Koozali.org: home of the SME Server
Contribs.org Forums => General Discussion => Topic started by: holck on November 27, 2007, 11:09:41 AM
-
Sometimes during my daily backups with dar2 to a USB-attached disk, a great number of errors/messages appear in /var/log/messages. An example:
Nov 27 03:00:16 anton esmith::event[2000]: S50rewind-tape=action|Event|pre-backup|Action|S50rewind-tape|Start|1196128815 609796|End|1196128816 550314|Elapsed|0.940518
Nov 27 04:17:24 antonio kernel: [<d88a3c72>] ext3_write_inode+0x22/0x3f [ext3]
Nov 27 04:17:24 antonio kernel: [write_inode+48/55] write_inode+0x30/0x37
Nov 27 04:17:24 antonio kernel: [<c0178026>] write_inode+0x30/0x37
Nov 27 04:17:24 antonio kernel: [__sync_single_inode+112/443] __sync_single_inode+0x70/0x1bb
Nov 27 04:17:24 antonio kernel: [<c017809d>] __sync_single_inode+0x70/0x1bb
Nov 27 04:17:24 antonio kernel: [sync_sb_inodes+423/628] sync_sb_inodes+0x1a7/0x274
Nov 27 04:17:24 antonio kernel: [<c0178449>] sync_sb_inodes+0x1a7/0x274
Nov 27 04:17:24 antonio kernel: [writeback_inodes+145/222] writeback_inodes+0x91/0xde
Nov 27 04:17:24 antonio kernel: [<c01785a7>] writeback_inodes+0x91/0xde
Nov 27 04:17:24 antonio kernel: [balance_dirty_pages+124/284] balance_dirty_pages+0x7c/0x11c
Nov 27 04:17:24 antonio kernel: [<c01452a7>] balance_dirty_pages+0x7c/0x11c
Nov 27 04:17:24 antonio kernel: [<d88a1f7d>] ext3_ordered_commit_write+0xb6/0xc5 [ext3]
Many more lines...
Nov 27 04:18:13 antonio kernel: [syscall_call+7/11] syscall_call+0x7/0xb
Nov 27 04:18:13 antonio kernel: [<c02d66a7>] syscall_call+0x7/0xb
Nov 27 04:18:13 antonio kernel: [unix_dgram_sendmsg+112/1117] unix_dgram_sendmsg+0x70/0x45d
Nov 27 04:18:13 antonio kernel: ] unix_dgram_sendmsg+0x70/0x45d
Nov 27 05:09:52 antonio esmith::event[11527]: Processing event: post-backup
As far as I can see, I receive no other error messages, and dar2 seems to finish OK. Can any of you help with a clue as to what is going on and whether I should worry?
Thanks,
Jesper
-
I get these too, and then the server crashes after a couple of hours of these messages. Had to switch backups off. I'm hoping to change to a faster server, and then turn backups back on.
I believe it has something to do with race-conditions in some kernel versions. Two questions: are you running on older hardware, or something very new? Secondly, what filesystem are you using on the external hard drive - FAT32, ext3, something else?
-- JJ
-
Thanks for your reply. Yes, I run on older hardware, a nice quiet PC that could no longer be used at my workplace. The USB-attached disk is ext3-formatted. So what does that tell you?
Thanks,
Jesper