[lustre-discuss] old Lustre 2.8.0 panic'ing continously
harenberg at physik.uni-wuppertal.de
Tue Mar 10 00:18:23 PDT 2020
thanks for all of your replies.
Am 09.03.20 um 13:32 schrieb Andreas Dilger:
> It would be better to run a full e2fsck, since that not only rebuilds
> the quota tables, but also ensures that the values going into the quota
> tables are correct. Since the time taken by "tune2fs -O quota" is
> almost the same as running e2fsck, it is better to do it the right way.
We already ran e2fsck -f on all LUNs after every crash, so it seems that
was all we could do, right?
During the last days (since thursday), our Lustre instance was
surprisingly stable. We lowered a bit the load by limiting the # of
running jobs which might also helped to stablize the system.
We enabled kdump, so if another crash is happening anytime soon, we hope
to get at least a dump for a hint where the problem is.
Dr. Torsten Harenberg harenberg at physik.uni-wuppertal.de
Fakultät 4 - Physik Tel.: +49 (0)202 439-3521
Gaussstr. 20 Fax : +49 (0)202 439-2811
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 5341 bytes
Desc: S/MIME Cryptographic Signature
More information about the lustre-discuss