hi gurus,

i hv one HP DL380 and 8units of HP DL360 all running with redhat 7.3.
basically, i hv clustered all machines using LSF 5.1 (website platform.com).
the cluster processing works fine. no problem.
the problem is, my machine got hung-up after 4-5 days.. what could possibly be wrong?

FYI, i did not install kernel patch or any patch at all as my software & LSF 5.1 do not require the patch to be installed.

any ideas? im sorry im new to redhat linux. this is my first assignment.

thx.

Hi,

I used to have a similar problem on my rh7.3 boxes, and yes they were also HP boxes .... dont remember the model number though .....

look into the output of /var/log/messages immediately after a hang and a hard reboot..... u might get some indication of a error in the code..... dont remember the exact error.

Solution: Recompile the kernel to something more recent. In my case a recompile to 2.4.21 and 2.4.23 solved the problem and this box gave an uptime of > 365 days . had to be brought down as we wanted more RAM on the box....

-Subhasis

Note: LSF will still work after the recompile...... just to be sure, check the platform website for the latest kernel supported.