Skip to contentSkip to conversation

ext3 on raid5 failure

DISCUSSION Jan Dittmer 9 participants 23 messages Jan 18, 2004 – Feb 19, 2004

lore
Share & tools
lore
#0 lore raw

Hello,

I've recently upgraded my fileserver to 2.6.1 and since then I had to reboot two times because of the following errors. Afterwards the blockdevice was set readonly, so I couldn't even remount the partition rw. Interestingly after reboot the raid is not marked dirty. So is this an ext3 only error? This never happenend with 2.4.23pre6-aa3 or any other 2.4 version which run for about a year on this system without any (linux caused) downtime.
What I can try to debug this? The first time it happened after about 3 or 4 days uptime, the second time after about 1 day. So it's pretty reproducible.

Thanks,

Jan

EXT3-fs error (device dm-1): ext3_readdir: bad entry in directory #9783034: rec_len % 4 != 0 - offset=0, inode=1846971784, rec_len=33046, name_len=154 Aborting journal on device dm-1.
ext3_abort called.
EXT3-fs abort (device dm-1): ext3_journal_start: Detected aborted journal
Remounting filesystem read-only

and

EXT3-fs error (device dm-1): ext3_readdir: bad entry in directory #8422045: rec_len % 4 != 0 - offset=0, inode=655393188, rec_len=1998, name_len=0 Aborting journal on device dm-1.
ext3_abort called.
EXT3-fs abort (device dm-1): ext3_journal_start: Detected aborted journal
Remounting filesystem read-only
EXT3-fs error (device dm-1) in start_transaction: Journal has aborted

$ cat /proc/mdstat
Personalities : [raid5]
read_ahead 1024 sectors
md0 : active raid5 ide/host2/bus1/target1/lun0/part1[3]
ide/host2/bus1/target0/lun0/part1[1]
ide/host2/bus0/target1/lun0/part1[2]
ide/host2/bus0/target0/lun0/part1[0]
ide/host0/bus1/target0/lun0/part1[4]
      468872704 blocks level 5, 4k chunk, algorithm 2 [5/5] [UUUUU]
$ vgdisplay:
  --- Volume group ---
  VG Name               myraid
  System ID
  Format                lvm2
  Metadata Areas        1
  Metadata Sequence No  7
  VG Access             read/write
  VG Status             resizable
  MAX LV                256
  Cur LV                2
  Open LV               2
  Max PV                256
  Cur PV                1
  Act PV                1
  VG Size               447.15 GB
  PE Size               4.00 MB
  Total PE              114470
  Alloc PE / Size       114470 / 447.15 GB
  Free  PE / Size       0 / 0
  VG UUID               0xPsvH-t204-YlD4-jVAW-mYgK-ndF5-nEVOTk
$ lvdisplay:
  --- Logical volume ---
  LV Name                /dev/myraid/lvol0
  VG Name                myraid
  LV UUID                X47Raw-ZGgE-PZ4I-F1o3-lYX9-TVQ1-KIWqrs
  LV Write Access        read/write
  LV Status              available
  # open                 1
  LV Size                100.00 GB
  Current LE             25600
  Segments               1
  Allocation             next free (default)
  Read ahead sectors     0
  Block device           254:1

--- Logical volume ---

  LV Name                /dev/myraid/lvol1
  VG Name                myraid
  LV UUID                LgnX1d-r3BP-3u7t-yLQb-iRke-544J-v7uBBd
  LV Write Access        read/write
  LV Status              available
  # open                 1
  LV Size                347.15 GB
  Current LE             88870
  Segments               1
  Allocation             next free (default)
  Read ahead sectors     0
  Block device           254:2
$ mount -l
/dev/mapper/myraid-lvol1 on /mnt/data/1 type ext3 (rw)
/dev/mapper/myraid-lvol0 on /mnt/backup type ext3 (rw)
#1 lore raw
Quoted 7 lines · Jan Dittmer

EXT3-fs error (device dm-1): ext3_readdir: bad entry in directory #9783034: rec_len % 4 != 0 - offset=0, inode=1846971784, rec_len=33046, name_len=154
Aborting journal on device dm-1.
ext3_abort called.
EXT3-fs abort (device dm-1): ext3_journal_start: Detected aborted journal Remounting filesystem read-only

Run fsck on the filesystem.

