[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [Xen-users] Filesystem corruptions
Dear xen-users,in the last two days, two different domUs on two different dom0s had suddenly corrupted filesystems which let to total data loss on both domUs. I have never suffered under this failure before and I guess it is related to XEN. Both domUs were running for long time without any problems and I am pretty sure, that the hard drives are fine (checked with smart checks). [1215379.302679] EXT3-fs (xvda2): error in ext3_new_inode: IO failure [1215379.302689] Aborting journal on device xvda2.[1215379.304193] EXT3-fs (xvda2): error: ext3_journal_start_sb: Detected aborted journal [1215379.304200] EXT3-fs (xvda2): error: remounting filesystem read-only [1215379.329043] EXT3-fs (xvda2): error: remounting filesystem read-only [1215379.329246] EXT3-fs (xvda2): error in ext3_create: IO failure [1215379.330502] __journal_remove_journal_head: freeing b_committed_dataAfter shuting down of domUs, fsck failed to repair the filesystem and restore of files was necessary. No updates of XEN or the kernel were made the days before. It was no possible to reproduce the problem yet. Setup for both physical hosts is exactly the same: dom0 ----- Debian GNU/Linux 7.5 wheezy sw-Raid via mdadm, LVM on top ââ$ uname -a Linux hostname 3.2.0-4-amd64 #1 SMP Debian 3.2.57-3+deb7u2 x86_64 GNU/Linux release : 3.2.0-4-amd64 version : #1 SMP Debian 3.2.54-2 machine : x86_64 xen_major : 4 xen_minor : 1 xen_extra : .4Each running about 10 domains, and only one was affected each by this problem. domU -------- Debian GNU/Linux 7.5 wheezy Boot via pygrub Linux hostname 3.2.0-4-amd64 #1 SMP Debian 3.2.57-3+deb7u2 x86_64 GNU/Linux Has somebody an idea what is going wrong here? Best regards, Patric -- web www.geops.de rss www.geops.de/blog/feed follow www.twitter.com/geops _______________________________________________ Xen-users mailing list Xen-users@xxxxxxxxxxxxx http://lists.xen.org/xen-users
|
Lists.xenproject.org is hosted with RackSpace, monitoring our |