and this is what I experienced after an electrical failure.
this time, /dev/sda1 was the victim.
the system restarted fine; it e2fsck'd all filesystems and were clean; then
quotacheck started. at that time it started with messages like mentioned
before. this time, the logs are usable, sorry for the long lines here.
Jun 18 11:32:08 grobbebol kernel: VFS: Mounted root (ext2 filesystem) readonly.
Jun 18 11:32:08 grobbebol kernel: Freeing unused kernel memory: 72k freed
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #102035: rec_len is smaller than minimal - offset=0, inode=0, rec_len=0, name_len=0
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #102033: rec_len is smaller than minimal - offset=0, inode=0, rec_len=0, name_len=0
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #69458: rec_len % 4 != 0 - offset=0, inode=1869364781, rec_len=27491, name_len=41
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #75516: rec_len % 4 != 0 - offset=0, inode=1647255843, rec_len=28265, name_len=47
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #75513: rec_len is smaller than minimal - offset=0, inode=0, rec_len=0, name_len=0
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #87755: rec_len is smaller than minimal - offset=0, inode=0, rec_len=0, name_len=0
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #87753: rec_len is smaller than minimal - offset=0, inode=0, rec_len=0, name_len=0
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #73498: rec_len % 4 != 0 - offset=0, inode=1684960547, rec_len=26213, name_len=32
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #67791: rec_len % 4 != 0 - offset=0, inode=561277216, rec_len=11305, name_len=10
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #85713: rec_len is smaller than minimal - offset=0, inode=0, rec_len=0, name_len=0
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #67805: rec_len % 4 != 0 - offset=0, inode=2337412747, rec_len=21015, name_len=232
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #73492: rec_len % 4 != 0 - offset=0, inode=1815044978, rec_len=28265, name_len=117
Jun 18 11:32:08 grobbebol kernel: EXT2-fs error (device sd(8,1)): ext2_readdir: bad entry in directory #116313: rec_len is smaller than minimal - offset=0, inode=0, rec_len=0, name_len=0
Jun 18 11:32:08 grobbebol kernel: EXT2-fs warning (device sd(8,1)): ext2_free_blocks: bit already cleared for block 446518
Jun 18 11:32:08 grobbebol last message repeated 11 times
Jun 18 11:32:08 grobbebol kernel: EXT2-fs warning (device sd(8,1)): ext2_free_blocks: bit already cleared for block 145414
Jun 18 11:32:08 grobbebol kernel: EXT2-fs warning (device sd(8,1)): ext2_free_blocks: bit already cleared for block 448671
Jun 18 11:32:08 grobbebol kernel: EXT2-fs warning (device sd(8,1)): ext2_free_blocks: bit already cleared for block 448671
Jun 18 11:32:08 grobbebol kernel: EXT2-fs warning (device sd(8,1)): ext2_free_blocks: bit already cleared for block 136433
Jun 18 11:32:08 grobbebol kernel: EXT2-fs warning (device sd(8,1)): ext2_free_blocks: bit already cleared for block 350077
Jun 18 11:33:50 grobbebol kernel: EXT2-fs warning (device sd(8,1)): ext2_free_inode: bit already cleared for inode 65306
ok, I didn't panic (as the disk is removable and I have 2 backups) so I
restarted the system. now, the system wewnt into single user emergency mode
and I started e2fsck manually.
it spewed several duplicate/bad blocks.... asking if I wanted to clone it.
then stuff like this (several messages each, shown one only)
entry 'kernel' in ??? (65381) is a link to dir ??? (34302)
bad inode # for '.' in dir inode 69735
unconnected dir inode 55223 (???) .. in /lost+found/#120549/config/Motif.tmpl
is /usr/X/server (120800), should be /lost+found/#120549/config (33108)
inode ref count is 5, should be 6
23, should be 20
54, should be 61
and also several instances of 'WARNING -- E2FSCK program bug (or bonehead
(me) tries to check live FS' (the filesystem was NOT mounted)
inode.link_info[32705]=0, inode.i_links_count=65532. should be the same
/dev/sda2 was also affected but the backup blocks helped here.
I know that some messages might be slightly off the real world due to the
fact that I had all other filesystems needed mounted r/o.
I also experienced one time, during the check an uncommanded reboot.
Like Harald, I'll revert to 2.2.5 or 2.2.7 and see if that keeps the
stability OK.
I hope it helps.
Roeland (now, lets reconstruct the data)
-- Grobbebol's Home | Don't give in to spammers. http://www.xs4all.nl/~bengel | Use your real e-mail address Linux 2.0.34 on an i586/64 MB | on Usenet.- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/