#2 lore raw
Quoted 10 lines · Mike Fedyk
Quoted 7 lines · Jan Dittmer

EXT3-fs error (device dm-1): ext3_readdir: bad entry in directory #9783034: rec_len % 4 != 0 - offset=0, inode=1846971784, rec_len=33046, name_len=154
Aborting journal on device dm-1.
ext3_abort called.
EXT3-fs abort (device dm-1): ext3_journal_start: Detected aborted journal Remounting filesystem read-only

Run fsck on the filesystem.

Jan, what distribution are you running? The superblock *should* have been marked "filesystems has errors", and so fsck should have been forced when you rebooted. Did fsck in fact run, and if so, did it detect and fix any problems?

- Ted

#3 lore raw
Quoted 13 lines · Theodore Ts'o
Quoted 9 lines · Mike Fedyk
Quoted 5 lines · Jan Dittmer

EXT3-fs error (device dm-1): ext3_readdir: bad entry in directory #9783034: rec_len % 4 != 0 - offset=0, inode=1846971784, rec_len=33046, name_len=154 Aborting journal on device dm-1.
ext3_abort called.
EXT3-fs abort (device dm-1): ext3_journal_start: Detected aborted journal
Remounting filesystem read-only

Run fsck on the filesystem.

Quoted 6 lines

Jan, what distribution are you running? The superblock *should* have been marked "filesystems has errors", and so fsck should have been forced when you rebooted. Did fsck in fact run, and if so, did it detect and fix any problems?

- Ted

Okay, I fscked all filesystems in single user mode, thereby fscked up my root filesystem, though I didn't even check it - so I restored it from backup (grub wouldn't even load anymore). After 2 days in my freshly setup debian (2.6.1-bk6), same error. But this time at least I know it's because I tried to delete those files in the lost+found directory... So, how do I delete these and why did fsck fail? It's pretty annoying to reboot because of this. It would be nice to just being able to remount,rw the partition.

Thanks,

Jan

EXT3-fs error (device dm-2): ext3_readdir: directory #16370 contains a hole at offset 8192 Aborting journal on device dm-2.
EXT3-fs error (device dm-2): ext3_readdir: directory #16370 contains a hole at offset 24576 ext3_abort called.
EXT3-fs abort (device dm-2): ext3_journal_start: Detected aborted journal Remounting filesystem read-only

#4 lore raw
Quoted 4 lines · Jan Dittmer

Okay, I fscked all filesystems in single user mode, thereby fscked up my root filesystem, though I didn't even check it - so I restored it from backup (grub wouldn't even load anymore).

What messages were printed by e2fsck while it was running --- and was all of the filesystems unmounted, excepted for the root filesystem, which should have been mounted read-only?

Quoted 3 lines

After 2 days in my freshly setup debian (2.6.1-bk6), same error. But this time at least I know it's because I tried to delete those files in the lost+found directory...

How did you come to that conclusion?

Quoted 3 lines

So, how do I delete these and why did fsck fail? It's pretty annoying to reboot because of this. It would be nice to just being able to remount,rw the partition.

How did fsck fail? Without a transcript of the fsck output, it's hard to say exactly what happened here.

An output of dumpe2fs of the filesystem in question would also be useful.

- Ted

#5 lore raw
Quoted 9 lines · Theodore Ts'o
Quoted 1 line · Jan Dittmer

Okay, I fscked all filesystems in single user mode, thereby fscked up my root filesystem, though I didn't even check it - so I restored it from backup (grub wouldn't even load anymore).

What messages were printed by e2fsck while it was running --- and was all of the filesystems unmounted, excepted for the root filesystem, which should have been mounted read-only?

Yep it was all unmounted. But I don't have a transcript of this session. I was directly logged in, sorry.

Quoted 5 lines
Quoted 2 lines

After 2 days in my freshly setup debian (2.6.1-bk6), same error. But this time at least I know it's because I tried to delete those files in the lost+found directory...

How did you come to that conclusion?

sfhq:/mnt/data/1/lost+found# ls -l
total 76
d-wSr-----    2 1212680233 136929556    49152 Jun  7  2008 #16370
-rwx-wx---    1 1628702729 135220664    45056 May  4  1974 #16380
sfhq:/mnt/data/1/lost+found# rm *
rm: cannot remove `#16370': Operation not permitted
rm: cannot remove `#16380': Operation not permitted
sfhq:/mnt/data/1/lost+found# rm *
rm: cannot remove `#16370': Operation not permitted
rm: cannot remove `#16380': Operation not permitted
sfhq:/mnt/data/1/lost+found# rm *
rm: cannot remove `#16370': Operation not permitted
rm: cannot remove `#16380': Operation not permitted
sfhq:/mnt/data/1/lost+found# dmesg
EXT3-fs error (device dm-2): ext3_readdir: directory #16370 contains a hole at offset 8192
Aborting journal on device dm-2.
EXT3-fs error (device dm-2): ext3_readdir: directory #16370 contains a hole at offset 24576
ext3_abort called.
EXT3-fs abort (device dm-2): ext3_journal_start: Detected aborted journal
Remounting filesystem read-only

This is with 2.6.1-bk6 and no concurrent access. With 2.4.25-pre7-pac1 this does not happen. So I rechecked the filesystem (with 2.4), hoping it won't get corrupted
and I would be able to delete the files, but:

sfhq:~# fsck /dev/myraid/lvol1
fsck 1.35-WIP (07-Dec-2003)
e2fsck 1.35-WIP (07-Dec-2003)
/dev/myraid/lvol1 contains a file system with errors, check forced.
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Directory inode 16370 has an unallocated block #2.  Allocate<y>? yes

Directory inode 16370 has an unallocated block #6. Allocate<y>? yes

Pass 3: Checking directory connectivity
Pass 3A: Optimizing directories
Pass 4: Checking reference counts
Pass 5: Checking group summary information

/dev/myraid/lvol1: ***** FILE SYSTEM WAS MODIFIED ***** /dev/myraid/lvol1: 2341/45514752 files (24.0% non-contiguous), 82194611/91002880 blocks

still no luck with undeleting these files

sfhq:/mnt/data/1/lost+found# rm -rf *
rm: cannot remove directory `#16370': Operation not permitted
rm: cannot remove `#16380': Operation not permitted
sfhq:/mnt/data# dumpe2fs /dev/myraid/lvol1
dumpe2fs 1.35-WIP (07-Dec-2003)
Filesystem volume name:   <none>
Last mounted on:          <not available>
Filesystem UUID:          ba3f83cd-04c8-4c4d-82f7-990077aabb73
Filesystem magic number:  0xEF53
Filesystem revision #:    1 (dynamic)
Filesystem features:      has_journal dir_index needs_recovery sparse_super
Default mount options:    (none)
Filesystem state:         clean
Errors behavior:          Continue
Filesystem OS type:       Linux
Inode count:              45514752
Block count:              91002880
Reserved block count:     3640115
Free blocks:              8808269
Free inodes:              45512411
First block:              0
Block size:               4096
Fragment size:            4096
Blocks per group:         32768
Fragments per group:      32768
Inodes per group:         16384
Inode blocks per group:   512
Filesystem created:       Thu Jul  3 13:57:02 2003
Last mount time:          Wed Jan 28 11:48:13 2004
Last write time:          Wed Jan 28 11:48:13 2004
Mount count:              2
Maximum mount count:      23
Last checked:             Wed Jan 28 11:47:06 2004
Check interval:           15552000 (6 months)
Next check after:         Mon Jul 26 12:47:06 2004
Reserved blocks uid:      0 (user root)
Reserved blocks gid:      0 (group root)
First inode:              11
Inode size:               128
Journal inode:            8
Default directory hash:   tea
Directory Hash Seed:      a7fd0c53-efe9-4316-9fa6-e3206623f4fe
Journal backup:           inode blocks

Do you need the complete output? Btw. I have the same error on another partition (also on raid5, dm).

Thanks,

Jan

#18 lore raw
In reply to Theodore Ts'o
Quoted 10 lines · Theodore Ts'o
Quoted 1 line · Jan Dittmer

Okay, I fscked all filesystems in single user mode, thereby fscked up my root filesystem, though I didn't even check it - so I restored it from backup (grub wouldn't even load anymore).

What messages were printed by e2fsck while it was running --- and was all of the filesystems unmounted, excepted for the root filesystem, which should have been mounted read-only?

Quoted 4 lines
Quoted 1 line

After 2 days in my freshly setup debian (2.6.1-bk6), same error. But this time at least I know it's because I tried to delete those files in the lost+found directory...

How did you come to that conclusion?

Argh, sorry, after chattr -aI the files could be deleted. sorry for the noise.

Jan

#6 lore raw
In reply to Jan Dittmer
Quoted 11 lines · Jan Dittmer
Quoted 6 lines
Quoted 3 lines

After 2 days in my freshly setup debian (2.6.1-bk6), same error. But this time at least I know it's because I tried to delete those files in the lost+found directory...

How did you come to that conclusion?

sfhq:/mnt/data/1/lost+found# ls -l
total 76
d-wSr-----    2 1212680233 136929556    49152 Jun  7  2008 #16370
-rwx-wx---    1 1628702729 135220664    45056 May  4  1974 #16380

Ok, this looks like random garbage has gotten written into inode table.

If you can make this happen consistently with 2.6 and not with 2.4, then that would be useful to know. There may be some kind of race condition or problem with either the raid5 code, or the combination of raid5 plus ext3. It's unlikely this kind of error would be caused by a flaw in the ext3 code alone, since this is indicative of complete garbage written to the inode table, or a block intended for another location on disk getting written to the inode table. The natural suspect is at the block device layer and below.

- Ted

#7 lore raw
Quoted 10 lines · Theodore Ts'o
Quoted 6 lines

sfhq:/mnt/data/1/lost+found# ls -l
total 76
d-wSr-----    2 1212680233 136929556    49152 Jun  7  2008 #16370
-rwx-wx---    1 1628702729 135220664    45056 May  4  1974 #16380

Ok, this looks like random garbage has gotten written into inode table.

Quoted 9 lines

If you can make this happen consistently with 2.6 and not with 2.4, then that would be useful to know. There may be some kind of race condition or problem with either the raid5 code, or the combination of raid5 plus ext3. It's unlikely this kind of error would be caused by a flaw in the ext3 code alone, since this is indicative of complete garbage written to the inode table, or a block intended for another location on disk getting written to the inode table. The natural suspect is at the block device layer and below.

I've seen this kind of problems on my notebook too. Among others, over 600MB of a huge cache directory (from a news reader) was having "funny" permissions. Maybe more files were affected. I used fsck.ext3 and changed the attributes with chmod.

It may be caused by crashes of the notebook (power failure and suspend/resume failure), but I would expect that the administration of the fs would survive that, as it did for years. Actually, the last time I had filesystem corruption was in kernel 1.2.xx days...

As my notebook is not using raid, I suspect something in the ext3 code. Kernel is 2.6.1 with ACPI, acpi-dsdt and swsusp2 patches.

Regards,

Bas.

#8 lore raw

Hi.

Are you using a swap partition or swap file? If we're talking swap file, I would suspect suspend2. I haven't had a chance to look yet (preparing to move to Aussie), but Michael has told me there are problems with the swapfile support.

If you had a crash while using suspend and swap file support, I wouldn't be totally surprised to see an emergency sync causing this. That said, the code has a number of safety nets aimed at stopping us syncing while suspend is running, to avoid precisely this sort of corruption. If it was suspend, I'd expect your superblock to have been messed too. Did that happen?

