Path: senator-bedfellow.mit.edu!bloom-beacon.mit.edu!nycmny1-snh1.gtei.net!news.gtei.net!newsfeed.mathworks.com!enews.sgi.com!news.tamu.edu!sgi-faq
From: [email protected] (The SGI FAQ group)
Newsgroups: comp.sys.sgi.misc,comp.answers,news.answers
Subject: SGI admin Frequently Asked Questions (FAQ)
Supersedes: <[email protected]>
Followup-To: comp.sys.sgi.misc
Date: 6 Jul 2001 05:59:41 GMT
Organization: Visualization Lab, Texas A&M University
Lines: 2226
Approved: [email protected]
Expires: 3 Aug 2001 06:00:15 GMT
Message-ID: <[email protected]>
Reply-To: [email protected] (The SGI FAQ group)
NNTP-Posting-Host: viz.tamu.edu
NNTP-Posting-Date: 6 Jul 2001 05:59:41 GMT
Originator: [email protected]
Xref: senator-bedfellow.mit.edu comp.sys.sgi.misc:57361 comp.answers:46106 news.answers:210658

Archive-name: sgi/faq/admin
Last-modified: Tue Jun 20  1:00:03 CDT 2000
Posting-Frequency: Twice monthly
URL: http://www-viz.tamu.edu/~sgi-faq/

   SGI admin Frequently Asked Questions (FAQ)

This is one of the Silicon Graphics FAQ series, which consists of:

   SGI admin FAQ - IRIX system administration
   SGI apps FAQ - Applications and miscellaneous programming
   SGI audio FAQ - Audio applications and programming
   SGI diffs FAQ - Changes to the other FAQs since the last posting
   SGI graphics FAQ - Graphics and user environment customization
   SGI hardware FAQ - Hardware
   SGI impressario FAQ - IRIS Impressario
   SGI inventor FAQ - IRIS Inventor
   SGI misc FAQ - Introduction & miscellaneous information
   SGI movie FAQ - Movies
   SGI performer FAQ - IRIS Performer
   SGI pointer FAQ - Pointer to the other FAQs
   SGI security FAQ - IRIX security

Read the misc FAQ for information about the FAQs themselves. Each FAQ is
posted to comp.sys.sgi.misc and to the news.answers and comp.answers
newsgroups (whose purpose is to store FAQs) twice per month. If you
can't find one of the FAQs with your news program, you can get it from

   ftp://viz.tamu.edu/pub/sgi/faq/
   ftp://rtfm.mit.edu/pub/usenet/news.answers/sgi/faq/

(rtfm.mit.edu is home to many other FAQs and informational documents,
and is a good place to look if you can't find an answer here.) The FAQs
are on the World Wide Web at

   http://www-viz.tamu.edu/~sgi-faq/

If you can't use FTP or WWW, send mail to [email protected] with
the word 'help' on a line by itself in the text, and it will send you a
document describing how to get files from rtfm.mit.edu by mail. Send the
command 'send usenet/news.answers/sgi/faq/misc' to get the SGI misc FAQ,
and similarly for the other FAQs. Send the command 'send
usenet/news.answers/internet-services/access-via-email' to get the
"Accessing the Internet by E-Mail FAQ".

You may distribute the SGI FAQs freely and we encourage you to do so.
However, you must keep them intact, including headers and this notice,
and you must not charge for or profit from them. Contact us for other
arrangements. We can't be responsible for copies of the SGI FAQs at
sites which we do not control, and copies published on paper or CD-ROM
are certain to be out of date. The contents are accurate as far as we
know, but the usual disclaimers apply. Send additions and changes to
[email protected].

Topics covered in this FAQ:
---------------------------
  -1- DIAGNOSTICS
  -2- How can I determine which release of IRIX I'm running and which
      patches are applied?
  -3- How can I determine my SGI's Ethernet (and/or FDDI) address?
  -4- My SGI crashed and generated a file, /usr/adm/crash/vmcore.1. How
      can I examine this file to see what crashed my system?
  -5- How can I find out just about everything about my system at once?
  -6- DISKS
  -7- How big can files and filesystems be?
  -8- My XFS filesystem is corrupt or inconsistent.  How do I fix it?
  -9- Does IRIX support sparse files?
 -10- Why is /debug or /proc full of huge files?
 -11- How do I remount the /proc filesystem after accidentally
      unmounting it?
 -12- How do I extend an existing filesystem onto a new disk?
 -13- How do I know if I need more memory and/or swap space?
 -14- How much swap space should I have per megabyte of memory?
 -15- How can I increase my swap space?
 -16- What are virtual and logical swap space? How do they work in IRIX
      3.x, 4.0.x and 5.x?
 -17- Why is there no way to set up a RAM disk under IRIX?
 -18- BOOTING
 -19- How can I boot directly into single-user mode?
 -20- How can I boot from a non-default disk?
 -21- How can I boot my machine using a server on the other side of a
      router?
 -22- How do I make a bootable tape from an IRIX CD?
 -23- Why can't I boot one of the stand-alone programs on a tape or CD?
 -24- INSTALLING
 -25- Is it possible to remotely install IRIX over a network?
 -26- Which IRIX CD is the program 'foo' on?
 -27- How can I extract a single file from an 'inst' subsystem?
 -28- Why doesn't 'inst' work?
 -29- Why doesn't 'inst' work remotely?
 -30- I just installed a new version of IRIX and file X is missing.
      What should I do?
 -31- I reinstalled an IRIX subsystem to restore a missing file or get
      rid of a corrupted file, but it didn't help. Why not?
 -32- Why do 'inst' and 'showprods' say that a subsystem is an "Unknown
      product entry"?
 -33- How can I install IRIX onto a second disk which I can then move
      to another machine?
 -34- How can I copy my system disk onto a second disk which I can then
      move to another machine?
 -35- How can I share parts of IRIX among multiple machines?
 -36- Can I install a "tardist" archive without the "tardist" software?
 -37- I'm running 6.5.x, why can't I install the compilers?
 -38- NETWORKING
 -39- Why isn't my network working?
 -40- How can I measure my network's reliability?
 -41- How do I add a static route?
 -42- How can I make the 'slip' command advertise the Ethernet address
      of the SLIP client?
 -43- I've just edited inetd.conf, and nothing changed. Why?
 -44- Why can't I 'rdist' files between Suns and SGIs?
 -45- Why are there "satwrite failure: inetd" messages in my SYSLOG in
      IRIX 5.3?
 -46- What is causing the "get interface flags" error in my syslog?
 -47- Why are network connections between SGIs and Suns much slower
      than between SGIs and SGIs or Suns and Suns?
 -48- How can I set up IP aliases?
 -49- Where is ipfilterd documented?
 -50- MAIL
 -51- How can I set up 'sendmail' to pass (or not pass) 8-bit
      characters?
 -52- Why are my mailbox files changing ownership?
 -53- Why isn't a valid user getting their mail?
 -54- How can SGIs and Suns share a mail spool?
 -55- What's an "unknown mailer error"?
 -56- What's "mailbox: Error 0"?
 -57- Why am I having problems with my NFS-mounted mail spool?
 -58- Why are incoming mail addresses case-sensitive in IRIX 5.3?
 -59- NFS
 -60- How can I tell what hostname to use in /etc/exports?
 -61- Why can't I export an NFS-mounted filesystem?
 -62- Why can't Ultrix automount SGI filesystems?
 -63- Why does 'tar' work strangely on a filesystem mounted from an
      SGI?
 -64- Why does 'df' report incorrect numbers for a filesystem mounted
      from a Sun?
 -65- Is 'pcnfsd' available for the SGI?
 -66- Can I export a CD-ROM from my SGI to a non-SGI?
 -67- How can I read an IRIX (EFS) CD-ROM on a machine which doesn't
      use EFS?
 -68- How can I get quotas to work on an NFS filesystem?
 -69- Why can't some NFS clients NFS-mount IRIX 5.2 filesystems?
 -70- What's NFS3?
 -71- Why can't Solaris 2.5 clients read directories on NFS3-mounted
      IRIX 6.2 filesystems?
 -72- Why does my NFS server print "fhtovp_end VFS_GET failed" or
      "fhtovp_end vp NULL" to the console/SYSLOG?
 -73- PRINTING
 -74- Why can't lp(1) read my file?
 -75- How can I tell 'lp' to turn banner printing or page reversal off
      or on?
 -76- How can I use 'lpr' instead of or as well as 'lp'?
 -77- How can I print to a printer with a JetDirect card?
 -78- Why won't lpr(1) print very large files?
 -79- SGI DAEMONS
 -80- Why isn't the objectserver working?
 -81- What is sending packets to the sgi-dog.mcast.net multicast
      address?
 -82- Why are the objectserver and directoryserver taking up so much
      memory in IRIX 5.3?
 -83- Why don't the desktop admin tools recognize root as a privileged
      user?
 -84- Why doesn't my desktop deal properly with the CD or tape icons?
 -85- Why doesn't my mediad start up under IRIX 5.3?
 -86- Why do cdman, cdplayer and capture say "permission denied"?
 -87- Why can't 'cformat' format floptical disks, and/or why can't
      'datman' play audio DATs?
 -88- Why is famd hammering my NFS server?
 -89- What is causing "/dev/imon: event queue overflow" in my syslog?
 -90- BUGS
 -91- Why do 'who', 'rusers', etc. show users who aren't really logged
      in?
 -92- What's wrong with ftpd in IRIX 5.2?
 -93- Why isn't /usr/adm/SYSLOG being updated?
 -94- I just edited /etc/inittab, and now I can't start up or shut down
      my SGI! What's wrong?
 -95- Why does timed say "bind: Cannot assign requested address"?
 -96- Why does "ALERT: i2cWaitForXferDone: Xfer Done timeout ERROR"
      appear in my SYSLOG?
 -97- Why has NIS (YP) stopped working under IRIX 6.x?
 -98- MISCELLANEOUS
 -99- How do I set the number of processes allowed on my machine?
-100- Where can I get a termcap file for 'iris-ansi-net' to install on
      my non-SGI system?
-101- How can I make my SGI understand strange terminal types from
      other Unix systems?
-102- Can I change my login shell or other password-file info without
      being superuser?
-103- How can I administer my Iris without a graphics terminal?
-104- How can I use the visual admin tools on a system with graphics to
      administer a system without graphics?
-105- What's /etc/ioctl.syscon?
-106- Can I log out users automatically?
-107- How can I change kernel variables and/or rebuild my kernel?

----------------------------------------------------------------------

Subject:    -1- DIAGNOSTICS
Date: 15 May 1994 00:00:01 EST

 These questions discuss how to find out things about your system.

------------------------------

Subject:    -2- How can I determine which release of IRIX I'm running
               and which patches are applied?
Date: 28 Apr 2000 11:58:53 PDT

 On pre-6.5 systems, a 'uname -a' will give you the Irix version of
 the currently running kernel. The 6.5.x release added the '-R' flag
 which reports the current revision number (e.g. 6.5.6m). See the
 uname(1) manpage for other options.

 Of more general use, since the running kernel does not always
 reflect installed software, is the versions(1M) command.  'versions'
 with no arguments provides a full listing all the installed software
 components.  'versions -b | grep patch ' provides a brief listing of
 installed patches.  Note that the use of patches is depricated in the
 6.5 release stream.  Software subsystems are now "updated" so in with
 a few exceptions (e.g. systems running legato networker), there should
 be no patches installed.

