ syslog; and all messages of priority 3 or higher into
/usr / adm/ cri tical. The users "eric", "kridle", and "kalash" will be
informed on any subalert messages.
The flags are:

-m

Set the mark interval to N (default 20 minutes).

-f

Specify an alternate configuration file.

-d

Turn on debugging (if compiled in).

To bring s y slog down, it should be sent a terminate signal. It logs that it is going
down and then waits approximately 30 seconds for any additional messages to come
in.
There are some special messages that cause control functions. "<*>N" sets the
default message priority to N. "<$>" causes s y slog to reconfigure (equivalent to a
hangup signal). This can be used in a shell file run automatically early in the
morning to truncate the log.
The syslog command creates the file / etc/ syslog. pid if possible containing a
single line with its process ID. This can be used to kill or reconfigure syslog.

Restrictions
LOG_ALERT and LOG_SUBALERT messages should only be allowed to privileged
programs.
Actually, syslog can not deal with kernel error messages in the current
implementation.

Files
/etc/syslog.conf
Configuration file
/etc/syslog.pid
Process id

Maintenance 8-363

syslog (8)
See Also
syslog(3)

8-364 Maintenance

talkd (Bc)
Name
talkd - inter-terminal communications server

Syntax
letc/talkd

Description
The talkd program is the server for the talk (1) program. The server provides a
rendezvous method for the requesting (possibly remote) talk and the local
responding tal k .
The talkd server is invoked by inetd(8c) when it receives a packet on the port
indicated in the talk service specification.

Restrictions
The talkd server does not strictly follow network byte order in its packet format
and may have difficulty in talking with implementations of talkd on other
architectures that do not take this into account.
The version of talk released with ULTRIX V3.0 uses a protocol that is
incompatible with the protocol used in earlier versions. Starting with ULTRIX V3.0,
the talk program communicates with other machines running ULTRIX, V3.0 (and
later), and machines running 4.3 BSD or versions of UNIX based on 4.3 BSD.
The talk command is not 8-bit clean. Typing in DEC Multinational Characters
(DECMCS) causes the characters to echo as a sequence of a carets (") followed by
the character represented with its high bit cleared. This limitation makes tal k
unusable if you want to communicate using a language which has DECMCS
characters in its alphabet.

See Also
talk( I), services( 5), inetd(8c)

Maintenance 8-365

tapex(8)
Name
tapex - tape exerciser program

Syntax
tapex [ option(s) ] [parameter(s) ]

Description
The tapex program tests tape driver functionality. These tests provide more
comprehensive functional coverage than the mtx utility which does simple start/stop
oriented read/write testing. Functions that are tested include:
•

Writing records onto a tape and verifying the records

•

Using records in a range of sizes

•

Record -length testing

•

Random record-size testing

•

Positioning tests for records and files

•

Writing and reading past the end of media

•

End-of-file testing

•

n-buffered I/O testing

•

Tape-transportability testing

•

Bandwidth performance analysis

•

Media loader testing

•

Reporting of tape contents

When tapex is run, a writable tape must be loaded in the drive being tested, and the
drive must be online.

Options
Some tapex options cause specific tests to be performed, for example, an end-ofmedia test. Other options modify the tests, for example, enabling caching. The
tapex options are as follows:

-a

Performance measurement test that calculates the tape transfer bandwidth for
writes and reads to the tape by timing data transfers.

-b

Continuously runs the write/read tests until the process is killed. This flag can
be used in conjunction with the -r or -g flag.

-c

Enables caching on the device, where supported. This does not specifically test
caching, but it enables the use of caching on a tape device while running the
other tests.

-C

Disables caching on TMSCP tape devices. If the tape device is a TMSCP unit,
then caching is the default mode of test operation. This flag causes the tests to
be run in noncaching mode.

8-366 Maintenance

tapex(8)
-d

Tests the ability to append to the media. First, the test writes records to the
tape. Then, it repositions back one record and appends additional records.
Finally, the test does a read verification. This test simulates the behavior of the
tar r switch.