Regards,

Nigel

Quoted 7 lines · Bas Mevissen
Quoted 6 lines · Theodore Ts'o
Quoted 6 lines

sfhq:/mnt/data/1/lost+found# ls -l
total 76
d-wSr-----    2 1212680233 136929556    49152 Jun  7  2008 #16370
-rwx-wx---    1 1628702729 135220664    45056 May  4  1974 #16380
#9 lore raw
Quoted 7 lines · Nigel Cunningham

Hi.

Are you using a swap partition or swap file? If we're talking swap file, I would suspect suspend2. I haven't had a chance to look yet (preparing to move to Aussie), but Michael has told me there are problems with the swapfile support.

You are reading everything, I beleave :-) Nope, only a swap partition. I read swsusp ML too :-)

I really suspect the power failure (forgot to plug in the AC adapter :-) ) was the "hard off" with the FS corruption.

Quoted 7 lines

If you had a crash while using suspend and swap file support, I wouldn't be totally surprised to see an emergency sync causing this. That said, the code has a number of safety nets aimed at stopping us syncing while suspend is running, to avoid precisely this sort of corruption. If it was suspend, I'd expect your superblock to have been messed too. Did that happen?

The superblock was fine. There was just a short file system check when booting again. I only noticed it a day later when the news reader didn't want to start anymore.

But thanks for thinking!

Regards,

Bas.

#10 lore raw
Quoted 1 line · Bas Mevissen

You are reading everything, I beleave :-)

Not everything! I have my email client set up to highlight messages that mention suspend or swsusp. :>

Nigel

Quoted 3 lines

The superblock was fine. There was just a short file system check when booting again. I only noticed it a day later when the news reader didn't want to start anymore.

Okay. Apparently not suspend related then.

Quoted 1 line

But thanks for thinking!

You're welcome.

Nigel

#13 lore raw
In reply to Bas Mevissen
Quoted 4 lines · Bas Mevissen

I've seen this kind of problems on my notebook too. Among others, over 600MB of a huge cache directory (from a news reader) was having "funny" permissions. Maybe more files were affected. I used fsck.ext3 and changed the attributes with chmod.

Was it just the permissions screwy? Was the contents of these files with the "funny" permission sane, or did they contain garbage? What about the modtime of the files?

In the original bug report random garbage had gotten written into the inode table. That sort of failure is generally caused at a level below that of the filesystem (i.e., device driver or hardware).

The question is whether the problems you are seeing seem to be caused by wholesale corruption of an entire block of the inode table, or is some other kind of problem. For example, if only the permissions are getting screwed up, when the rest of the inode data is correct, then yes, it would most likely be a filesystem bug. I haven't noticed any such problem myself, but it's possible that something like that might be going on. On the other hand, if it is an entire block in the inode table getting corrupted then I'd be less likely to presume it to be a filesystem flaw.

- Ted

#14 lore raw
Quoted 5 lines · Theodore Ts'o

Was it just the permissions screwy? Was the contents of these files with the "funny" permission sane, or did they contain garbage? What about the modtime of the files?

Only permissions. Something like r-Sr-S--- . File contents were OK.

Quoted 10 lines

The question is whether the problems you are seeing seem to be caused by wholesale corruption of an entire block of the inode table, or is some other kind of problem. For example, if only the permissions are getting screwed up, when the rest of the inode data is correct, then yes, it would most likely be a filesystem bug. I haven't noticed any such problem myself, but it's possible that something like that might be going on. On the other hand, if it is an entire block in the inode table getting corrupted then I'd be less likely to presume it to be a filesystem flaw.

It looks like this only appeared once. The FS looks fine now. So I guess
I won't be able to reproduce it. Let's just go to 2.6.[23] and see if
it happens again.

Regards,

Bas.

#11 lore raw
In reply to Nigel Cunningham

Hi!

