Backup Server: 4.142.2.58: [0] The ‘shmget’ call failed with error number 22 (Invalid argument). Refer to your operating system documentation for further details.

To confirm, if maximum shared memory has been manually set,

# projects -l

user.asetest
projid : 101
comment: “Sybase ASE”
users  : (none)
groups : (none)
attribs: project.max-shm-memory=(privileged,4294967296,deny)

For instance, here we have set the Maximum Shared memory to 4GB to the user account which owns and runs the Sybase ASE. And, the last line “attribs: project.max-shm-memory=(privileged,4294967296,deny)” basically, sets the maximum shared memory for the user processes to 4GB (4294967296).

This is equivalent to doing

# cat /etc/projects

If something similar doesn’t exist then the ASE is using the system defaults.

In any case, we need to increase the Maximum Shared memory. In our case increasing it to 7GB fixed the problem.

To add a new Project entry,

# projadd -p 101 -U asetest -c “Sybase ASE” -K “max-shm-memory=(privileged,7GB,deny)” user.asetest

This would create a new project with ID 101, name user.asetest for user asetest and set the Maximum Shared Memory to 7GB.

To modify an existing Project entry,

# projmod -sK “project.max-shm-memory=(privileged,7GB,deny)” user.asetest

This should help!

Oct  8 12:26:14 myserver sshd[29288]: [ID 800047 auth.error] error: Could not get shadow information for NOUSER

It would look like a security incident with the “nouser” system account but not so to be. The error gets written to /var/adm/messages when a user tries to SSH to the server using a user account which doesn’t exist on the system (a typo error can contribute) or on your NIS/NIS+ systems.

So no panic!!!!

Let’s look at the “which” utility in Solaris and find which package installed the file.

The file path for which is /usr/bin/which

# which which
/usr/bin/which

To find the package which installed it use the “pkgchk” command with the “-l” “and -p” options and the full file path of the file as follows,

# pkgchk -l -p /usr/bin/which
Pathname: /usr/bin/which
Type: regular file
Expected mode: 0555
Expected owner: root
Expected group: bin
Expected file size (bytes): 1288
Expected sum(1) of contents: 25204
Expected last modification: Jan 21 23:32:39 2005
Referenced by the following packages:
SUNWcsu

You can see that the which utility is installed by the package “SUNWcsu”.

There are many ways to do this. Let’s start with the easy one look at each of them

1. Use dos2unix utility

Solaris pre-installs dos2unix utility into the system to do this job for you. As the name says, it converts the file from DOS format to UNIX format. To do this the syntax is

# dos2unix <file> <new file>

for instance

# dos2unix test.txt test2.txt

Where the test.txt is the file you want the ^M characters removed and test2.txt is the file stripped of the ^M characters.

You may also overwrite the existing file by mentioning the source and destination files as same:

# dos2unix test.txt test.txt

2. Using VI

Open the file with “vi” editor and type the following:

:%s/^M//g

NOTE: To get the ^M in there, you should type CTRL+V+M

3. Using “tr” utility

“tr” utility is used to translate characters. Using with “-d” deletes a listed string.

# tr -d ‘\r’ <old.file > <new.file>

Alternatively, use its octel representation as follows:

# tr -d ‘\015′ <old.file > <new.file>

To simply display the number of Physical processors, simply run the command with the -p option as follows:

root@sunserver # psrinfo -p
2

where 2 implies that there are 2 physical processors installed on the system.

If you would like to check the number of Virtual Processors on each of these Physical processors then type the command with the “-pv” arguement as follows:

root@ server:/root$ uname -a

SunOS server 5.10 Generic_137111-02 sun4v sparc SUNW,SPARC-Enterprise-T5120

root@server:/root$ psrinfo -pv

The physical processor has 32 virtual processors (0-31)

  UltraSPARC-T2 (cpuid 0 clock 1165 MHz)

The above indicates that there was only one physical processor (UltraSPARC-T2) on the T5120 server which has 32 Virtual processors. Each virtual processor is an entity with its own interrupt ID, capable of executing independent threads.

In simple terms, the number of Virtual Processors supported by a physical CPU is

“Number of Core” x “Number of threads”

For instance, the above is on a T5120 server with the UltraSPARC-T2. This CPU  has 4 cores and each core can support 8 threads and that gives us 32 Virtual processors.

The number of Virtual processors on a Server is simply the total Virtual processors supported on each of the physical processor.

