[Dunder-users] status update regarding /nobackup/rossby9 and /nobackup/smhid4

Peter Kjellstrom cap at nsc.liu.se
Wed May 30 17:35:32 CEST 2007


Tornado and Dunder users,

Quick summary for the impatient:
* smhid4, very likeyly perfectly ok, back soon (kept off-line to ease 
diagnostics on rossby9)
* rossby9, probably ok except for a few possibly lost files, could take a week 
before it's back


/nobackup/rossby9 and /nobackup/smhid4 are still unavailable after the power 
outage last Sunday. When we brought the system back up internal sanity checks 
flagged rossby9 as damaged. Both dunder and tornado were kept off-line to 
stop jobs expecting access to the filesystem from failing (and ease 
diagnostics). Later the clusters were restored to normal status without 
rossby9 and smhid4.

We have made good progress in bringing up the filesystems and it now seems 
like both filesystems are in a good shape. However, there is a small number 
of files on /nobackup/rossby9 that we unfortunately suspect will be lost.

We are now in the process of assessing the consistency of the file systems 
before resuming normal operations. This can take several days 
since /nobackup/rossby9 is very large. If this takes a long time it's likely 
that we will provide access to /nobackup/smhid4 earlier.

Stay tuned for more updates.  We appreciate you patience in this matter.

Regards / NSC support team
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://www.nsc.liu.se/pipermail/dunder-users/attachments/20070530/9de7750d/attachment.bin


More information about the dunder-users mailing list