Quoted 2 lines

Not everything! I have my email client set up to highlight messages that mention suspend or swsusp. :>

Clever ;-). [How does one do that in mutt?]
Pavel

Signature · 2 lines
When do you have a heart between your knees?
[Johanka's followup: and *two* hearts?]
#12 lore raw
Quoted 4 lines · Pavel Machek
Quoted 2 lines

Not everything! I have my email client set up to highlight messages that mention suspend or swsusp. :>

Clever ;-). [How does one do that in mutt?]

color index yellow black '(~b suspend | ~b swsusp)'

But if you use remote mailboxes (such as imap or pop) this sucks. So it's better to put something like:

folder-hook . "uncolor index *"
folder-hook . "color index yellow black '(~b suspend | ~b swsusp)'" folder-hook ^imaps?:// "uncolor index *"
folder-hook ^pops?:// "uncolor index *"

in your .muttrc to avoid downloading _all_ message bodies in the index when reading imap(s) or pop(s) mailboxes.

Signature · 1 line
Kiko
#15 lore raw
In reply to Bas Mevissen
Quoted 21 lines · Bas Mevissen
Quoted 5 lines · Theodore Ts'o

Was it just the permissions screwy? Was the contents of these files with the "funny" permission sane, or did they contain garbage? What about the modtime of the files?

Only permissions. Something like r-Sr-S--- . File contents were OK.

Quoted 10 lines

The question is whether the problems you are seeing seem to be caused by wholesale corruption of an entire block of the inode table, or is some other kind of problem. For example, if only the permissions are getting screwed up, when the rest of the inode data is correct, then yes, it would most likely be a filesystem bug. I haven't noticed any such problem myself, but it's possible that something like that might be going on. On the other hand, if it is an entire block in the inode table getting corrupted then I'd be less likely to presume it to be a filesystem flaw.

Quoted 3 lines

It looks like this only appeared once. The FS looks fine now. So I guess
I won't be able to reproduce it. Let's just go to 2.6.[23] and see if
it happens again.

Did this go away on reboot, or did you have to fix it? If it went away on reboot, it could be that the copy of the inode in memory was borked.

Signature · 3 lines
bill davidsen <[email protected]>
  CTO TMR Associates, Inc
  Doing interesting things with small computers since 1979
#19 lore raw
In reply to Theodore Ts'o
Quoted 8 lines · Theodore Ts'o
Quoted 8 lines · Jan Dittmer
Quoted 7 lines

EXT3-fs error (device dm-1): ext3_readdir: bad entry in directory #9783034: rec_len % 4 != 0 - offset=0, inode=1846971784, rec_len=33046, name_len=154
Aborting journal on device dm-1.
ext3_abort called.
EXT3-fs abort (device dm-1): ext3_journal_start: Detected aborted journal Remounting filesystem read-only

Has this been resolved? I have a machine due to enter production, am
considering going back to 2.4 if there is no further information.

Signature · 4 lines
Leandro Guimarães Faria Corsetti Dutra           +55 (11) 5685 2219
Av Sgto Geraldo Santana, 1100 6/71               +55 (11) 5686 9607
04.674-000  São Paulo, SP                                    BRASIL
http://br.geocities.com./lgcdutra/
#20 lore raw
Quoted 14 lines · Leandro Guimarães Faria Corsetti Dutra
Quoted 9 lines · Theodore Ts'o
Quoted 9 lines · Jan Dittmer
Quoted 8 lines

EXT3-fs error (device dm-1): ext3_readdir: bad entry in directory #9783034: rec_len % 4 != 0 - offset=0, inode=1846971784, rec_len=33046, name_len=154
Aborting journal on device dm-1.
ext3_abort called.
EXT3-fs abort (device dm-1): ext3_journal_start: Detected aborted journal Remounting filesystem read-only

Quoted 4 lines

Has this been resolved? I have a machine due to enter production, am
considering going back to 2.4 if there is no further information.