-e

End-of-media test. This test first writes data to fill up a tape, which may take a
long time for long tapes. It then does reads and writes past the end of media,
which should fail. Next it enables writing past end of media, writes to the tape,
and reads back the records for validation.

-E

Runs an extensive series of tests in sequential order. Due to the large number
of tests, this option takes a long time to complete. Depending on tape type and
cpu type, this series of tests can take up to 10 hours to complete.

-f /dev/rmt#?FP
Specifies the name of the device special file that corresponds to the tape unit
being tested. The number sign (#) symbol represents the unit number. The
question mark (?) argument can be the letter h for the high density device or
the letter 1 for the low density device. The default tape device is
/dev/rmtOh.
-F

File-positioning tests. First, files are written to the tape and verified. Next,
every other file on the tape is read. Then, the previously unread files are read
by traversing the tape backwards. Finally, random numbers are generated; the
tape is positioned to those locations, and the data is verified. Each file uses a
different record size.

-G

File-positioning tests on already-written tape. This flag can be used in
conjunction with the - F flag to run the file position tests on a tape that has
already been written to by a previous version of the - F test. For this to work,
the same test parameters, for example record size and number of files, must be
used as when the the tape was written. No other data should have been written
to the tape since the previous - F test.

-g

Random record-size tests. This test writes records of random sizes. It reads in
the tape, specifying a large read size; however, only the amount of data in the
randomly-sized record should be returned. This test only checks return values
and does not validate record contents.

-b

Displays a help message describing the tape exerciser.

-i

Interactive mode. Under this mode, the user is prompted for various test
parameters. Typical parameters include the record size and the number of
records to write. The following scaling factors are allowed:
k or K

for kilobyte (1024 * n)

b orB

for block (512 * n)

morM

for megabyte (1024 * 1024 * n)

For example, 10k would specify 10240 bytes.
-j

Write phase of the tape-transportability tests. This test writes a number of files
to the tape, and then verifies the tape. After a successful verification, the tape
is brought offline to be moved to another tape unit and read in with the - k
option. The purpose of this test is to prove that a tape can be written on one
drive and read by another drive. Note that the test parameters for the - k phase
of the transportability test must match the parameters of the - j test. Any
Maintenance 8-367

tapex(8)
changes of test parameters from the defaults should also be changed during the
-k test.
-k

Read phase of the tape-transportability tests. This test reads a tape that was
written by the - j test and verifies that the expected data is read from the tape.
Success of this test proves that a tape can be written on one drive and read on
another. As stated in the the description of the - j option, any parameters
changed in the - j test must also be changed in the - k test.

-L

Media loader test. For sequential stack loaders, the media is loaded, written to,
and verified. Then the media is unloaded, and the test repeats on the next piece
of media. This verifies that all the media in the input deck is writable. To run
this test in read-only mode, also specify the -w option.

-I

End-of-file test. This test verifies that a zero byte count is returned when a tape
mark is read and that another read will fetch the first record of the next tape
file.

-m

Displays tape contents. This is not a test; it reads the tape sequentially and
prints out the number of files on the tape, the number of records in each file,
and the size of the records within the file. The contents of the tape records are
not examined.

-N

Disables the usage of n-buffered I/O on tests that support its usage. (See
nbuf(4) for a description of n-buffered I/O.)

-0 filename

Sends output to the specified filename. The default is to not create an output
file and send output to the terminal.
-p

Runs both the record and file positioning tests. (See the -R and -F options.)

-q

Command timeout test. This test verifies that the driver allows enough time
for completion of long operations. The test consists of writing files to fill up
the tape. Next a rewind is performed followed by a forward skip out to the last
file. The test is successful if the forward skip operation completes without
error.

-r

Record-size test. A number of records are written to the tape and then verified.
This process is repeated over a range of record sizes.

-R

Record-positioning test. First, records are written to the tape and verified.
Next, every other record on the tape is read. Then, the other records are read by
traversing the tape backwards. Finally, random numbers are generated; the
tape is positioned to those locations, and the data is verified.

-s

Record-size behavior test. Verifies that a record read will return at most one
record or the read size, whichever is less.

