Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: Gentoo: PPC-User

Filesystem Corruption revisited

 

 

Gentoo ppc-user RSS feed   Index | Next | Previous | View Threaded


ben.ricker at gmail

Dec 28, 2004, 7:57 AM

Post #1 of 4 (489 views)
Permalink
Filesystem Corruption revisited

I believe there is a kernel issue (at least with my kernel). Less then
24 hours after fixing the /dev/hda8 filesystem which was fubar, I now
have BOTH hda8 and hda9 go read-only (the first symptom I see).

Here is some information from the logs. I am seeing a ton of the
following error:

Dec 28 06:03:57 anime kernel: evbug.c: Event. Dev: adb2:2.02/input,
Type: 1, Code: 57, Value: 1
Dec 28 06:03:57 anime kernel: evbug.c: Event. Dev: adb2:2.02/input,
Type: 0, Code: 0, Value: 0

This looks like something with the adb driver (keyboard/mouse are
using it). Not sure why it would affect filesystems.

I also get this one. I am running a static kernel, so I am not sure
why this is even happening. Again, I do not see how it would affect
filesystems:

Dec 27 21:27:21 anime modprobe: FATAL: Could not load
lib/modules/2.6.9-gentoo-r1/modules.dep: No such file or directory
Dec 27 21:27:21 anime modprobe: FATAL: Could not load
/lib/modules/2.6.9-gentoo-r1/modules.dep: No such file or directory

There is one other error I see in the kernel log:

Dec 28 06:52:31 anime kernel: evbug.c: Event. Dev: macio/input0, Type:
18, Code: 2, Value: 750
Dec 28 06:52:31 anime kernel: evbug.c: Event. Dev: macio/input0, Type:
18, Code: 2, Value: 0
Dec 28 06:52:31 anime kernel: evbug.c: Event. Dev: macio/input0, Type:
18, Code: 1, Value: 0

Finally, the coup de grace. No time stamps since this is in dmesg, but
it is the last thing logged:

EXT3-fs error (device hda8): ext3_free_blocks: Freeing blocks not in
datazone - block = 1327899396, count = 1
Aborting journal on device hda8.
ext3_abort called.
EXT3-fs error (device hda8): ext3_journal_start: Detected aborted journal
Remounting filesystem read-only
EXT3-fs error (device hda8) in start_transaction: Journal has aborted
EXT3-fs error (device hda8) in start_transaction: Journal has aborted
EXT3-fs error (device hda8): ext3_free_blocks: Freeing blocks not in
datazone - block = 855704320, count = 1
ext3_reserve_inode_write: aborting transaction: Journal has aborted in
__ext3_journal_get_write_access<2>EXT3-fs error (device hda8) in
ext3_reserve_inode_write: Journal has aborted
EXT3-fs error (device hda8) in ext3_truncate: Journal has aborted
ext3_reserve_inode_write: aborting transaction: Journal has aborted in
__ext3_journal_get_write_access<2>EXT3-fs error (device hda8) in
ext3_reserve_inode_write: Journal has aborted
EXT3-fs error (device hda8) in ext3_orphan_del: Journal has aborted
ext3_reserve_inode_write: aborting transaction: Journal has aborted in
__ext3_journal_get_write_access<2>EXT3-fs error (device hda8) in
ext3_reserve_inode_write: Journal has aborted
EXT3-fs error (device hda8) in ext3_delete_inode: Journal has aborted
__journal_remove_journal_head: freeing b_committed_data
EXT3-fs error (device hda8) in start_transaction: Journal has aborted

I ran a dd yesterday that copied the entire /dev/hda8 to /dev/null and
it completed with no errors. I also ran an fsck -c and there were no
bad blocks found.

Unless I can get some direction here, I am thinking of trying to
install YellowDog4 and see if it is more stable. Of course, if it does
not fail, then I know it was Gentoo. If it does, then it must be
hardware related.

Any help would be be appreciated,

Ben Ricker


--
Ben Ricker
He's just this guy, you know?

--
gentoo-ppc-user [at] gentoo mailing list


josejx at gentoo

Dec 28, 2004, 8:31 AM

Post #2 of 4 (464 views)
Permalink
Re: Filesystem Corruption revisited [In reply to]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Flaffer wrote:
| I believe there is a kernel issue (at least with my kernel). Less then
| 24 hours after fixing the /dev/hda8 filesystem which was fubar, I now
| have BOTH hda8 and hda9 go read-only (the first symptom I see).
|
| Here is some information from the logs. I am seeing a ton of the
| following error:
|
| Dec 28 06:03:57 anime kernel: evbug.c: Event. Dev: adb2:2.02/input,
| Type: 1, Code: 57, Value: 1
| Dec 28 06:03:57 anime kernel: evbug.c: Event. Dev: adb2:2.02/input,
| Type: 0, Code: 0, Value: 0
|
| This looks like something with the adb driver (keyboard/mouse are
| using it). Not sure why it would affect filesystems.
|

Yes, this is the input event debugging system, turn it off in your
kernel config, it's useless and just clutters things up.

| I also get this one. I am running a static kernel, so I am not sure
| why this is even happening. Again, I do not see how it would affect
| filesystems:
|
| Dec 27 21:27:21 anime modprobe: FATAL: Could not load
| lib/modules/2.6.9-gentoo-r1/modules.dep: No such file or directory
| Dec 27 21:27:21 anime modprobe: FATAL: Could not load
| /lib/modules/2.6.9-gentoo-r1/modules.dep: No such file or directory

Build make modules_install even if you don't use modules and this should
go away.