In the following T5140 server there are 2 Physical UltraSPARC-T2+ processors with 6 cores, each supporting 8 threads which means we get 48 Virtual processors per Physical processor and hence a total of 96 Virtual processors (sum of VPs on individual processors) for the server:

root@ bserver:/root$ uname -a

SunOS bserver 5.10 Generic_137111-02 sun4v sparc SUNW,T5140

root@ bserver:/root$ psrinfo -pv

The physical processor has 48 virtual processors (0-23 32-55)

  UltraSPARC-T2+ (cpuid 0 clock 1167 MHz)

The physical processor has 48 virtual processors (64-71 80-119)

  UltraSPARC-T2+ (cpuid 64 clock 1167 MHz)

In earlier versions of Solaris, the -p arguement is not supported and hence wouldn’t provide this summarised output on the counts of the physical and Virtual CPUs.

For instance, in the following crontab file there is a blank line between the last two cron jobs.

root@sunsolaris# crontab -l
# The root crontab should be used to perform accounting data collection.
#
# The rtc command is run to adjust the real time clock if and when
# daylight savings time changes.
#
10 1 * * 0,4 /etc/cron.d/logchecker
10 2 * * 0  /usr/lib/newsyslog
15 3 * * 0 /usr/lib/fs/nfs/nfsfind

30 4 * * * /usr/local/bin/disk_check,sh
;;;;
;;;
;;
;

To resolve the problem edit the crontab file and look for the blank line and delete the line. In the above, after editing the crontab, it should look lie the follows:

root@sunsolaris# crontab -l
# The root crontab should be used to perform accounting data collection.
#
# The rtc command is run to adjust the real time clock if and when
# daylight savings time changes.
#
10 1 * * 0,4 /etc/cron.d/logchecker
10 2 * * 0  /usr/lib/newsyslog
15 3 * * 0 /usr/lib/fs/nfs/nfsfind
30 4 * * * /usr/local/bin/disk_check,sh
;;;
;;
;

You can see the  blank line removed from the crontab file.

Following procedure should help you mount an ISO image in Sun Solaris

Attach a Block Device

sunsolaris# lofiadm -a /export/software/iso_image.iso
/dev/lofi/1

Mount the ISO Image block device

sunsolaris# mount -F hsfs -o ro /dev/lofi/1 /mnt

Where /mnt is the mount point.

This should mount the ISO image.

To confirm, change directory to /mnt and do a “ls” to lis the files

sunsolaris# cd /mnt

sunsolaris# ls

if at anytime, you want to look at these block devices simply type “lofiadm” command with no arguement.

sunsolaris# lofiadm
Block Device File
/dev/lofi/1 /export/software/iso_image.iso

When we are done with the files on the mounted ISO, we can unmount and detach the Block device we attached earlier:

sunsolaris# umount /mnt

sunsolaris# lofiadm -d /dev/lofi/1

For instructions on how to mount an ISO image in Linux, click here

There are also more convenient tools to make a Screenshot. You should use more reliable import (part of ImageMagick package) instead of xwd.

To Take Screenshot of a Frame

When logged onto the CDE, open a terminal session and run the command as follows:

sunsolaris# xwd -frame > screenshot.xwd

This will prompt you to select a window (the cursor becomes a +), then it will take a picture of the window (as it appears on the screen), the “-frame” options will have it also include the window managers decorative borders, then the file is converted to png.

Fullscreen Screenshot

A full screen screenshot can be done without any frame being set

sunsolaris# xwd -root > screenshot.xwd

Screenshot and Convert togethor

sunsolaris# xwd -frame | convert xwd:- screenshot.png

Screenshot after “n” seconds

Now, comes the interesting bit, you can take screenshots after a delay time (say 10 seconds). All you do is to use a sleep command as follows

sunsolaris# sleep 10; xwd -root -out screenshot.xwd

To print the image captured using Laserjet

xpr -device ljet /tmp/myimage | lp -d printer

where “printer” is the installed printer.

This file can now be converted to PNG, JPG,GIF using GIMP or any other image editing software.

The at utility reads commands from standard input and groups them together as an at job, to be executed at a later time.

List AT jobs

To list all the at jobs scheduled on the Sun Server

sunsolaris# at -l
user = root     1210161000.a    Wed May  7 12:50:00 2008
user = root     1210161001.a    Wed May  7 12:50:01 2008
user = root     1210198500.a    Wed May  7 23:15:00 2008
user = root     1210331460.a    Fri May  9 12:11:00 2008
user = root     1210198320.a    Wed May  7 23:12:00 2008

To run a command at a particular time

sunsolaris# at -t 200805072312
at> reboot
at> <EOT>
  commands will be executed using /sbin/sh
  job 1210198320.a at Wed May  7 23:12:00 2008

In the above -t switch defines the time that follows it as the time at which the one or more command that follows at the “at>” prompt to be run. Once, all the commands are entered press “CTRL+D” to make “at” aware of the end of commands.