 IRIX 5.2's System Manager ('chost') has the IRIX version number under
 "IRIX Version" and a listing of installed software under "Software"
 (the "Show Installed" button).

------------------------------

Subject:    -3- How can I determine my SGI's Ethernet (and/or FDDI)
               address?
Date: 4 Jun 1997 00:00:01 EST

 Many thanks to Miguel Sanchez <[email protected]> for
 providing the original version of the following discussion, and to
 Dave Olson <[email protected]> for comments. Andrew Cherenson
 <[email protected]> reminded us that all these methods except the first
 apply to FDDI as well, but we'll just say "Ethernet" below.

 Every system on an Ethernet network must have a unique Ethernet
 address for the network to operate properly. The physical Ethernet
 address of your system is the unique number assigned to the Ethernet
 hardware on your system. This unique number is assigned to the
 manufacturer of your Ethernet hardware by the IEEE (formerly by
 Xerox, one of the original developers of Ethernet). This is not to be
 confused with the IP address, which can be set arbitrarily.

 You may need to determine your system's Ethernet address if your
 network manager requires it before connecting your system to a
 network.  How to do so depends on whether IRIX is running and what
 operating system version is loaded.  Method 1 only provides the
 Ethernet address of the primary interface.  If you have multiple
 Ethernet interfaces (boards) in a system, use method 2, 3, 4 or 5 to
 determine the address(es) of any other interface(s).

 METHOD 1: eaddr

     If IRIX is not running, and the system is a Personal IRIS (4D20,
     25, 30, or 35), Indigo, Crimson, Onyx, Challenge, Indy, O2,
     or Indigo2, you can obtain the Ethernet address by typing 'eaddr'
     (older machines) or 'printenv eaddr' (newer) at the PROM monitor .
     prompt.  On some machines (4D30 or later) you can say 'nvram eaddr'
     while IRIX is running to get the same result.

 METHOD 2: netstat

     Under IRIX 4.0.1 or later, you can use the netstat command. For
     example,

     % /usr/etc/netstat -ia
     Name  Mtu    Network   Address            Ipkts   Ierrs   Opkts  Oerrs  Coll
     ec0   1500   siligrph  luey7              7765678 21648  384477     0  30338
                           192.48.200.251
                           192.0.0.1
                           08:00:69:06:17:c2
     lo0   32880  loopback localhost           41438       0   41438     0      0
                           192.0.0.1

     As seen on the fourth address line, the address of the system
     luey7's primary Ethernet interface, "ec0", is 08:00:69:06:17:c2.

 METHOD 3: arp

     You can obtain the Ethernet address of a Silicon Graphics system
     by using another system on your network. 'ping' the system whose
     Ethernet address you want, then use 'arp'. For example,

     % /usr/etc/ping -c 1 luey6
     PING luey6.sgi.com (192.48.200.250): 56 data bytes
     64 bytes from 192.48.200.250: icmp_seq=0 ttl=255 time=0 ms
     ----luey6.sgi.com PING Statistics----
     3 packets transmitted, 3 packets received, 0% packet loss
     round-trip (ms)  min/avg/max = 0/0/0
     % /usr/etc/arp luey6
     luey6 (192.48.200.250) at 8:0:69:6:c:40
     %

 METHOD 4: NetVizualyzer/FDDIVizualyzer and the like

     SGI's NetVizualyzer/FDDIVizualyzer network monitoring software
     and at least one public domain equivalent ('netman', at
     ftp://ftp.cs.curtin.edu.au/pub/netman/) allow you to find the
     Ethernet address corresponding to any IP address. Read the
     manual.

 METHOD 5: System Manager

     The Network Setup part ('cnet') of the Indigo Magic System Manager
     tool ('chost') shows the Ethernet address of each interface.

 4DDN: A Special Case

     DECnet uses a one-to-one relationship between the DECnet node ID
     and the Ethernet address. If the DECnet address is changed the
     Ethernet address is changed. DECnet Ethernet addresses always
     start with aa:, so you can identify systems running DECnet with
     'arp -a'.

     4DDN is Silicon Graphics' DECnet interconnection product. The
     Ethernet address of an IRIS running 4DDN will change when 4DDN is
     started.  Method 1 will return the original Ethernet address for
     the system.  Methods 2-5 will show the Ethernet address currently
     in use.

 sysinfo

     /etc/sysinfo is intended to return a unique identifier, which on
     some machines includes part or all of the Ethernet address. This
     is best regarded as an amusing coincidence, like HAL's name in
     "2001".  Don't rely on it.

 You can find an Ethernet address from a program most efficiently by
 using the SIOCGIFADDR ioctl on a raw socket (SOCK_RAW) using the
 RAWPROTO_SNOOP protocol (thanks to David Peter
 <[email protected]>) but the program must run as root. If you
 can't run as root, call one of the above programs with system().

------------------------------

Subject:    -4- My SGI crashed and generated a file,
               /usr/adm/crash/vmcore.1. How can I examine this file to
               see what crashed my system?
Date: 28 Apr 2000 11:58:53 PDT

 Under Irix 5.3 and above a system crash will generate a number of
 files in the /var/adm/crash directory. These files appear in numbered
 sets, with each number (n) corresponding to an event. These files
 include: unix.n (a copy of the kernel), vmcore.n.comp (a compressed
 dump of the memory), and analysis.n (an analysis file generated by the
 system using the icrash kernel debugger). Note that these files are
 created only is savecore is chkconfig'd on.

 Simply viewing the analysis file can often lead to clues regarding the
 reason behind a system crash. 'icrash' can also be run manually. See the
 icrash(1) manpage for details.

 SGI's 'icrash' utility can generate helpful reports and allow
 experienced administrators and support people to sift through the core
 file.  Get patch 813 for IRIX 5.3 or patch 769 for IRIX 6.1.  icrash
 is part of eoe.sw.unix (and therefore installed by default) on IRIX
 6.2 and above.

 For machines which lack the icrash utility:

   dbx -k /var/adm/crash/{unix,vmcore}.#
   t
   &putbuf/1000s

 Some machines have a special 'dbx' for crash dumps,
 /usr/adm/crash/dbx.  If it exists, use it instead of /usr/bin/dbx.

 The IRIX 5.x Electronic Services package includes a script, 'crpt',
 which does this and more automagically. A copy of the IRIX 5.2
 version lives at ftp://viz.tamu.edu/pub/sgi/software/admin/crpt.

------------------------------

Subject:    -5- How can I find out just about everything about my system
               at once?
Date: 06 Apr 1996 00:00:01 EST

 IRIX comes with hinv(1M) and gfxinfo(1G). If that's not enough, get
 Michael Cooper <[email protected]>'s 'sysinfo', recently ported
 to IRIX, from ftp://usc.edu/pub/sysinfo/. Change "PROG" in the
 Makefile from "sysinfo" to "Sysinfo" so it doesn't get in the way of
 /sbin/sysinfo, which isn't nearly as entertaining but which other
 software may require. This is probably also a good place to mention
 Dave Olson's 'scsicontrol', at
 http://reality.sgi.com/employees/olson/Olson/.

------------------------------

Subject:    -6- DISKS
Date: 15 May 1994 00:00:01 EST

 These questions deal with disks and swap space.

------------------------------

Subject:    -7- How big can files and filesystems be?
Date: 15 Jun 1996 00:00:01 EST

 A file on an EFS filesystem can be only 2G in size. The filesystem can
 be no more than 8G in size, whether or not it is on a logical volume.
 A physical or logical volume can be larger than 8G, but not usefully
 so: mkfs still can't make a filesystem larger than 8G on it. (Each
 block (512 bytes) has a unique number which must fit into 24 bits; 512
 * 2^24 = 8G. See also inode(4).)

 Note that 'fx' had a bug in versions of IRIX before 4.0.5H which
 prevented it from *exercising* disks larger than 2G, but other 'fx'
 functions and other parts of IRIX have no trouble with disks
 (filesystems) up to 8G.

 Files and filesystems on SGI's new XFS filesystem can be 1 terabyte
 (1K gigabytes) in size. On 64-bit machines running IRIX 6.2, a file
 will be able to be 9 exabytes (9 gigagigabytes) in size and a
 filesystem will be able to be 18 exabytes in size.  See the misc FAQ
 under "WWW pages" for a pointer to a demonstration.

------------------------------

Subject:    -8- My XFS filesystem is corrupt or inconsistent.  How do I
               fix it?
Date: 28 Apr 2000 11:58:53 PDT

 Under IRIX 6.2, get the latest XFS rollup patch and install it.
 (At the time of this writing, the latest XFS rollup patch was patch
 1768.)  This patch comes with xfs_repair, a tool that can often fix
 corrupted XFS filesystems.  (If the partition you wish to repair is
 your boot partition, you will have to boot from the miniroot that
 comes with the patch and run xfs_repair from it.)  Actual installation
 of the patch requires the installation of one of the recent kernel
 rollup patches; read the XFS rollup patch release notes before doing
 anything else with it.

 Under the 6.5.x release stream, 'xfs_repair' and 'xfs_check' are
 installed by default as part of eoe.sw.base. Please note that xfs
 check and repair operations are performed on unmounted filesystems.
 See the xfs_repair(1) manpage for details. Also, it is best to
 do an initial scan on a damaged filesystem using 'xfs_repair -n',
 (no modify mode) as extremely corrupted filesystems can occasionally
 cause an xfs_repair to fail with a core dump.

------------------------------

Subject:    -9- Does IRIX support sparse files?
Date: 09 Dec 1995 00:00:01 EST

 EFS does not; XFS does.

------------------------------

Subject:   -10- Why is /debug or /proc full of huge files?
Date: 10 Dec 1993 00:00:01 EST

 Those aren't disk files, they're interfaces to running processes.
 Read the debug(4) (IRIX 4.0.x) and/or proc(4) (IRIX 5.x) manpages.

------------------------------

Subject:   -11- How do I remount the /proc filesystem after accidentally
               unmounting it?
Date: 4 Jun 1997 00:00:01 EST

 Run /etc/mntproc as root.

------------------------------

Subject:   -12- How do I extend an existing filesystem onto a new disk?
Date: 28 Apr 2000 11:58:53 PDT

 Back up the existing filesystem (just in case) then run 'mklv' and
 'growfs'. 'mklv' and 'growfs' are nondestructive, so you don't need
 to restore the backup unless you screw up. Don't use 'mkfs', which
 does destroy existing data.

 Under modern IRIX versions (w/ the XFS filesystem) lv (the original
 logical volume manager) has been replaced with xlv. The requisite
 subsystems are _not_ installed by default. If you wish to use logical
 volumes, you must install eoe.sw.xlv, and if you wish to use mirroring
 (RAID 1), you must also install eoe.sw.xlvplex. This option requires a
 license be purchased from sgi.

 To simply grow the filesystem to another disk, you must first create a
 concatenated logical volume. This will change the volume headers of the
 target disks and let the system know that they should be treated as one
 volume. You can then use an 'xfs_growfs' to expand the filesystem on to
 the new disk. see the manpages for xlv_make(1), xlv_mgr(1), and
 xfs_growfs(1) for more details.