| There is one other error I see in the kernel log:
|
| Dec 28 06:52:31 anime kernel: evbug.c: Event. Dev: macio/input0, Type:
| 18, Code: 2, Value: 750
| Dec 28 06:52:31 anime kernel: evbug.c: Event. Dev: macio/input0, Type:
| 18, Code: 2, Value: 0
| Dec 28 06:52:31 anime kernel: evbug.c: Event. Dev: macio/input0, Type:
| 18, Code: 1, Value: 0
|

This is also input event debugging.

| Finally, the coup de grace. No time stamps since this is in dmesg, but
| it is the last thing logged:
|
| EXT3-fs error (device hda8): ext3_free_blocks: Freeing blocks not in
| datazone - block = 1327899396, count = 1
| Aborting journal on device hda8.
| ext3_abort called.
| EXT3-fs error (device hda8): ext3_journal_start: Detected aborted journal
| Remounting filesystem read-only
| EXT3-fs error (device hda8) in start_transaction: Journal has aborted
| EXT3-fs error (device hda8) in start_transaction: Journal has aborted
| EXT3-fs error (device hda8): ext3_free_blocks: Freeing blocks not in
| datazone - block = 855704320, count = 1
| ext3_reserve_inode_write: aborting transaction: Journal has aborted in
| __ext3_journal_get_write_access<2>EXT3-fs error (device hda8) in
| ext3_reserve_inode_write: Journal has aborted
| EXT3-fs error (device hda8) in ext3_truncate: Journal has aborted
| ext3_reserve_inode_write: aborting transaction: Journal has aborted in
| __ext3_journal_get_write_access<2>EXT3-fs error (device hda8) in
| ext3_reserve_inode_write: Journal has aborted
| EXT3-fs error (device hda8) in ext3_orphan_del: Journal has aborted
| ext3_reserve_inode_write: aborting transaction: Journal has aborted in
| __ext3_journal_get_write_access<2>EXT3-fs error (device hda8) in
| ext3_reserve_inode_write: Journal has aborted
| EXT3-fs error (device hda8) in ext3_delete_inode: Journal has aborted
| __journal_remove_journal_head: freeing b_committed_data
| EXT3-fs error (device hda8) in start_transaction: Journal has aborted
|
| I ran a dd yesterday that copied the entire /dev/hda8 to /dev/null and
| it completed with no errors. I also ran an fsck -c and there were no
| bad blocks found.
|
| Unless I can get some direction here, I am thinking of trying to
| install YellowDog4 and see if it is more stable. Of course, if it does
| not fail, then I know it was Gentoo. If it does, then it must be
| hardware related.
|
| Any help would be be appreciated,
|
| Ben Ricker
|

It may be worth trying again without the event debugging to see if any
other errors catch your eye. Is your kernel compiled with PREEMPT on?
Which kernel are you using? Can you post your config somewhere where we
can see it? I'd lean on the side of hardware problem, but I'm not positive.

Good luck!
- -Joe

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFB0XxBwGq7BLLARfoRAkxvAKCMuXG9i8Bqe62z9cnNvZPqmlWpeACg0NjU
W72woXCbwii0Pjl4I3a4fEQ=
=pwWN
-----END PGP SIGNATURE-----

--
gentoo-ppc-user [at] gentoo mailing list


fusero at wanadoo

Dec 28, 2004, 9:51 AM

Post #3 of 4 (462 views)
Permalink
Re: Filesystem Corruption revisited [In reply to]

El mié, 29-12-2004 a las 02:31 +1100, Joseph Jezak escribió:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Flaffer wrote:
> | I believe there is a kernel issue (at least with my kernel). Less then
> | 24 hours after fixing the /dev/hda8 filesystem which was fubar, I now
> | have BOTH hda8 and hda9 go read-only (the first symptom I see).
> |
> | Here is some information from the logs. I am seeing a ton of the
> | following error:
> |
> | Dec 28 06:03:57 anime kernel: evbug.c: Event. Dev: adb2:2.02/input,
> | Type: 1, Code: 57, Value: 1
> | Dec 28 06:03:57 anime kernel: evbug.c: Event. Dev: adb2:2.02/input,
> | Type: 0, Code: 0, Value: 0
> |
> | This looks like something with the adb driver (keyboard/mouse are
> | using it). Not sure why it would affect filesystems.
> |
>
> Yes, this is the input event debugging system, turn it off in your
> kernel config, it's useless and just clutters things up.
only wanted to point that it can be also a security problem

> und.
> |
> | Unless I can get some direction here, I am thinking of trying to
> | install YellowDog4 and see if it is more stable. Of course, if it does
> | not fail, then I know it was Gentoo. If it does, then it must be
> | hardware related.
> |
> | Any help would be be appreciated,
> |
> | Ben Ricker
> |
>
> It may be worth trying again without the event debugging to see if any
> other errors catch your eye. Is your kernel compiled with PREEMPT on?
> Which kernel are you using? Can you post your config somewhere where we
> can see it? I'd lean on the side of hardware problem, but I'm not positive.
>

Why don't you try with the latest kernel? It seems that 2.6.10 comes
with tons of fixes...

See you, Fernando


--
gentoo-ppc-user [at] gentoo mailing list


ben.ricker at gmail

Dec 28, 2004, 10:03 AM

Post #4 of 4 (465 views)
Permalink
Re: Filesystem Corruption revisited [In reply to]

On Tue, 28 Dec 2004 17:51:54 +0100, Fernando Usero <fusero [at] wanadoo> wrote:

> Why don't you try with the latest kernel? It seems that 2.6.10 comes
> with tons of fixes...

I will be using that. I may also do the .9 version just to see if the
issue still exists. Then, I will go to .10. Or maybe I will just go to
.10; depends on how lazy I get when I get to the point of doing it!

Thanks,

Ben Ricker

--
Ben Ricker
He's just this guy, you know?

--
gentoo-ppc-user [at] gentoo mailing list

Gentoo ppc-user RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.