UNIX and LINUX Administration

Enable EMS on HPUX

2013-08-29T12:03:00.003-07:00

EMS (Event Management System) is an amazing tool that comes along with the HPUX OS that can be used to monitor for any hardware failure events that happens in the server hardware like CPU, Memory, Fan, power supply, disk failures etc... among other things.

EMS has the capability to monitor for occurance of such events on the server and can also notify the team via email communication, loggin into the /var/adm/syslog/syslog.log among other mode of communication.

This tool is very helpful and can help us to keep track of any failures in the server hardware and we can take timely actions to reduce the effect of failure on the server functionality. monconfig is the command to configure EMS. the path of the file is /etc/opt/resmon/lbin/monconfig

read manual page of monconfig for more information.

different unix and linux equivalent commands

2013-02-01T11:39:00.001-08:00

Just imagine different unix linux flavours... say HPUX, Red Hat linux, AIX, solaris and what not... not all but among the above mentioned a few more flavours and the equivalent commands.

I have a link that shows a table with commands equivalent in different flavours.

can be helpful sometimes when we are not familiar with some flavours.

http://www.unixguide.net/unixguide.shtml

Enjoy!!!

MC SG Cluster and package creation

2013-01-11T06:52:00.002-08:00

Here is a really nice post on MC SG Cluster and package creation procedure.

http://kmit4u.blogspot.in/2010/06/cluster-create-mc-serviceguard-cluster.html

Force umount a vxfs on HPUX

2012-11-14T05:15:00.000-08:00

Have you come across a situation where you want to unmount a file system on a HPUX system and the file system is on VXFS?

Well, i have.

These situation does happens. I tried fuser, lsof. but nothing seems to be holding the File System which may disallow unmounting it. Only solution left is to reboot the server and go ahead with the task.

But what if the reboot of the server costs lots of business impact and financial losses. What if the downtiem window is not enough to have the server rebooted?

That is when we came across the command /sbin/vxumount This will unmount the file system forcefully without a reboot and we avoid a reboot.

usage : - /sbin/vxumount -o force /mount_point_name

Credit goes to my colleague mr Subodh Bagade for discovering this command!!!

Hope this comes to some ones help....

Resetting a GSP Console of HPUX from OS.

2012-09-30T15:34:00.002-07:00

There is this server whose GSP console credentials we do not have. So i was left with the option to get the server's GSP console resetted from onsite person since the box sites at remote location. Another option is i can do the same from the server's OS as well. here goes the command.

stty +resetGSP < /dev/GSPdiag1

See if the GSP is functional:
stty +queryGSP < /dev/GSPdiag1

Once we are done with this we can login to the server using the normal credentials Admin/Admin.

UNIX/LINUX User Add

2012-08-13T23:14:00.001-07:00

And i have a script that adds user on multiple UNIX/LINUX servers at a go.....

Sorry, if you want that?? Its really simple. Put in a comment with your email address and i will send you the script.

DIAGNOSTIC SYSTEM Warning -- HPUX

2012-08-06T03:11:00.002-07:00

Got a ticket for diagnostic system warning on a HPUX box.

below is the complete error log

Aug 3 16:33:47 sapsrvr vmunix: DIAGNOSTIC SYSTEM WARNING:

Aug 3 16:33:47 sapsrvr vmunix: The diagnostic logging facility has started receiving excessive

Aug 3 16:33:47 sapsrvr vmunix: errors from the I/O subsystem. I/O error entries will be lost

Aug 3 16:33:47 sapsrvr vmunix: until the cause of the excessive I/O logging is corrected.

Aug 3 16:33:47 sapsrvr vmunix: If the diaglogd daemon is not active, use the Daemon Startup command

Aug 3 16:33:47 sapsrvr vmunix: in stm to start it.

Aug 3 16:33:47 sapsrvr vmunix: If the diaglogd daemon is active, use the logtool utility in stm

Aug 3 16:33:47 sapsrvr vmunix: to determine which I/O subsystem is logging excessive errors.

started digging in and found that the diaglogd deamon is not running on the server.

But above error message doesnt suggest any thing that says the log is cause of diaglogd deamon not running.

so first step was to start the diaglogd and diagmond deamons.
/usr/sbin/stm/uut/bin/sys is the location where both the deamons executables are located and the diagmond deamon was renamed to diagmond.old which was restricting the start of diaglogd deamon.

so i renamed it back to original name and could be able to start the diagmond and diaglogd deamons and could be able to map devices from cstm.