 Please not that it is NOT POSSIBLE to have a successfully striped or
 concatenated root filesystem. The only available logical volume
 configuration for the root filesystem is a mirror (RAID1). Any
 attempt to use another logical volume configuration will most
 likely work until the next system reboot. Since the Logal volume
 management software funtions on the OS level, it is unavailable
 during the inital phases of the boot process. Syncing of the
 mirrored filesystem will occur after the system comes up.

------------------------------

Subject:   -13- How do I know if I need more memory and/or swap space?
Date: 20 Feb 1994 00:00:01 EST

 If processes are killed due to lack of memory/swap, you need more
 memory and/or swap space. If your CPU is always waiting for swapping
 (run 'osview' and look at the "%Swap" entry under "Wait Ratio") you
 need more memory.

------------------------------

Subject:   -14- How much swap space should I have per megabyte of
               memory?
Date: 4 Jun 1997 00:00:01 EST

 An oft-recommended ratio is X memory:2.5 X swap, but this may be too
 low. Decide how much of your favorite program (plus IRIX) needs to
 be resident for good performance and how much doesn't, and make sure
 you have enough memory for the former and enough memory plus swap for
 the latter. Put "rmem" and "swp" in your ~/.grosview file, run
 'gr_osview' and run your favorite program to see what it needs.

------------------------------

Subject:   -15- How can I increase my swap space?
Date: 28 Apr 2000 11:58:53 PDT

 The Jan/Feb 1993 and May/Jun 94 Pipelines have detailed writeups on how
 to do this in IRIX 4.0.x and 5.x respectively. The Jul/Aug Pipeline
 has a correction to the latter article. If you like you can call the
 TAC and have them fax you the very latest version.

 It is also possible to add swapfiles within the filesystem structure.
 You can allocate space for swap in the filesystem by doing a 'mkfile'
 and adding this file to the system as swap space. See mkfile(1) for more
 information. The created file can be added by using a 'swap -a <file>'.
 If you intend this to be a permanent resource, make sure that you add
 an appropriate line to the /etc/fstab file. See the fstab(4) manpage
 for details.

------------------------------

Subject:   -16- What are virtual and logical swap space? How do they
               work in IRIX 3.x, 4.0.x and 5.x?
Date: 05 Jul 1994 00:00:01 EST

 Two terms whose meanings should already be clear: Physical swap space
 is an area on disk, either a partition or (in IRIX 5.x) a swap file.
 Virtual memory is the sum of physical memory and swap space.

 IRIX 3.x accepts a memory request only if enough virtual memory is
 free. Even if a process isn't using most of the memory it requested
 (which happens often, e.g. when a large process forks and execs a
 small process, or with Fortran 77 programs which allocate all storage
 statically), its memory is unavailable to other processes until it
 exits. IRIX 3.x has no virtual or logical swap space.

 In IRIX 4.0.x, IRIX accepts every memory request, and does not
 allocate virtual memory until a process actually tries to use it.
 This allows programs which request more memory than they use to run
 with much less memory than would otherwise be required. If too many
 processes actually use their memory requests so that virtual memory
 is in danger of filling up, IRIX kills one or more processes. IRIX
 usually kills the process which is using the most virtual memory,
 which may well not be the process which most recently requested
 virtual memory.

 IRIX 5.x works like IRIX 4.0.x, but one can set the amount of virtual
 memory which IRIX is allowed to overallocate. This amount is called
 "virtual swap space". "Logical swap space" is the sum of physical and
 virtual swap. There is no virtual swap space by default, so IRIX 5.x
 behaves like IRIX 3.x. One can set virtual swap to any amount of
 memory; if it is set sufficiently high, memory requests will always
 be granted, just like IRIX 4.0.x. Using jargon retroactively, IRIX
 4.0.x has an infinite amount of virtual swap space.

 Large or infinite amounts of virtual swap space work well for many
 people, because most programs don't use all the virtual memory they
 request, at least not at once. If your programs do use all their
 virtual memory, they'll be killed and you'll see "Process killed due
 to insufficient memory/swap" messages in your SYSLOG.

 Under IRIX 4.0.x, you can only turn virtual swap off completely by
 setting the kernel variable availsmem_accounting to 1. Doing so makes
 IRIX 4.0.x behave like IRIX 3.x, allocating memory only if it is
 actually available.

 Under IRIX 5.x, you can turn virtual swap on or off by doing
 'chkconfig vswap off' or 'chkconfig vswap on', or change the size of
 virtual swap by editing /etc/config/vswap.options, and rebooting. You
 can also use 'swap -v' to do any of these things directly and without
 rebooting.

 Remember that IRIX 5.x comes with virtual swap turned off and set to
 zero.  If you were happy with IRIX 4.0.x, you should turn virtual
 swap on and set its size to a very large number. If programs are
 killed, decrease the size of virtual swap or turn it off.

 See the swap(1M) and swapctl(2) manpages for details.

------------------------------

Subject:   -17- Why is there no way to set up a RAM disk under IRIX?
Date: 29 Apr 1995 00:00:01 EST

 IRIX has no way to set up a RAM disk (a personal computerish term for
 a part of memory which can be used like a very fast disk) because it
 already has a sophisticated disk caching mechanism, which will
 probably do a better job of minimizing disk reads than you would if
 you were deciding what to put on a RAM disk.

------------------------------

Subject:   -18- BOOTING
Date: 15 May 1994 00:00:01 EST

 As the song says, "There must be fifty ways to boot your Iris."

------------------------------

Subject:   -19- How can I boot directly into single-user mode?
Date: 8 Mar 1996 00:00:01 CST

 Use the PROM monitor's 'single' command.

 For machines earlier than 4D35s, whose PROMs don't have that command,
 first boot into sash from the PROM monitor with "boot", then type
 "boot initstate=s".

------------------------------

Subject:   -20- How can I boot from a non-default disk?
Date: 31 May 1995 00:00:01 CST

 Says Justin Mason <[email protected]>: If your disk is SCSI ID 4, do

   boot -f dksc(0,4,8)sash dksc(0,4,0)unix root=dks0d4s0

 or

   setenv bootfile dksc(0,4,8)sash
   setenv path dksc(0,4,8)
   setenv root dks0d4s0                # This is the tricky part
   auto

 from the PROM. The first method works once, so that subsequent
 reboots use SCSI ID 1, and the second method sets the PROM to boot
 from ID 4 every time (until you reset the PROM variables).

 Dave Olson <[email protected]> adds: "If you have a separate / and /usr,
 you'll probably want to add 'initstate=s' at least the first time, or
 the usr partition on your normal root will be mounted.  Boot in single
 user, then do 'cd /dev; ./MAKEDEV disklinks; multi'.  After this, you
 won't need 'initstate=s' since the link will now be to the usr
 partition on the alternate boot disk.

------------------------------

Subject:   -21- How can I boot my machine using a server on the other
               side of a router?
Date: 24 Jan 1994 00:00:01 EST

 Tell the router to forward BOOTP packets. If it can't, NFS-mount the
 remote volumes on another machine on the same subnet and use the
 nearby machine for your boot server.

------------------------------

Subject:   -22- How do I make a bootable tape from an IRIX CD?
Date: 03 Sep 1995 00:00:01 EST

 See the Sep/Oct 1993 Pipeline and/or
 ftp://viz.tamu.edu/pub/sgi/software/admin/making-bootable-tape for a
 detailed description, or just follow Dave Olson <[email protected]>'s
 summary: Take a look at the distcp(1M) manpage, and do something like

     tapehost# mount -o ro cdhost:/CDROM /mnt
     tapehost# distcp /mnt/dist /dev/nrtape

 Note that 'fx', 'ide', and 'sash' for all machines are in the dist/sa
 file.  'sa' is an image of the first part of the tape; use 'mkbootape
 -f sa -l' to see the contents.

------------------------------

Subject:   -23- Why can't I boot one of the stand-alone programs on a
               tape or CD?
Date: 03 Apr 1994 00:00:01 EST

 One reason is that some CPU names are preceded by periods and some
 aren't. Another is that the Indigo R4000 and later CPUs use the
 suffix 'ARCS', not 'IP20' or whatever as one might expect from
 'hinv'. For example, the correct command to boot fx directly from the
 PROM monitor on an Indigo R4000 is 'boot -f dksc(ctlr,unit,8)sashARCS
 dksc(ctlr,unit,7)stand/fx.ARCS'. Note the use of 'ARCS' instead of
 'IP20' and the missing period in 'sashARCS'.

------------------------------

Subject:   -24- INSTALLING
Date: 15 May 1994 00:00:01 EST

 These questions discuss software installation.

------------------------------

Subject:   -25- Is it possible to remotely install IRIX over a network?
Date: 20 May 1993 00:00:01 CST

 Yes. You can install IRIX from a remote machine which has a CD-ROM, a
 tape drive, or an IRIX distribution directory.  All of these
 scenarios (and several others) are described in detail in the "IRIS
 Software Installation Guide".  Examples are provided.

------------------------------

Subject:   -26- Which IRIX CD is the program 'foo' on?
Date: 25 May 1994 00:00:01 EST

 Mount the CD and try 'grep foo /CDROM/dist/*.idb'. If you don't get
 any output, 'foo' isn't on that CD. If you do, it is, and one of the
 fields is the subsystem in which 'foo' lives. Entries in *.idb files
 don't have a leading slash so you must leave it out if you grep for a
 full path, e.g. 'grep usr/bin/lp /CDROM/dist/*.idb', not 'grep
 /usr/bin/lp /CDROM/dist/*.idb'.

------------------------------

Subject:   -27- How can I extract a single file from an 'inst'
               subsystem?
Date: 05 Feb 1995 00:00:01 EST

 Under IRIX 5.2 and earlier, you need to be clever. 'inst' guru Paul
 Jackson <[email protected]> reveals all:

 - Find the subsystem in which the file lives, as described in the
   previous question. For this example we'll extract /sbin/ed, which
   lives in eoe1.sw.unix.

 - Follow the bouncing prompt:

   > su
   > cd /usr/tmp
   > mkdir -p tmproot/var/inst
   > inst -f /CDROM/dist/eoe1 -r /usr/tmp/tmproot
   > Inst> keep *
   > Inst> install eoe1.sw.unix
   > Inst> go
   > Inst> q
   > ls -l /usr/tmp/tmproot/sbin/ed
   -rwxr-xr-x    1 root     sys        75480 May 24 13:57 /usr/tmp/tmproot/sbin/ed

 - Move your file somwhere else and 'rm -r /usr/tmp/tmproot'.

 - That was under IRIX 5.x. Under IRIX 4.0.x or earlier, use
   '/usr/tmp/tmproot/usr/lib/inst' for a temporary inst directory
   instead of '/usr/tmp/tmproot/var/inst'.

 In IRIX 5.3, inst's '-Y' flag allows you to install a single file. See
 inst(1M) for more.

------------------------------

Subject:   -28- Why doesn't 'inst' work?
Date: 16 Jan 1994 00:00:01 EST

 One possibility is that you're using an old 'inst' with new
 software.  Always use an 'inst' at least as new as what you're
 installing.

------------------------------

Subject:   -29- Why doesn't 'inst' work remotely?
Date: 05 May 1994 00:00:01 EST

 Usually because it can't log in to the machine with the distribution
 media.  'inst' uses the guest account to do so, so make sure that
 guest on the machine on which you want to install software can rlogin
 to guest on the machine with the distribution media without a
 password.

