[Gimle-users] Login node rebooted due to a kernel panic

Peter Kjellstrom cap at nsc.liu.se
Fri Nov 7 15:39:04 CET 2008


Short version: A panic forced us to reboot the login-node on Gimle no jobs 
running or queued should have been affected.

Longer version:
Around 13.30 today the login-node hung due to a kernel panic. Since we had a 
similar hang on the system-node last week and on the longin-node this monday 
we were already preparing a new software image to be deployed during the 
comming week.

However, when the login-node died again today we took the opportunity to 
upgrade it directly. The rest of the cluster will be upgraded as planned next 
week with minimal planned down-time. The upgrade includes, among other 
things, a minor version bump for the kernel and the lustre file system 
software.

Have a nice week-end and please do report any oddities to us,
 Peter K

-- 
------------------------------------------------------------
  Peter Kjellström               | E-mail: cap at nsc.liu.se
  National Supercomputer Centre  |
  Sweden                         | http://www.nsc.liu.se
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part.
Url : http://www.nsc.liu.se/pipermail/gimle-users/attachments/20081107/f3842ac7/attachment.bin


More information about the Gimle-users mailing list