Run a job now

sunsolaris# at now
at>reboot
at> <EOT>
commands will be executed using /sbin/sh
job 1210100190.a at Tue May  6 19:56:30 2008

Run at noon (12pm)

sunsolaris# at noon
at> uname -a
at> <EOT>
commands will be executed using /sbin/sh
job 1210158000.a at Wed May  7 12:00:00 2008

Run at midnight

sunsolaris# at midnight
at> uname -a
at> <EOT>
commands will be executed using /sbin/sh
job 1210114800.a at Wed May  7 00:00:00 2008

To run a job “n” mins from now

Say you want to run the command in 5 minutes from now

sunsolaris# at now + 5minutes
at> uname -a
at> <EOT>
commands will be executed using /sbin/sh
job 1210100631.a at Tue May  6 20:03:51 2008

To run a script or a file

sunsolaris# at -f /root/myscript now

The above runs the script /root/myscript immediatly. The switch “-f” defines the file to be run instead of commands.

To remove/cancel an at job

sunsolaris# at -r <jobname>

For instance, in the following list of jobs I want to cancel the last job “1210100631.a” job.

sunsolaris# at -l
user = root     1210161000.a    Wed May  7 12:50:00 2008
user = root     1210161001.a    Wed May  7 12:50:01 2008
user = root     1210198500.a    Wed May  7 23:15:00 2008
user = root     1210331460.a    Fri May  9 12:11:00 2008
user = root     1210198320.a    Wed May  7 23:12:00 2008
user = root     1210158000.a    Wed May  7 12:00:00 2008
user = root     1210114800.a    Wed May  7 00:00:00 2008
user = root     1210100631.a    Tue May  6 20:03:51 2008

sunsolaris# at -r 1210100631.a

sunsolaris# at -l
user = root     1210161000.a    Wed May  7 12:50:00 2008
user = root     1210161001.a    Wed May  7 12:50:01 2008
user = root     1210198500.a    Wed May  7 23:15:00 2008
user = root     1210331460.a    Fri May  9 12:11:00 2008
user = root     1210198320.a    Wed May  7 23:12:00 2008
user = root     1210158000.a    Wed May  7 12:00:00 2008
user = root     1210114800.a    Wed May  7 00:00:00 2008

Now, you can see the last job removed from the scheduled list.

at job using a desired shell

To use C shell (csh)

sunsolaris -c -f /etc/myfile now

-c specifies the shell to be used as bash.

To use bash (sh)

sunsolaris -s -f /etc/myfile now

-s specifies the shell to be used as bash.
To email user

To email user who scheduled the at job once the at command is run

sunsolaris -m -f /etc/myfile now

The -m specifies at job to email the user once the job is run.

The at jobs are saved as files under

/var/spool/cron/atjobs

To see whats in a job, do a “cat”

sunsolaris# cat 1210114800.a
: at job
: jobname: stdin
: notify by mail: no
: project: 1
export HOME; HOME=’/’
export HZ; HZ=”
export LANG; LANG=’C’
export LOGNAME; LOGNAME=’root’
export MAIL; MAIL=’/var/mail/root’
export PATH; PATH=’/usr/sbin:/usr/bin’
export SHELL; SHELL=’/sbin/sh’
export TERM; TERM=’xterm’
export TZ; TZ=’GB’
$SHELL << ‘…the rest of this file is shell input’
#ident  “@(#).proto     1.6     00/05/01 SMI”   /* SVr4.0 1.2   */
cd /etc/dhcp
umask 22
ulimit unlimited
uname -a

In instances where vold is not running or when there is a need to manually mount the CDROM the following the procedure should help.

Logon to the Sun Server and switch user to root

sunsolaris$ su -
Password:

Indentify Device Name for CDROM

The device name for the CDROM is normally /dev/dsk/c0t6d0s2. However, you can find this as follows:

sunsolaris# ls -al /dev/sr* |awk ‘{print “/” $11}’

This returns the device name for the CDROM. In my case, /dev/dsk/c0t6d0s2

Mount CDROM

Assuming the mount point “/cdrom” already exists, mount the CDROM as follows:

sunsolaris# mount -F hsfs -o ro /dev/dsk/c0t6d0s2 /cdrom

where /dev/dsk/c0t6d0s2 is the device name for the CDROM

/cdrom is the mount point

Now, cd to CDROM to see the files from the CDROM mounted.

Mounting NFS shared remote CDROM

In order to mount a CDROM froma  remote server sharing the CDROM, it has to be exported with root access.  Mount the CDROM from the remote server as follows:

sunsolaris# mount remoteserver:/cdrom/cdrom0 /cdrom

where /cdrom/cdrom0 is the sharename on the remote server.