------------------------------

Subject:   -30- I just installed a new version of IRIX and file X is
               missing. What should I do?
Date: 14 Oct 1995 00:00:01 EST

 'inst' doesn't deal well with some cases where a file moves from one
 subsystem to another between IRIX versions. Upgrading may cause such a
 file to disappear entirely. To bring it back, reinstall the subsystem
 to which the file belongs in the newer IRIX with "set neweroverride
 on" in 'inst' (or, equivalently, remove and reinstall the subsystem).
 You might want to install only the missing file; see the entry on
 installing a single file above.

 Popular examples of this include several headers in /usr/include/sys
 in IRIX 4.0.5/IDO 4.1.1, which are in eoe1.sw.unix, the mount(1M)
 manpage in IRIX 5.3, which is in eoe1.man.unix, and
 /usr/lib/X11/dyDDX/glx.so, also in 5.3, which is in x_eoe.sw.Server.

 One way to detect files which have disappeared in this manner in IRIX
 5.3 is 'showfiles -B'.

------------------------------

Subject:   -31- I reinstalled an IRIX subsystem to restore a missing
               file or get rid of a corrupted file, but it didn't help.
               Why not?
Date: 13 Apr 1994 00:00:01 EST

 'inst' doesn't bother to install a subsystem if the same or a newer
 version is already installed. Tell it to install anyway by saying
 'set neweroverride' before you say 'go'. Removing the subsystem and
 reinstalling it will do more or less the same thing.

------------------------------

Subject:   -32- Why do 'inst' and 'showprods' say that a subsystem is an
               "Unknown product entry"?
Date: 12 May 1996 00:00:01 EST

 Paul "Mr. inst" Jackson of SGI <[email protected]> explains two reasons:

 - The "orphan" subsystem is a special subsystem for directories in
   which more than one product keeps files. inst will remove these
   directories as they become empty. Do NOT remove the subsystem
   yourself. In fact, you can only see it due to a cosmetic bug in some
   versions of inst.

 - The product spec file in /var/inst which describes that subsystem is
   damaged, due either to the usual sort of file-damaging problems or
   to a bug in IRIX 5.3's inst. This also makes inst think that that
   product is version 0, and thus incompatible with every other
   subsystem. There is no patch for the bug. The workaround is to
   reinstall the affected subsystems with all files restricted, so that
   no files are actually installed but the product spec file is
   updated, like so:

   inst -f whatever -X/
   Inst> set exclusions                # Should be set to "Value /"
   Inst> keep *
   Inst> install upgrade
   Inst> list i                        # Should see the unknown entries
   Inst> go
   Inst> quit

------------------------------

Subject:   -33- How can I install IRIX onto a second disk which I can
               then move to another machine?
Date: 20 Jan 1994 00:00:01 EST

 With difficulty. Many parts of the installation process assume that
 you're installing IRIX onto your system disk (SCSI ID 1). Just fiddle
 with SCSI ID switches and/or move disks around to make the disk onto
 which you want to install IRIX the system disk for the duration of
 the installation.

 Furthermore, IRIX has many hardware dependencies, so you should only
 move system disks between absolutely identical machines. If you want
 to make a system disk for a machine without a network connection,
 CD-ROM or tape drive, the easiest and safest way is to borrow another
 CD-ROM or tape drive.

 If you want to try anyway, Justin Mason <[email protected]> reports that
 the following works under IRIX 5.1.1:

 Set up the disk, e.g. with SCSI id 4, fx a generic "[bo]otable"
 partition setup onto it, and mkfs the partitions. Copy sash, etc.
 from your system disk to the new disk with dvhtool. Boot up the
 miniroot as usual, go into inst, choose "admin" from the menu and do
 the following, replacing SCSI IDs and partition numbers as
 appropriate:

   umount /root
   umount /root/usr
   mount /dev/dsk/dks0d4s0 /root
   mount /dev/dsk/dks0d4s6 /root/usr
   mount                               # Just to check
   return                              # Go back to main inst menu

 Then install as you like.

------------------------------

Subject:   -34- How can I copy my system disk onto a second disk which I
               can then move to another machine?
Date: 03 Sep 1995 00:00:01 EST

 See the article in the Jul/Aug 1992 Pipeline and the addendum in the
 Nov/Dec 1992 Pipeline, and note that the warning about hardware
 dependencies in the previous question applies here too.  Steve
 Kotsopoulos <[email protected]> has written a script which does
 this automatically; you can FTP it from
 ftp://viz.tamu.edu/pub/sgi/software/admin/clonedisk. Be sure to read
 the comments before running it!

 If your machine has FlashPROMs, which are normally updated by 'inst',
 you'll need to update them yourself; see flashio(1M) and heed its
 warnings.

------------------------------

Subject:   -35- How can I share parts of IRIX among multiple machines?
Date: 15 May 1994 00:00:01 EST

 Diskless machines, for one. Share trees, for another. Look in the
 misc FAQ under "WWW pages" for the "IRIX share trees" WWW page.

------------------------------

Subject:   -36- Can I install a "tardist" archive without the "tardist"
               software?
Date: 23 Sep 1995 00:00:01 EST

 Certainly. A tardist archive is just a tar archive of an 'inst'
 distribution.  Retrieve the tardist file directly to disk (in Mosaic
 or Netscape one does this by shift-clicking the link to the tardist
 file), name it something.tar, untar it as you would any tar file, and
 install it as you would any inst distribution.

------------------------------

Subject:   -37- I'm running 6.5.x, why can't I install the compilers?
Date: 28 Apr 2000 11:58:53 PDT

 This is a common problem, having to do with the way that inst deals
 with version numbering. Essentially, the distributions on the
 Development Foundation and Development Libraries are looking for the
 stock 6.5 version of eoe.sw.base. In order to install successfully
 from these distributions, you must have the following distributions
 open (inst under 6.5 supports having multiple open distributions):
 irix-6.5-foundation-1, and overlay disks for your current os level
 (e.g irix-6.5.6-installation-tools-and-overlays-1-of-2-11-99, and
 irix-6.5.6-overlays-2-of-2-11-99)

 Once these distributions are open, you should issue a keep *, so
 that nothing will be installed from these distributions. You should
 now be able to install the development system with minimal difficulty.

------------------------------

Subject:   -38- NETWORKING
Date: 15 May 1994 00:00:01 EST

 These questions discuss general networking.

------------------------------

Subject:   -39- Why isn't my network working?
Date: 03 Sep 1995 00:00:01 EST

 A list of good things to try is at
 ftp://viz.tamu.edu/pub/sgi/software/admin/network-checklist.

------------------------------

Subject:   -40- How can I measure my network's reliability?
Date: 22 Oct 1995 00:00:01 EST

 Don't worry about collisions. They are part of normal operation on a
 crowded Ethernet. You *should* worry about late collisions (which are
 logged to the console) and lost packets (which you can easily measure
 with the command 'ping -fs 3000 -c 1000 someotherhost'), which usually
 mean network hardware problems or a misconfigured bridge or router.

 See the Ethernet FAQ (posted to comp.dcom.lans.ethernet and FTPable
 from ftp://rtfm.mit.edu/pub/usenet/comp.dcom.lans.ethernet/ and other
 news.answers archives), Chapter 18 of the "IRIX Advanced Site and
 Server Administration Guide" and the Sep/Oct 1993 Pipeline for more.

------------------------------

Subject:   -41- How do I add a static route?
Date: 07 Sep 1996 00:00:01 EST

 Some sites handle IP routing by designating a routing machine and
 having all other hosts define a static route to that machine. The way
 to do this on SGIs is in the /etc/init.d/network.local script.

 1) Read the paragraph just before the copyright at the top of
    /etc/init.d/network and make the links it specifies.

 2) Put something like the following in /etc/init.d/network.local,
    replacing ROUTER'S.IP.ADDRESS.HERE with the address of your router.
    Under IRIX 6.x, omit the "1" at the end of the same line.

    #! /bin/sh
    IS_ON=/sbin/chkconfig
    case "$1" in
    'start')
        if $IS_ON network; then        # network must be chkconfig'ed on
            /usr/etc/route add default ROUTER'S.IP.ADDRESS.HERE 1
        fi
        ;;
    'stop')
        /usr/etc/route delete default ROUTER'S.IP.ADDRESS.HERE ;;
    *)
        echo "Usage: $0 {start|stop}" ;;
    esac

 Check the script with 'sh -v /etc/init.d/network.local'.

 The above setup will not allow you to NFS-mount (or unmount) disks
 from the other side of the static route gracefully, because the route
 will be added after the mount attempt during startup and deleted
 before the unmount attempt during shutdown. Instead of putting the
 routing commands in a separate script, put them in /etc/init.d/network
 itself. Put the 'route add' just after the "$ROUTE $RFLUSHFLAG" line
 and the 'route delete' after the '/sbin killall mount ...' line.

 There is no reason to define a static route and also to run routed. If
 you do so, routed will delete the static route and you'll be confused.
 'chkconfig routed off'.

 Under IRIX 6.2 or later, remove the '1' from the end of the 'route
 add' line.

------------------------------

Subject:   -42- How can I make the 'slip' command advertise the Ethernet
               address of the SLIP client?
Date: 10 Dec 1993 00:00:01 EST

 You can't. Just add something like

   /usr/etc/arp -s $USER `netstat -ia | grep :` pub

 to the shell script in which you start the SLIP process. $USER is the
 SLIP client. The 'netstat | grep' part gets the host's Ethernet
 address, and 'arp' advertises the host as an ARP server for $USER.
 See also the arp(1M) manpage.

------------------------------

Subject:   -43- I've just edited inetd.conf, and nothing changed. Why?
Date: 15 Feb 1995 00:00:01 EST

 You need to make 'inetd' reread inetd.conf. Do 'killall -HUP inetd'.
 If that doesn't work, comment out the entry you modified, 'killall
 -HUP inetd', uncomment the entry and 'killall -HUP inetd' again.  If
 that doesn't work, 'killall -TERM inetd'. If that doesn't work,
 reboot.

------------------------------

Subject:   -44- Why can't I 'rdist' files between Suns and SGIs?
Date: 30 Jan 1995 00:00:01 EST

 Sun's 'rdist' expects SGI's 'rdist' to live in /usr/ucb, but it's
 actually in /usr/bsd. Make a symbolic link from /usr/ucb/rdist to
 /usr/bsd/rdist and all will be well.

 In IRIX 5.3 or later, you may need to use ordist(1).

------------------------------

Subject:   -45- Why are there "satwrite failure: inetd" messages in my
               SYSLOG in IRIX 5.3?
Date: 4 Jun 1997 00:00:01 EST

 Says Ellen Desmond <[email protected]>:
 This is due to a bug in inetd that is manifest only when eoe2.sw.audit
 has been installed.  Inetd calls satwrite() to generate an audit
 record every time it spawns a server.  The bug is that the call to
 satwrite() is made after the process has done a setuid() to the user
 specified for that server in inetd.conf.  Because satwrite() is a
 superuser-only call, it fails for servers that don't run as root, such
 as fingerd.

 The bug is fixed by patch 1268. If you don't use auditing, you can just
 ignore the messages or remove the eoe2.sw.audit subsystem and rebuild
 your kernel.

------------------------------

Subject:   -46- What is causing the "get interface flags" error in my
               syslog?
Date: 06 Apr 1996 00:00:01 EST

 Walter Roberson <[email protected]> says: As part of its
 licensing verification, Framemaker 4.0 attempts to get interface flags
 on a fixed ethernet interface instead of using the hardware inventory
 functions to figure out what the name of the ethernet is. This results
 in an error like

 broadcast: ioctl (get interface flags): No such device or address

 Speak with Frame Technologies.