-S

Single record size test. This option modifies the record-size test ( - r option).

-T

Copies output to standard output. This flag is useful if you want to log output
to a file with the -0 option and also have the output displayed on standard
output. This flag must be specified after the -0 flag in the command line.

-v

Verbose mode. This option causes more detailed terminal output of what the
tape exerciser is doing. For example, it lists operations the exerciser is
performing, such as record counts, and more detailed error information.

-V

Very verbose mode. This option causes more output to be generated than

8-368 Maintenance

tapex(8)
either the default mode or the -v flag. The output consists of additional status
information on exerciser operation.
-w

Opens the tape as read-only. This mode is only useful for tests that do not
write to the media. For example, it allows the -m test to be run on a writeprotected media.

-Z

Initializes read buffer to the nonzero value 0130. This may be useful for
debugging purposes. If the - z flag is not specified, all elements of the read
buffer will be initialized to O. Many of the tests first initialize their read buffer
and then perform the read operation. After reading a record from the tape,
some tests validate that the unused portions of the read buffer remain at the
value to which they were initialized. As a debugging tool, it may in some cases
be useful to have this initialized value set to be nonzero. In those cases, the
arbitrary character 0130 can be used.

Parameters
You can change the default test parameters either by using the - i option described
previously or by specifying the parameters in the command line. This section
describes the parameters you can set in the command line, listed with the associated
test.
To specify a value, type the parameter name followed by a space and then the
number. For example -min rs 512 specifies a minimum record size of 512 blocks.
The following scaling factorsare allowed:
k or K

for kilobyte (1024 * n)

b orB

for block (512 * n)

morM

for megabyte (1024 * 1024 * n)

For example, 10K would specify 10240 bytes.
These parameters are associated with the option -a:

-perf_num The number of records to write and read.
-perf_rs

The size of records.

These parameters are associated with the option -d:

-tar num The number of additional and appended records.
-tar size

The record size for all records written in this test.

These parameters are associated with the option -e.
Note that specifying too much data to be written past BOM could cause a reel-to-reel
tape to go off the end.
-end_ num The number or records written past BOM.
-end rs

The record size.

Maintenance 8-369

tapex(8)
These parameters are associated with the option· - F:

-num_fi
-pos_ra

The number of files.

-pos_rs
-ree fi

The record size.

The number of random repositions.
The number of records per file.

This parameter is associated with the option -g:

-rand_num The number of records to write and read.
These parameters are associated with the options - j and - k:

-tran file The number of files to write or read.
-tran_ree The number of records contained iIi each file.
-tran_rs

The size of each record.

These parameters are associated with the option - R:

-pos_num The number of records.
--pos_ra The number of random repositions.
-pos_rs
The record size.
these parameters are associated with the options - rand - S:

-ine
The record increment factor.
-max rs The maximum record size.
-min rs The minimum record size.
-nUin_ree The number of records.
-t

Sets a time limit in minutes on how long to run the record-size test ( -r
option). The default is to run the test to completion.

These parameters are associated with the option - s:

-num_ree The number of records.
-size ree The record size.
This parameter is used in any test which supports n-buffered I/O:

-num_nbuf The number of buffers to use.
This parameter is associated with all tests:

-err_lines The threshold on error printouts.

Examples
This example runs a series of tests on tape device rmt lh and sends all output to a
file called tapex. out.
tapex -f /dev/rrntlh -E -0 tapex.out

8-370 Maintenance

tapex(8)
This example runS the end-of-media test on tape device rmt4h. Verbose mode is
specified, which causes additional output. By default, output is directed to the
terminal.
tapex -f /dev/rmt4h -v -e

This example performs read/write record testing. By default, this test runs on the
default tape device /dev/rmtOh and·output is sent to the terminal.
tapex -r

This example performs read/write record testing using record sizes in the range 10k
to a maximum record size of 20k. By default, this test runs· on the default tape device
/ dev / rmt 0 h and output is sent to the terminal.
tapex -r -min_rs 10k -max_rs 20k