I haven't tried it with 2.6 since this incident. But considering that the machine in question crashed a couple of times afterwards, it may well be, that a hardware fault caused this initially. But I simply don't dare to put 2.6 again on this machine as I've no real backup of most of the data, and restoring some 100 gb from cds is really annoying.

Jan

#16 lore raw
In reply to Bill Davidsen
Quoted 8 lines · Bill Davidsen
Quoted 4 lines

It looks like this only appeared once. The FS looks fine now. So I guess
I won't be able to reproduce it. Let's just go to 2.6.[23] and see if
it happens again.

Did this go away on reboot, or did you have to fix it? If it went away on reboot, it could be that the copy of the inode in memory was borked.

I really had to fix it. But, it never appeared again until now. So maybe this was just caused by some crash while experimenting with swsusp2.

Regards,

Bas.

#21 lore raw
In reply to Jan Dittmer

Em Qui, 2004-02-19 às 05:07, Jan Dittmer escreveu:

Quoted 10 lines
Quoted 9 lines · Leandro Guimarães Faria Corsetti Dutra
Quoted 8 lines
Quoted 8 lines · Jan Dittmer
Quoted 7 lines

EXT3-fs error (device dm-1): ext3_readdir: bad entry in directory #9783034: rec_len % 4 != 0 - offset=0, inode=1846971784, rec_len=33046, name_len=154
Aborting journal on device dm-1.
ext3_abort called.
EXT3-fs abort (device dm-1): ext3_journal_start: Detected aborted journal Remounting filesystem read-only

Quoted 6 lines
Quoted 2 lines

Has this been resolved? I have a machine due to enter production, am
considering going back to 2.4 if there is no further information.

I haven't tried it with 2.6 since this incident. But considering that the machine in question crashed a couple of times afterwards, it may well be, that a hardware fault caused this initially.

I don't believe so, because I have already run both memtest86 (all
tests, a dozen passes) and cpuburn (burnP6, several hours). I am now trying to figure out how to run fsx-linux, documentation seems scarce.

Quoted 3 lines

But I simply don't
dare to put 2.6 again on this machine as I've no real backup of most of the data, and restoring some 100 gb from cds is really annoying.

That's the position I think I will have to face. A pity, I was already
hooked into 2.6's speed and LVM2.

Signature · 4 lines
Leandro Guimarães Faria Corcete Dutra
Maringá, PR, BRASIL                 +55 (11) 5685 2219
http://br.geocities.com./lgcdutra/  +55 (44) 3028 7467
Soli Deo Gloria!                    +55 (44) 3028 8335
#22 lore raw
Quoted 2 lines · Leandro Guimarães Faria Corcete Dutra

I am now
trying to figure out how to run fsx-linux, documentation seems scarce.

If anyone is interested on fsx-linux output for a PROBLEM report,
please tell me how to run it. Google and even the sources were of no help to me.

No big deal, I will produce a PROBLEM report with or without it.

Signature · 4 lines
Leandro Guimarães Faria Corsetti Dutra           +55 (11) 5685 2219
Av Sgto Geraldo Santana, 1100 6/71               +55 (11) 5686 9607
04.674-000  São Paulo, SP                                    BRASIL
http://br.geocities.com./lgcdutra/
#17 lore raw
In reply to Bas Mevissen
Quoted 6 lines · Bas Mevissen
Quoted 5 lines · Bill Davidsen
Quoted 4 lines

It looks like this only appeared once. The FS looks fine now. So I guess
I won't be able to reproduce it. Let's just go to 2.6.[23] and see if
it happens again.

Quoted 6 lines
Quoted 3 lines

Did this go away on reboot, or did you have to fix it? If it went away on reboot, it could be that the copy of the inode in memory was borked.

I really had to fix it. But, it never appeared again until now. So maybe this was just caused by some crash while experimenting with swsusp2.

Another WAG bites the dust, but at least it's clear that you had damage to the metadata on the media. Hopefully that would be useful if it comes back to visit again.

Glad you COULD fix it!

Signature · 3 lines
bill davidsen <[email protected]>
  CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.