 Patch 1092 includes fixes for "frame license daemon" and might fix the
 above problem; we don't know yet.

------------------------------

Subject:   -47- Why are network connections between SGIs and Suns much
               slower than between SGIs and SGIs or Suns and Suns?
Date: 13 Aug 1995 00:00:01 EST

 Dave Olson <[email protected]> explains: Some older Sun systems (I don't
 remember which, but not current systems) have ethernet *chips* (this
 isn't an OS issue) that can't handle the recommended 9.6 usec
 interpacket gap. This is not a problem when sending packets, but when
 receiving packets from faster systems (like SGI Indigos and later) it
 results in lost packets, retransmission and major slowdown.

------------------------------

Subject:   -48- How can I set up IP aliases?
Date: Wed Sep 22 16:50:53 CDT 1999

 An IP alias is an extra IP address which may be in a completely
 different subnet and domain than a host's primary IP address. This is
 often desired by WWW presence providers. An IP aliasing is a step
 beyond a DNS alias, which is an extra hostname for the same IP
 address. HTTP requests do not include the host name, so a server has
 no way of knowing the host name to which a particular request was
 sent. A server *can* determine the IP address to which a particular
 request was sent and respond accordingly, so IP aliasing allows one to
 have multiple independent servers on a single machine.

 Patch 1356 to IRIX 5.3 allows IP aliasing. IRIX 6.2 and later does IP
 aliasing without patches. The interim solution once available from
 Silicon Surf is no longer available.

 See also

     http://www.amazing.com/internet/virtual-homer.html
     http://www.apache.org/docs/virtual-host.html
     http://www.thesphere.com/~dlp/TwoServers/

 for general information and descriptions of other implementations.

------------------------------

Subject:   -49- Where is ipfilterd documented?
Date: 15 Jun 1996 00:00:01 EST

 In IRIX 6.1 and earlier, ipfilterd itself is part of eoe.sw.ipgate but
 the ipfilterd(1M) manpage is part of eoe2.sw.netman and the
 netsnoop(1M) manpage (ipfilterd uses netsnoop syntax) is part of the
 extra-cost NetVisualyzer package. A copy of netsnoop(1M) is at
 ftp://viz.tamu.edu/pub/sgi/software/admin/netsnoop.1, and both
 manpages are in the uman database (see the misc FAQ under "WWW
 pages"). In IRIX 6.2 and later, ipfilterd(1M), netsnoop and
 netsnoop(1M) are all part of the base OS.

 Note that patch 1249 fixes a number of ipfilterd problems.

------------------------------

Subject:   -50- MAIL
Date: 15 May 1994 00:00:01 EST

 These questions discuss mail configuration and problems.

------------------------------

Subject:   -51- How can I set up 'sendmail' to pass (or not pass) 8-bit
               characters?
Date: 30 Jan 1995 00:00:01 EST

 Many experts say "don't try". RFC822 requires mail transport agents to
 *clear* the eighth bit, and many hosts do.  Some which don't may crash
 when they get mail with the eighth bit set. Instead, use a
 MIME-compatible mail program. MIME, described in RFC1521, is a
 standard for enclosing non-RFC822 material in your mail. The apps FAQ
 discusses several mail programs which support it.

 In IRIXes up to 5.2, one can flout this doctrine by running sendmail
 with the '-o8' flag to allow 8-bit characters in message bodies (*not*
 headers). Read the description of that option in sendmail(1) for more
 reasons not to use it. In IRIX 5.3, sendmail runs in 8-bit mode by
 default, but one can use the '-o7' flag to get the old behavior.

------------------------------

Subject:   -52- Why are my mailbox files changing ownership?
Date: 15 Oct 1995 00:00:01 EST

 If your mail directory is mounted from another machine, your machine
 does not have root access to that directory, and the other machine has
 BSD-style "restricted chown" (the restricted_chown kernel variable is
 turned on or it's not an SGI) /bin/mail will change mail file
 ownership when delivering local mail. Without unrestricted chown *or*
 root access, /bin/mail is unable to give a mail file back to its owner
 after delivering mail. You can fix the problem by turning off re-
 stricted chown on the other machine (if it's an SGI) or exporting the
 mail directory with root access for your machine. This problem was to
 have been fixed in IRIX 5.2, but apparently is still present in 5.3.

------------------------------

Subject:   -53- Why isn't a valid user getting their mail?
Date: 24 Jan 1994 00:00:01 EST

 IRIX' mail system requires "valid users" to have both valid password
 file entries (whether local or via NIS) and home directories. The
 latter often trips one up when installing POP servers and whatnot,
 where home directories aren't really necessary. Just make a fake
 one.

------------------------------

Subject:   -54- How can SGIs and Suns share a mail spool?
Date: 05 Feb 1994 00:00:01 EST

 Paul Riddle <[email protected]> has written up how he did it. Read
 ftp://ftp.umbc.edu/pub/sgi/shared-spool.text.

------------------------------

Subject:   -55- What's an "unknown mailer error"?
Date: 20 Feb 1994 00:00:01 EST

 See ftp://viz.tamu.edu/pub/sgi/software/mail/mail-errors.

------------------------------

Subject:   -56- What's "mailbox: Error 0"?
Date: 05 Mar 1994 00:00:01 EST

 It's a harmless bug; don't worry about it. It is fixed in IRIX
 versions 4.0.5H/4.0.5IOP and later.

------------------------------

Subject:   -57- Why am I having problems with my NFS-mounted mail spool?
Date: 02 Mar 1996 00:00:01 EST

 Perhaps for one of the following reasons:

 - IRIX 5.2 NFS has a bug which prevents writing to a zero-length file
   if the writing process has group write permission but not group read
   (or user write) permission. IRIX 5.2 /bin/mail likes to set the
   permissions of /var/mail/<user> to mode 620, exactly what is needed
   to exercise the NFS bug. There is no patch for these bugs, but both
   are fixed in IRIX 5.3. Meanwhile, you can work around the problem by
   chmod'ing your mail files to 660 and setting the appropriate option
   in your mail program to make sure that empty mail files are not
   removed.

 - lockd must be running on the NFS server to allow programs on the
   client to lock mail files. If it is not, Zmail/MediaMail will
   complain that "file cannot be locked". One could tell MediaMail to
   not request a lock with "unset dot_lock", but that would risk mail
   file corruption.

 - IRIX 5.3 lockd is broken. Typical symptoms include "Warning: could
   not lock /var/mail/user after 100 trys." (from /usr/sbin/Mail) and
   the presence of user.lock and user.rolock files in /var/mail. Patch
   1128 fixes at least some of these problems.  If the client and
   server are both IRIX 5.3, you can also work around the problem by
   changing the mount type to NFS3. (See notes on NFS3 elsewhere in
   this FAQ.)

 See also "NFS Mounting Mail with IRIX 5.3 and IRIX 6.0.1" in the
 Sep/Oct 1995 Pipeline.

------------------------------

Subject:   -58- Why are incoming mail addresses case-sensitive in IRIX
               5.3?
Date: 31 May 1995 00:00:01 EST

 It's a configuration change. To change it back, find each line in your
 /etc/sendmail.cf which begins with "M"; these are mailer definitions.
 Each has a field beginning with "F="; these are the mailer flags.
 Remove the "u" from each mailer flags field.

------------------------------

Subject:   -59- NFS
Date: 15 May 1994 00:00:01 EST

 These questions discuss NFS.

------------------------------

Subject:   -60- How can I tell what hostname to use in /etc/exports?
Date: 07 Feb 1994 00:00:01 EST

 NFS servers may need a particular form of a client's name in
 /etc/exports to allow the client access. This may not be obvious, for
 example if the server is also a router. Log in from the client to the
 server and say 'echo $REMOTEHOST' to see what the server thinks the
 client is called, and put that in /etc/exports.

 The System Manager ('chost') should be able to determine the correct
 hostname for you.

------------------------------

Subject:   -61- Why can't I export an NFS-mounted filesystem?
Date: 10 Dec 1993 00:00:01 CST

 This is known as multi-hop NFS. It is not allowed or supported in
 (Sun's) NFS because it is not in general possible to detect errors
 such as infinite mount loops, on either the client or the server.

------------------------------

Subject:   -62- Why can't Ultrix automount SGI filesystems?
Date: 10 Dec 1993 00:00:01 CST

 Ultrix's automount uses an "untrusted" port for mount requests. Add
 an '-n' to the mountd lines in /usr/etc/inetd.conf (/etc/inetd.conf
 in IRIX 5.x), like so:

 mountd/1    stream  rpc/tcp wait    root    /usr/etc/rpc.mountd     mountd -n
 mountd/1    dgram   rpc/udp wait    root    /usr/etc/rpc.mountd     mountd -n

 then 'killall mountd' and 'killall -HUP inetd' or reboot.

------------------------------

Subject:   -63- Why does 'tar' work strangely on a filesystem mounted
               from an SGI?
Date: 15 Oct 1995 00:00:01 EST

 When user A extracts a file owned by user B from a tar archive, 'tar'
 makes the file owned by user A unless user A is the superuser.  Some
 systems allow users to give files away (e.g. IRIX); some do not
 (e.g. SunOS).  On some systems with the restricted behavior (SunOS
 among them), 'tar' tries to give the file to user B whether or not
 user A is the superuser, assuming that the chown system call will fail
 if user A is not.  This is not true if user A is using 'tar' on (e.g.)
 a Sun to extract files onto a filesystem NFS-mounted from (e.g.) an
 SGI.  'tar' may create zero-length files or give away directories and
 then be unable to extract files into them.

 Work around the problem by doing the 'tar' on the SGI or extracting
 onto a Sun filesystem. It is possible that third-party versions of
 'tar' (e.g. GNU tar) are smarter; if so, let us know. You could turn
 the restricted_chown kernel variable on on the SGI, but be aware that
 some programs may depend on unrestricted chown, notably /bin/mail as
 discussed elsewhere in this FAQ.

------------------------------

Subject:   -64- Why does 'df' report incorrect numbers for a filesystem
               mounted from a Sun?
Date: 29 Apr 1995 00:00:01 EST

 Sun's filesystem reserves space, usually 10%, and Sun's 'df' reports
 only the unreserved space. SGI's filesystem (efs) does not reserve
 space, so SGI's 'df' doesn't correct for reserved space.

------------------------------

Subject:   -65- Is 'pcnfsd' available for the SGI?
Date: 20 Apr 1996 00:00:01 EST

 For IRIX 4.0.x, look in ftp://ftp.sgi.com/support/pcnfsd.sysV/. (Note
 that although SGI makes this available, they do not support it.) For
 IRIX 5.x, look in ftp://viz.tamu.edu/pub/sgi/software/pcnfsd/ or
 http://www.sgi.com/Technology/Connectivity/pcnfsd.html. See also the
 pcnfsd bug mentioned in the security FAQ.

------------------------------

Subject:   -66- Can I export a CD-ROM from my SGI to a non-SGI?
Date: 10 Dec 1993 00:00:01 EST

 Not in IRIX 4.0.x. You can in IRIX 5.x, as you would any other
 filesystem.

