C
chipitsine
Guest
I seen similar messages regarding bug in kernel 2.6.18, but that is definetly not our case. we are running pvetest (2.6.32-5).
/var/log./messages contains regular errors about EXT3 things
Aug 1 00:42:57 KT0015 kernel: kjournald starting. Commit interval 5 seconds
Aug 1 00:42:57 KT0015 kernel: EXT3 FS on dm-3, internal journal
Aug 1 00:42:57 KT0015 kernel: EXT3-fs: dm-3: 17 orphan inodes deleted
Aug 1 00:42:57 KT0015 kernel: EXT3-fs: recovery complete.
Aug 1 00:42:57 KT0015 kernel: EXT3-fs: mounted filesystem with ordered data mode.
Aug 1 01:04:03 KT0015 kernel: lost page write due to I/O error on dm-3
usually it is ok, but often system becomes unresponsive during vzdump (virtual machines do not respond, host refuses to reboot using "reboot" command").
it looks like old known issue, but it is not.
according to that forum, people are using ProxMox in production environment and it is not common issue to stop at nigthly vzdump ?
what could I miss ?
P.S. it is not about our RAID hardware. we are running different RAID arrays, problem with EXT3 occures from time to time on any of them.
/var/log./messages contains regular errors about EXT3 things
Aug 1 00:42:57 KT0015 kernel: kjournald starting. Commit interval 5 seconds
Aug 1 00:42:57 KT0015 kernel: EXT3 FS on dm-3, internal journal
Aug 1 00:42:57 KT0015 kernel: EXT3-fs: dm-3: 17 orphan inodes deleted
Aug 1 00:42:57 KT0015 kernel: EXT3-fs: recovery complete.
Aug 1 00:42:57 KT0015 kernel: EXT3-fs: mounted filesystem with ordered data mode.
Aug 1 01:04:03 KT0015 kernel: lost page write due to I/O error on dm-3
usually it is ok, but often system becomes unresponsive during vzdump (virtual machines do not respond, host refuses to reboot using "reboot" command").
it looks like old known issue, but it is not.
according to that forum, people are using ProxMox in production environment and it is not common issue to stop at nigthly vzdump ?
what could I miss ?
P.S. it is not about our RAID hardware. we are running different RAID arrays, problem with EXT3 occures from time to time on any of them.
Last edited by a moderator: