system checklist

[Dan Copeland_1]

Hi,

I need to give my operations folk a checklist of things to do after a machine comes back up. I could do this myself but I wanted to check here first to see if anyone can point me to this type of documentation or has it themselves. I know that certain things will be specific to individual machines, but anything would be appreciated.

Tia,

Dan

[Bill McNAMARA_1]

Other than looking at /etc/rc.log and shutdownlog and syslog,
check that
mount -p matches /etc/fstab
strings /etc/lvmtab matches vgdisplay for the vgs
ioscan -fnk shows same as backed up ioscan.good
(remember state of NO_HW vanish after reboot)
top (make sure system load in reasonable range)

Later,
Bill

[A. Clay Stephenson · Accepted solution]

Hi Dan,

My answer to this is look at your ITO message browser - no news is good news. That is probably not want you want to hear so as a basic list:

0) Examine /etc/rc.log - Ant strange stuff?
1) bdf - Do you see what you expect? Are NFS mountpoints mounted?
2) ping a couple of nodes by hostname
3) lpstat -t - Is the scheduler up and running?
4) Are the application processes running (e.g. Oracle, ERP, CAD, CAM, ...)

Having said this, my best answer is still look at ITO. If you've written the templates and alarms to cover your applications, you don't have to ask 'are we there yet?'.

Regards, Clay

This is not meant to be an exhaustive list but
in the vast majority of cases if your box passes these tests, it's up and running.

NOTE: This does not mean that you have no data integrity problems caused by the crash.

[Victor_5]

Gather the following configuration and keep it:
1. uname -a
2. bdf
3. swapinfo
4. ioscan -fn
5. vgdisplay -v
6. lanscan
7. netstat -r
8. /etc/hosts
9. /etc/checklist
10. /etc/passwd
11. /etc/group
12. /etc/profile
13. ~/.profile
14. /etc/inittab
15. /etc/inetd.conf
16. /etc/services
17. lpstat -t