------------------------------

Subject:   -67- How can I read an IRIX (EFS) CD-ROM on a machine which
               doesn't use EFS?
Date: 18 Jun 1995 00:00:01 EST

 You want 'efslook', at
 ftp://viz.tamu.edu/pub/sgi/software/efslook.tar.gz.

------------------------------

Subject:   -68- How can I get quotas to work on an NFS filesystem?
Date: 16 Dec 1994 00:00:01 EST

 Mount the filesystem with the 'quotas' option, by adding it to
 /etc/fstab or the automounter map as appropriate, and make sure the
 nfs.sw.nis subsystem, which contains the NFS quota daemon
 (/usr/etc/rpc.rquotad) is installed. That's nfs.sw.nis, not
 nfs.sw.nfs! See fstab(4), rquotad(1M) and perhaps automount(1M) for
 details.

------------------------------

Subject:   -69- Why can't some NFS clients NFS-mount IRIX 5.2
               filesystems?
Date: 07 Sep 1996 00:00:01 EST

 IN IRIX 5.2, SGI's mount daemons (/usr/etc/mount_*) can fool some NFS
 clients (TGV's Multinet NFS for VMS, for one) into thinking that the
 mount daemons are NFS servers and trying to connect to them as
 such. If the client can mount filesystems from your machine when no
 mount daemons are running, but not when they are, this is your
 problem. Fix it by upgrading to IRIX 5.3 or, if you're daring,
 stealing the mount daemons from an IRIX 5.3 machine.

 Zsolt Bagoly <[email protected]> reports that DEC OSF1 and Linux
 clients can work around this problem with the option "port=2049",
 e.g. 'mount -t nfs -o port=2049 server:/path /mount-point'.

------------------------------

Subject:   -70- What's NFS3?
Date: 21 Nov 1995 00:00:01 EST

 An improved NFS which comes with IRIX 5.3. It is described in detail
 in ftp://gatekeeper.dec.com/pub/standards/nfs/NFS3.spec.ps.Z and the
 Nov/Dev 1995 Pipeline. Its quirks are still being discovered by users;
 one prominent quirk is that it is not supported by the automounter.
 Another is that a bug causes ~/.Xauthority files on an NFS3 volume to
 be truncated; patch 216 fixes the bug. Another is that it interacts
 poorly with XFS; patch 547 fixes that and many other problems.

------------------------------

Subject:   -71- Why can't Solaris 2.5 clients read directories on NFS3-
               mounted IRIX 6.2 filesystems?
Date: 8 Mar 1997 00:00:01 EST

 Solaris 2.5, a 32-bit operating system, can't understand 64-bit data
 from IRIX 6.2 NFS3 servers. This will be fixed in Solaris 2.6.
 Meanwhile, export the IRIX 6.2 filesystems with the 32bitclients
 option. See also 'relnotes nfs 5'.

------------------------------

Subject:   -72- Why does my NFS server print "fhtovp_end VFS_GET failed"
               or "fhtovp_end vp NULL" to the console/SYSLOG?
Date: 8 Mar 1997 00:00:01 CST

 Walter Roberson <[email protected]> explains:  You installed an
 NFS patch on your server, but you missed installing the NFS client
 patch on your 5.3 system.  You can either force the mount to use NFS2
 or patch your 5.3 NFS client system.

------------------------------

Subject:   -73- PRINTING
Date: 15 May 1994 00:00:01 EST

 These questions discuss printing.

------------------------------

Subject:   -74- Why can't lp(1) read my file?
Date: 10 Dec 1993 00:00:01 EST

 lp(1) is setuid, so it can only read world-readable files. You can say
 'lp < file' if you don't want to make your file world-readable.

------------------------------

Subject:   -75- How can I tell 'lp' to turn banner printing or page
               reversal off or on?
Date: 13 Nov 1994 00:00:01 EST

 'lp' controls printers via shell scripts, called 'models', which live
 in /var/spool/lp/model. When you install a printer, the appropriate
 model script is copied to /var/spool/lp/interface/<name-of-printer>.

 To temporarily change a printer's behavior, look at the manpage for
 its interface script (or, if there is none, the script itself) to see
 what options it wants, and pass them to the script with 'lp's '-o'
 option.  For example, 'lp -o"-nobanner" file' tells a "Generic
 Postscript" printer (described in the gpsinterface(1) manpage) to
 print 'file' without a banner page.

 To permanently change a printer's behavior, edit its interface
 script.  The following are true for "Generic Postscript" printers,
 but the idea is the same for others:

 - To turn banner printing off or on, change the line 'BANNER=1' to
   'BANNER=0' or vice versa.

 - To turn page reversal off or on, change the line
   'send=/usr/lib/print/lptops' to 'send="/usr/lib/print/lptops -U"'
   (note the quotes) or vice versa.

 In IRIX 5.x, you can change these settings in the printpanel. You can
 also turn banner printing off on a per-user basis by doing 'echo
 nobanner >> /var/spool/lp/settings/<printername>/<yourusername>'.

------------------------------

Subject:   -76- How can I use 'lpr' instead of or as well as 'lp'?
Date: 06 Apr 1996 00:00:01 EST

 IRIX is based on System V Unix and as such uses the lp(1) printing
 protocol. The lp software is sufficient (and preferred) for printing
 to local printers, printing to network printers attached to SGIs or
 other machines using the lp protocol, and acting as a print server to
 machines using the lp protocol. In IRIX 5.3/6.1 or later, it can also
 print to remote printers using the lpr protocol and act as an lpr
 server. SGI's standard lp interface scripts handle several common
 file formats (text, PostScript, RGB images) automatically, and
 Impressario (which works through the lp software) handles much more.
 The lp software is installed by default.

 lpr(1) is the BSD Unix printing protocol. SGI provides and supports
 the lpr software only for printing to network printers attached to
 computers which use the lpr protocol, e.g. a BSD Unix machine, an HP
 printer with an appropriately configured JetDirect card or a
 Macintosh running lpDaemon. It can also be used, unofficially, for
 controlling a local printer and for a print server. It comes with no
 support for handling different file formats at all. The lpr software
 is not installed by default. It is in the eoe2.sw.bsdlpr subsystem in
 IRIX 4.0.x and print.sw.bsdlpr in IRIX 5.x.

 You can use lpr to print to a local printer in one of two ways:

 - Set the printer up as an lp printer and write a printcap(4) entry
   with an output filter which is just a wrapper around lp. If that
   isn't crystal-clear, ask SGI for their writeup on "Integrating The
   AT&T Spooler With The BSD LPR Print Spooler". A not-guaranteed-to-
   be-up-to-date copy is at
   ftp://viz.tamu.edu/pub/sgi/software/print/lpr-to-lp.

 - Write a printcap entry and set up the printer as an lpr printer
   just as you would on a BSD system. SGI doesn't support this use of
   lpr, and comp.sys.sgi.* won't be much help either.

   Now that the printer is set up as an lpr printer, you won't be able
   to use lp with it directly. You can make lp(1) send files to a
   local lpr printer in one of two ways:

   - Use an lp interface script that calls lpr. You can use
     mkbsdpr(1M), which comes with Impressario 1.1 and later (see the
     Impressario FAQ), write one yourself or ask SGI for their writeup
     "LPTOLPR, A Model File for LP". A not-guaranteed-to-be-up-to-date
     copy of the last is at
     ftp://viz.tamu.edu/pub/sgi/software/print/lp-to-lpr.

   - Replace lp with a script which calls lpr. One such script is
     ftp://viz.tamu.edu/pub/sgi/software/print/lp-wrapper-for-lpr.

 You can print to an lpr server in one of two ways:

 - Under IRIX 5.3/6.1 or later, use the lp software. Just use
   printers(1M) to add the remote printer. The "list printer" button
   will not list lpr printers; just enter the name of the remote
   printer manually. This may or may not work for a printer with a
   JetDirect card; see the next entry.

 - Under older IRIXes, or if you're used to it, use the lpr software.
   See SGI's "Advanced Site and Server Administration Guide".

 You can set up an lpr server in one of two ways:

 - Under IRIX 5.3/6.1 or later, you don't need to, because the lp
   server is also an lpr server. This is officially documented only in
   'relnotes print 3'. There are some caveats:

   - Don't define printers to the lp and lpr software (i.e. in
     /var/spool/lp/interface and /etc/printcap) with the same name.
   - Clients must be in ~lp/.rhosts or /etc/hosts.equiv; lpsched
     won't look in /etc/hosts.lpd.
   - lpd must not be running when lpsched starts up. If you want to
     switch from lpd to lpsched, do '/etc/init.d/bsdlpr stop;
     /etc/init.d/lp stop; /etc/init.d/lp start'. /etc/init.d/bsdlpr
     can't be chkconfig'ed off, so if you want to leave the lpr
     software installed you'll have to move /etc/init.d/bsdlpr to
     another name so it won't run on bootup.
   - The client must send the control file before the data file. Most
     Unixes do; some PCs do not. It might be informative to run lpsched
     with the -debug flag, or just try it and see if it works. lpd
     accepts either file order; lpsched should, and will in IRIX 6.2.

 - Under older IRIXes, or if you can't get the previous solution to
   work, or if you're a BSD head, use the lpr software. Set up your
   local printer as an lpr printer (see above) and edit /etc/hosts.lpd
   to grant access to the clients. Again, SGI won't support this.

------------------------------

Subject:   -77- How can I print to a printer with a JetDirect card?
Date: 18 May 1996 00:00:01 EST

 Configure the JetDirect card to act as an lpr print server (see the
 JetDirect documentation) and set up your SGI to use it as you would
 any remote lpr printer (see above). You may also want to set up a
 bridge to the 'lp' system; again, see above.

 Unless you have a very recent JetDirect card, you will not be able
 to use printers(1M) to set up the JetDirect as a remote lpr printer.
 You can either use the lpr software on the SGI (see above), or you
 can circumvent printers(1M) in one of the following ways:

 - Use printers(1M) to add a normal 'lpr' print server with the local
   printer name you'll want to use for the JetDirect printer.
 - Do '/usr/lib/lputil add /dev/null netface <local-printer-name>'.
 - Do '/usr/lib/lpadmin -p<local-printer-name> -mnetface -v /dev/null',
   and create by hand all of the directories and little files in
   /var/spool/lp.

 Each of the above will create an interface script in
 /var/spool/lp/interface/<local-printer-name>. Since you subverted the
 usual installation process, some of the settings in the script won't
 be right yet. Change HOSTNAME to the JetDirect's hostname, NETTYPE to
 "bsd" and (thanks to Ken Harris <[email protected]> for this one) SYNC
 to 0. If it's a Postscript printer, change TYPE to "PostScript" and
 CONVTYPE to "PostScriptFile".

 Bill Poitras <[email protected]> adds: To print from a PC using SAMBA
 via your SGI, you can either modify the interface script so that
 specifying "-oraw" to lp (as suggested in the SAMBA documentation)
 will set CONVTYPE to nothing, or set up a second interface to the same
 printer and set its CONVTYPE to nothing permanently.

 If your JetDirect card is very old, it may not understand the lpr
 protocol at all. If it is very, very old, you may have to resort to
 sending the file over TCP port 9100 directly. If it is only very old,
 you may be able to upgrade the card to a version which does understand
 lpr. Upgrading the JetDirect is beyond the scope of this FAQ.

