Server crash

[jeena_2]

Hi,

After crash the server /var/adm/crash given below mentioned error....

error savecrash: Insufficient space to save full core dump

Could you please help on this...

Thanks in Adv.

[Dennis Handly]

It means what it says, you have too much memory for the size of the filesystem under /var/adm/crash/.

[jeena_2]

Hi,
If /var file system is full server will automatically restart or not...? because before this outage /var was 100%...how can i find out the RC.

Thanks in Adv.

[Bill Hassell]

First, save the contents of /var/adm/crash onto a tape and then remove all the files under the crash directory. That will fix the 100% problem in /var.

If you need to find the reason for the crash, you can contact HP support and send them the crash files.

[Kevin Lister]

Jeena,

Dennis is right. You need to have enough space to save the crashdump.

So, try this:

1) Boot the system in single user mode.
2) Mount the /var filesystem.
3) Clean it out. (Remove junk in /var/adm/crash)
4) If you have available space in your Volume Group then create a new Logical Volume with a new VxFS filesystem with a mount point of /var/adm/crash. Make sure this filesystem is bigger than the total amount of system memory installed on the system.
5) Bring the system up in multi-user mode and if it crashes the dump will be saved in the new filesystem that you created for it.
6) Then you can get HP to analyze the dump.

kev

[Hakki Aydin Ucar]

You have assigned points to 0 of 17 responses to my questions.

http://forums11.itrc.hp.com/service/forums/helptips.do?#28