------------------------------

Subject:   -78- Why won't lpr(1) print very large files?
Date: 24 Aug 1996 00:00:01 EST

 lpd(1M) intentionally places an upper limit on spool file size. To
 remove it, add "mx#0" to your printcap(4) entry.

------------------------------

Subject:   -79- SGI DAEMONS
Date: 08 Oct 1995 00:00:01 EST

 This section deals with the objectserver and its tricky pals, the
 directoryserver, famd and mediad.

------------------------------

Subject:   -80- Why isn't the objectserver working?
Date: 24 Aug 1996 00:00:01 EST

 Install patch 1096. If you still have problems, read on.

 First, consider whether you really need the objectserver. Without it,
 you'll lose "business cards" and the graphical admin software. They're
 probably not worth the headache.

 Anne Eagle <[email protected]> posted most of the following:

 - Its database may be corrupt. If the objectserver appears to start
   OK but crashes later, this is probably the case. Rebuild it like
   so:

     /etc/init.d/cadmin stop
     /etc/init.d/cadmin clean
     /etc/init.d/cadmin start

   If the preceding doesn't work, try this

     /etc/init.d/cadmin stop
     mv /var/Cadmin/data /var/Cadmin/data.old
     /usr/Cadmin/bin/parseclasses
     /etc/init.d/cadmin start

   Note that either method destroys "Privileged User" and "Business
   Card" information. (This is the ONLY known drawback of rebuilding
   your objectserver database, and the ONLY reason why SGI
   documentation recommends that you consult with the TAC before doing
   so. For most people that means that there's no reason why you
   shouldn't rebuild whenever the need arises.)

 - One of your system configuration files (including but not limited to
   /etc/exports, /etc/fstab, /etc/inittab, /etc/mtab, /etc/passwd and
   /etc/printcap) may have minor format problems which don't bother
   IRIX proper but do bother the objectserver. Such problems include a
   last line which doesn't end with a linefeed, a backspace not
   preceded by a space in /etc/exports, or unprintable characters. Gary
   Lin <[email protected]> suggests that you ensure that /etc/exports
   has explicit -ro or -rw export options and that you remove
   continuation lines (\) from /etc/printcap. Ken Gant
   <[email protected]> points out that, as specified in gettydefs(4),
   the last line of /etc/gettydefs must be blank. One sign that you
   have such a problem is a core file in /var/Cadmin/data. If you find
   and fix a problem, rebuild the databases as above.

   If you can't find the problem, try the following:

     par -s -i -N open -l -SS /usr/Cadmin/bin/objectserver -d

   The last file objectserver opens is probably where the problem is.
   If you're really desperate, the TAC will give you an objectserver
   compiled with -g and help you run dbx on it.

 - You may be swamping the objectserver with NIS (YP) users. There are
   several ways around this:

   - Start a directoryserver on a machine on your local network.

   - Use netgroups or the "+user" form in /etc/passwd instead of just
     a "+" and rebuild the databases as above.

   - Most severely, remove the NIS object definition files so that the
     objectserver will not create NIS objects, rebuild the
     objectserver database (without the NIS objects) and restart the
     objectserver as follows. You will not be able to manipulate NIS
     users with Cadmin if you do this.

     killall fm
     mediad -k
     killall objectserver
     mv /var/Cadmin/data /var/Cadmin/data.orig
     cp -pr /usr/Cadmin/classes /usr/Cadmin/classes.orig
     rm /usr/Cadmin/classes/groupObject.op
     rm /usr/Cadmin/classes/nisAccountObject.op
     rm /usr/Cadmin/classes/peopleNISObject.op
     rm /usr/Cadmin/classes/peopleObject.op
     /usr/Cadmin/bin/parseclasses
     /usr/Cadmin/bin/objectserver
     ps -ef | grep obj

     Wait until you see 2 objectserver processes running, then do

     mediad
     fm -lrb &

 - Chris Riney <[email protected]> says: "We have just discovered
   here at our site that if you do not have a route defined for the
   SGI multicast subnet, then objectserver will gobble up memory.  I
   established a route for 224.0.0.0, and objectserver has been up for
   over a week without consuming additional memory." This route is
   defined in the stock /etc/init.d/network.

 - Andreas Klingler <[email protected]> fixed his
   objectserver by removing /usr/Cadmin/classes/printerObject.op and
   then rebuilding /var/Cadmin/data as above.

 - David Carrigan <[email protected]> fixed his objectserver by editing
   his /etc/passwd file so userids were in ascending order.

 - Tovar ? <[email protected]> suggests shutting off your objectserver,
   then running 'objecterver -d'.

 - Urpo Kotipalo <[email protected]> had trouble with shadow
   passwords and the objectserver, which he fixed by waiting until
   '/etc/init.d/cadmin clean' had finished running pwconv(1M) before
   doing '/etc/init.d/cadmin start'.


 See also "Indigo Magic Tips and Tricks" in the Sep/Oct 1994 Pipeline
 and the entry on the imon queue below.

------------------------------

Subject:   -81- What is sending packets to the sgi-dog.mcast.net
               multicast address?
Date: 06 Apr 1996 00:00:01 EST

 The objectserver. It's using that address intentionally; SGI just
 didn't bother to define a new one. Scott Henry <[email protected]> points
 out that if you don't use any directoryservers and want to get rid of
 the objectserver multicast packets, you can add '-t 0' to
 /etc/config/objectserver.options and they will go away.

------------------------------

Subject:   -82- Why are the objectserver and directoryserver taking up
               so much memory in IRIX 5.3?
Date: 02 Mar 1996 00:00:01 EST

 They really aren't. Doing 'ps -ef' in IRIX 5.3 and looking at the SZ
 column shows that these programs have extremely large address spaces,
 about 128M each. That's not the same thing as the physical memory they
 occupy, which is shown in the RSS column and which should be much
 smaller. See ps(1) and the IRIX 5.3 Cadmin release notes for more.

 However, there is a bug in the objectserver which makes it grow
 continuously when it can't find a directoryserver.  SGI claims to have
 fixed this in patch 1096; however, some disagree.

------------------------------

Subject:   -83- Why don't the desktop admin tools recognize root as a
               privileged user?
Date: 28 Jul 1995 00:00:01 EST

 The objectserver isn't working. Clean the database as described above.

------------------------------

Subject:   -84- Why doesn't my desktop deal properly with the CD or tape
               icons?
Date: 02 Mar 1996 00:00:01 EST

 Install patch 1096. If you still have problems, read on.  Thanks to
 Anne Eagle <[email protected]> for the following:

 This is a known bug both in 5.2 and 5.3.  Essentially, the desktop
 icons stop responding to insertion/removal of mediad, although the
 media itself is mounted properly and is accessible either from the
 command line or by opening your home directory and then changing to
 /CDROM or /floppy or starting cdman by hand or whatever is
 appropriate for the media.  Double clicking on the icon for the
 peripheral results in a dialog saying that the drive is unloaded,
 which of course it is not.

 Here are a number of workarounds. If the first doesn't work, try the
 second, and so on.  Below, if the command begins with the "%" symbol,
 that command can be executed by any user.  Commands beginning with
 "#" must be executed by root.

 * Restart the File Manager:

       % /usr/lib/desktop/telldesktop quit
       % fm -lrb

 * Restart both the File Manager and mediad:

       % /usr/lib/desktop/telldesktop quit
       % su -
       # /etc/init.d/mediad stop
       # /etc/init.d/mediad start
       # exit
       % fm -lrb

 * Restart the File Manager, mediad and the objectserver:

       % /usr/lib/desktop/telldesktop quit
       % su -
       # /etc/init.d/mediad stop
       # /etc/init.d/cadmin stop
       # /etc/init.d/cadmin start
       # /etc/init.d/mediad start
       # exit
       % fm -lrb

 * See if the icons from the Shared Resources in the upper right hand
   corner of the System Manager are responsive to insertion/removal of
   media.  If so, access the icons from the System Manager.

 * Reboot the system:

       # reboot

 * The media will have to be accessed by hand.  Open a directory
   view either by double-clicking on the home directory icon or
   choosing "Desktop/Home Directory" from the Toolchest.  Then change
   to the appropriate directory for the device.  For instance, the
   floppy or floptical drives are access by default from /floppy and
   the CDROM from /CDROM.

   Find the "cdman" icon via "Find/An Icon" and then double click on
   it to listen to music CD's over the scsi audio port.

 See also the Sep/Oct 1994 Pipeline.

 Greg Morlock <[email protected]> fixed his device icons by adding
 entries to the local copy of /etc/passwd for each NIS user, with no
 password and a '+' at the beginning of each userid.

 Alternatively, mediad may never have started; see the next question.

------------------------------

Subject:   -85- Why doesn't my mediad start up under IRIX 5.3?
Date: 04 Feb 1995 00:00:01 EST

 If your mediad is dying with the message "mediad: Initialization of my
 address failed.  Cannot contact objectserver." to the syslog, your
 machine may not be able to get its IP address.  Try adding the line

     hostresorder local bind

 or (if you use NIS)

     hostresorder local nis bind

 to your /etc/resolv.conf file.  Make sure the local machine's hostname
 and IP address are in the local /etc/hosts file, and then either
 restart the objectserver, mediad, and desktop or reboot.

 This problem manifests itself when the objectserver cannot grab the
 local machine's IP address from DNS (aka bind).  When the resolver is
 told to look at the /etc/hosts file first, it should find the local
 machine's address without asking DNS about it.

 Thanks to Dave Olson and Alexis Cousein (both of SGI) for tracking
 this down.

------------------------------

Subject:   -86- Why do cdman, cdplayer and capture say "permission
               denied"?
Date: 02 Feb 1996 00:00:01 EST

 You've installed patch 708 or 1048, which come with a broken mediad.
 You really should have patch 1048 or later, which fixes a bad security
 hole in the objectserver (see the security FAQ).  If you don't care
 about automounting CDs, turn off mediad with '/etc/init.d/mediad stop;
 /etc/chkconfig mediad off'. If you do, use the mediad from IRIX 5.3,
 (which you'll find in /var/inst/patchbase/usr/etc/mediad after
 installing the patch) or upgrade to patch 1096 (but see below).

------------------------------

Subject:   -87- Why can't 'cformat' format floptical disks, and/or why
               can't 'datman' play audio DATs?
Date: 16 Mar 1996 00:00:01 EST

 You've installed patch 1096, which breaks cformat(1M) and datman(1).
 You really should have patch 1048 or later, which fixes a bad security
 hole in the objectserver (see the security FAQ), so you should either
 use mkfp(1M) instead of cformat and turn off mediad(1M) when you want
 to use datman, or back down to patch 1048 (but see above).

------------------------------

Subject:   -88- Why is famd hammering my NFS server?
Date: 24 Oct 1995 00:00:01 EST

 It's partly a bug; get patch 165 for IRIX 5.2 and patch 166 for IRIX
 5.3. It's partly just famd's nature; you can try to calm it down by
 changing its polling interval (6 seconds by default, specified by the
 '-t 6' flag) in /etc/inetd.conf.  If you do install the patch, don't
 forget to add the '-l' flag to famd's line in /etc/inetd.conf.

------------------------------

Subject:   -89- What is causing "/dev/imon: event queue overflow" in my
               syslog?
Date: 24 Oct 1995 00:00:01 EST

 Probably famd. Try the measures in the previous entry. If those don't
 work, try increasing the value of QSIZE in /var/sysgen/master.d/imon
 and rebuilding your kernel.

------------------------------

Subject:   -90- BUGS
Date: 03 Jun 1995 00:00:01 EST

 These questions discuss miscellaneous bugs in IRIX. Note also the
 entry on bugs and patches in the misc FAQ, and Walter Roberson's patch
 lists in the "WWW pages" section of the same FAQ.

------------------------------

Subject:   -91- Why do 'who', 'rusers', etc. show users who aren't
               really logged in?
Date: 30 Jan 1995 00:00:01 EST

 Several bugs in different versions of IRIX cause /etc/utmp to not be
 updated properly after a user logs out. Programs incorrectly reporting
 logins are correctly reporting the incorrect contents of /etc/utmp.
 One such bug is present in an unspecified part of IRIX 4.0.x and fixed
 in IRIX 5.x, and another is in ftpd in IRIX 5.2 and fixed in patch 162
 and IRIX 5.3.

 If one can't fix the bug which causes the incorrect /etc/utmp
 entries, one can run a separate program to remove them from /etc/utmp
 after the fact. Jeremy DuBois <[email protected]>, David Hinds
 <[email protected]> and Patrick M. Ryan <[email protected]>
 have written such programs, which can be found in
 ftp://viz.tamu.edu/pub/sgi/software/utmp/. SGI provides such a
 program, /usr/sbin/chkutent, in IRIX 5.3; the root crontab runs it
 daily.

------------------------------

Subject:   -92- What's wrong with ftpd in IRIX 5.2?
Date: 30 Jan 1995 00:00:01 EST

 It doesn't maintain utmp properly (see the previous entry) and it dies
 during 'mget's. Get patch 162 or upgrade to 5.3.

------------------------------

Subject:   -93- Why isn't /usr/adm/SYSLOG being updated?
Date: 20 Dec 1996 00:00:01 CST

 Popular causes include:

 - running out of disk space. Once syslogd is unable to write to
   /usr/adm/SYSLOG, it won't try again until it is `killall -HUP
   syslogd`ed.

 - installing IRIX 4.0.x and failing to heed the nagging from
   the system when it is rebooted to run 'versions changed' and
   combine new and old configuration files.  In this case, the trouble
   is in /usr/spool/cron/crontabs/root.

 - Separating fields in /etc/syslog.conf with spaces instead of tabs.
   If you use spaces, syslogd will silently segv when it reads that
   file.

------------------------------

Subject:   -94- I just edited /etc/inittab, and now I can't start up or
               shut down my SGI! What's wrong?
Date: 03 Dec 1994 00:00:01 EST

 If the last line of /etc/inittab is a comment, init will screw up
 horribly.  If your machine is still running, remove the comment and
 everything will be OK. If not, go to the miniroot, run the shell and
 remove the comment from there. The following sequence of commands is
 one possible way to do this:

   cd /root/etc
   cat inittab         # Decide how many lines to remove (say three)
   wc inittab          # See how many lines inittab has (say 120)
   head -117 inittab > inittab.new   # Keep the first 120 - 3 lines
   mv inittab inittab.old
   mv inittab.new inittab
   cat inittab         # Just making sure

 and reboot. Don't forget the 'cd'; from the miniroot's point of view,
 /etc/inittab is /root/etc/inittab.

 The problem should be fixed in IRIX 5.3.

------------------------------

Subject:   -95- Why does timed say "bind: Cannot assign requested
               address"?
Date: 29 Oct 1994 00:00:01 EST

 timed is incompatible with the rld which comes with patchSG0000023,
 which is needed for DeltaCC. There are two solutions to this problem
 (thanks to Alan Davis <[email protected]>):

 - Get a new timed from SGI.

 - Replace the following line in /etc/init.d/network.options (line 664
   in an unmodified IRIX 5.2 file)

   /usr/etc/timed -M `cat $CONFIG/timed.options 2> /dev/null` &

   with

   env _RLD_ARGS="-clearstack" /usr/etc/timed `cat $CONFIG/timed.options 2> /dev/null` &

------------------------------

Subject:   -96- Why does "ALERT: i2cWaitForXferDone: Xfer Done timeout
               ERROR" appear in my SYSLOG?
Date: 22 Oct 1995 00:00:01 EST

 These messages appear when the Indycam is unplugged. They are
 harmless, but if they bother you you can banish them with patch 486.

------------------------------

Subject:   -97- Why has NIS (YP) stopped working under IRIX 6.x?
Date: 04 Jul 1996 00:00:01 EST

 Joshua Hart <[email protected]> writes: Under earlier IRIXes,
 NIS domain names could contain uppercase letters. Under IRIX 6.x they
 may not; only lowercase letters are permitted.

------------------------------

Subject:   -98- MISCELLANEOUS
Date: 15 May 1994 00:00:01 EST

 Everything else.

------------------------------

Subject:   -99- How do I set the number of processes allowed on my
               machine?
Date: 13 Nov 1994 00:00:01 EST

 Use systune(1M) to change 'nproc' (in the 'numproc' group of
 parameters) and reboot.

------------------------------

Subject:  -100- Where can I get a termcap file for 'iris-ansi-net' to
               install on my non-SGI system?
Date: 4 Jun 1997 00:00:01 CST

 SGIs use terminfo, so you need to translate the terminfo description
 to termcap. 'infocmp -Cr iris-ansi-net' will produce an iris-ansi-net
 termcap file. See infocmp(1) for more. Note that 'infocmp' is in the
 eoe2.sw.terminf subsystem in IRIX 5.x, 6.0.x, and 6.1, and is not
 installed by default.  (infocmp is part of eoe.sw.unix in IRIX 6.2
 and above and is installed by default.)

------------------------------

Subject:  -101- How can I make my SGI understand strange terminal types
               from other Unix systems?
Date: 4 Jun 1997 00:00:01 EST

 If the other system uses terminfo:
 Brent L. Bates <[email protected]> has been able to copy binary
 terminfo files directly between several different types of systems.
 (On IRIX 5.x and above, the terminfo files are located under
 /usr/share/lib/terminfo.)  However, if for some reason this does not
 work, use 'infocmp -I whatever > file' to extract the source
 erminfo entry for the terminal. Transfer the file to your SGI and
 do 'tic file' (as root) to put the entry into the terminfo database.

 If the other system uses termcap:
 Snip the termcap entry out of /etc/termcap (or wherever) with an
 editor, transfer it to your SGI and (as root) do 'captoinfo file
 > newfile' and 'tic newfile'.

 See also the infocmp(1), captoinfo(1), tic(1) and terminfo(4)
 manpages, and make sure you've installed eoe2.sw.terminf (under
 IRIX 5.x, 6.0.x, and 6.1), which is not installed by default.
 On IRIX 6.2 and above, these utilities are part of eoe.sw.unix.

 Many unusual terminal devices can be taken care of in IRIX 6.2 and
 above by installing eoe.sw.terminfo, which includes terminfo files
 for many devices.

------------------------------

Subject:  -102- Can I change my login shell or other password-file info
               without being superuser?
Date: 27 Jan 1996 00:00:01 EST

 At a graphics terminal, use the User Manager. At a text terminal, if
 your account is on NIS (Yellow Pages) use 'ypchpass'.  If it isn't
 you're stuck, because IRIX has no way to change a local user's
 password-file info from the command line.  You might ask your
 superuser to install one of the many free implementations of 'chsh'
 and/or 'chfn' (one is in volume 3 of comp.sources.unix, at
 ftp://ftp.uu.net/usenet/comp.sources.unix/volume3/) but they'll want
 to make sure that those programs know how to lock /etc/passwd
 properly.

-----------------------------

Subject:  -103- How can I administer my Iris without a graphics
               terminal?
Date: 13 Apr 1994 00:00:01 EST

 The visual admin tools in IRIX 4.0.x ('vadmin') need GL, and do not
 work on X terminals or workstations without GL. You can use 'sysadm'
 on text terminals for some tasks, but beware of bugs and
 inadequacies: SGI judged 'sysadm' to be too buggy to be worth
 updating for IRIX 5.x.

 The visual admin tools in IRIX 5.2 and later should display on any X
 display, *except* for the backup/restore tool which is an exact port
 from IRIX 4.0.x and requires GL. Some images will be missing when GL
 is unavailable, but the tools will function properly. As for text
 terminals, you're out of luck: 'sysadm' does not exist in IRIX 5.x.

 Of course, you can always use a text editor and write scripts, or see
 the next question.

------------------------------

Subject:  -104- How can I use the visual admin tools on a system with
               graphics to administer a system without graphics?
Date: 12 Feb 1994 00:00:01 EST

 rlogin to the graphics-less system and run 'vadmin' (IRIX 4.0.x) or
 'chost' (IRIX 5.x).  Make sure that the DISPLAY environment variable
 is set correctly and that both the vadmin/sysadmdesktop and the
 shared library subsystems are installed on the graphics-less system
 (which they are in the default installation).

 Under IRIX 5.x, look at the READMEs in /var/sysadmdesktop/rsysmanapps
 and /var/sysadmdesktop/sysmanapps to find out how to use 'chost' to
 run commands on remote systems. Finally, in a future release of IRIX
 5.x, the sysadmdesktop tools will be able to manage remote systems
 *without* doing an rlogin.

------------------------------

Subject:  -105- What's /etc/ioctl.syscon?
Date: 09 Apr 1995 00:00:01 EST

 The default 'stty' setings to use on the console in single user mode,
 in 'stty -g' format.

------------------------------

Subject:  -106- Can I log out users automatically?
Date: 29 Apr 1995 00:00:01 EST

 It depends on the sense of "log out" you want. If your users use the
 'tcsh' shell you can set the autologout shell variable (and possibly
 make it read-only) in a startup script to log out inactive shells.
 This won't do anything to an instantiation of tcsh which is waiting
 for another program to quit (e.g. a 'more' process in mid-file) or to
 programs other than tcsh (e.g. a 'jot' window), and it won't log one
 out of the graphics console. One can imagine a program similar to a
 screen saver which would log one out after so many minutes of
 keyboard/mouse inactivity, but we don't know of one. However, if
 you're only worried about security (rather than resource hogging) it
 may suffice to use 'xlock' to lock the screen rather than actually
 logging one out. For that matter, tcsh can also lock itself rather
 than logging one out; this wouldn't be helpful on a graphics console
 but might be on a simple terminal.

------------------------------

Subject:  -107- How can I change kernel variables and/or rebuild my
               kernel?
Date: 14 Oct 1995 00:00:01 EST

 To change a kernel variable under IRIX 5.x, use systune(1M). It saves
 the tuned kernel in /unix.install; if this file exists, it is moved to
 /unix on the next reboot. systune will change the parameters in the
 running kernel if it can, and tell you to reboot if it can't.

 To actually rebuild the kernel, use 'autoconfig -f'. This leaves the
 rebuilt kernel in /unix.install. You must reboot to begin using the
 rebuilt kernel; "rebuild your kernel" elsewhere in the FAQs implies
 rebooting.

 IRIX 4.0.x does not have systune. To change a kernel variable, edit
 the file in /usr/sysgen/master.d which contains the variable, rebuild
 the kernel and reboot.

------------------------------

End of sgi/faq/admin Digest
******************************
--
The SGI FAQ group <[email protected]>   http://www-viz.tamu.edu/~sgi-faq/
Finger us for info on the SGI FAQs, or look in ftp://viz.tamu.edu/pub/sgi/.

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.