Linux Archive

Linux Archive (http://www.linux-archive.org/)
-   EXT3 Users (http://www.linux-archive.org/ext3-users/)
-   -   bind mounts, chroots, and mismatching kernel with userland (http://www.linux-archive.org/ext3-users/694177-bind-mounts-chroots-mismatching-kernel-userland.html)

Jayen 08-15-2012 10:16 AM

bind mounts, chroots, and mismatching kernel with userland
 
I have a 32-bit debian system, but I recently needed to mmap large
files, so I started using a 64-bit kernel with my 32-bit userland and
installed a 64-bit debian chroot to run inside. I use bind mounts for
proc, sys, tmp, dev, and home. It all works fine, but occasionally, the
filesystem gets corrupted (/ and /home are on the same system). I am
running linux 3.2.21 (3.2.0-3 in debian).

Have I done something I shouldn't do?

I hope this is the right mailing list for this question; I couldn't find
another linux fs user mailing list.

Thanks,
Jayen

_______________________________________________
Ext3-users mailing list
Ext3-users@redhat.com
https://www.redhat.com/mailman/listinfo/ext3-users

Christian Kujau 08-16-2012 05:15 AM

bind mounts, chroots, and mismatching kernel with userland
 
On Wed, 15 Aug 2012 at 20:16, Jayen wrote:
> proc, sys, tmp, dev, and home. It all works fine, but occasionally, the
> filesystem gets corrupted (/ and /home are on the same system). I am
> running linux 3.2.21 (3.2.0-3 in debian).

"Corrupted, as in....?" - Please post error messages, mountoptions and
fsck outputs, if attempted.

Thanks,
Christian.
--
BOFH excuse #116:

the real ttys became pseudo ttys and vice-versa.

_______________________________________________
Ext3-users mailing list
Ext3-users@redhat.com
https://www.redhat.com/mailman/listinfo/ext3-users

Jayen 08-16-2012 07:19 AM

bind mounts, chroots, and mismatching kernel with userland
 
Corrupted, as in /boot disappeared once and fsck reports heaps of
errors. (fsck did not successfully find /boot that time, and only found
one initrd (out of 3), no kernels).

i don't have any error messages at the moment (because the disk becomes
read only), but i'll take a picture next time it happens. is there any
way to record the fsck output to a text file if it's running from the
drive itself?

mount options:
/dev/sda1 on / type ext4
(rw,noatime,errors=remount-ro,user_xattr,barrier=1,data=ordered,discard)
tmpfs on /home/jayen/pkg/64/tmp type tmpfs
(rw,nosuid,nodev,relatime,size=814512k)
/dev/sda1 on /home/jayen/pkg/64/home type ext4
(rw,noatime,errors=remount-ro,user_xattr,barrier=1,data=ordered,discard)
proc on /home/jayen/pkg/64/proc type proc (rw,nosuid,nodev,noexec,relatime)
udev on /home/jayen/pkg/64/dev type devtmpfs
(rw,relatime,size=10240k,nr_inodes=213489,mode=755 )
sysfs on /home/jayen/pkg/64/sys type sysfs (rw,nosuid,nodev,noexec,relatime)

fstab:
proc /proc proc defaults 0 0
UUID=2b04563a-22b9-4b8a-b7c3-98774b749d7e / ext4
errors=remount-ro,noatime,discard 0 1

/tmp /home/jayen/pkg/64/tmp bind defaults,bind 0 0
/home /home/jayen/pkg/64/home bind defaults,bind 0 0
/proc /home/jayen/pkg/64/proc bind defaults,bind 0 0
/dev /home/jayen/pkg/64/dev bind defaults,bind 0 0
/sys /home/jayen/pkg/64/sys bind defaults,bind 0 0

Thanks,
Jayen

On 16/08/12 15:15, Christian Kujau wrote:
> On Wed, 15 Aug 2012 at 20:16, Jayen wrote:
>> proc, sys, tmp, dev, and home. It all works fine, but occasionally, the
>> filesystem gets corrupted (/ and /home are on the same system). I am
>> running linux 3.2.21 (3.2.0-3 in debian).
>
> "Corrupted, as in....?" - Please post error messages, mountoptions and
> fsck outputs, if attempted.
>
> Thanks,
> Christian.
>

_______________________________________________
Ext3-users mailing list
Ext3-users@redhat.com
https://www.redhat.com/mailman/listinfo/ext3-users

Bruno Wolff III 08-16-2012 12:48 PM

bind mounts, chroots, and mismatching kernel with userland
 
On Thu, Aug 16, 2012 at 17:19:27 +1000,
Jayen <jayen@jayenashar.org> wrote:


i don't have any error messages at the moment (because the disk becomes
read only), but i'll take a picture next time it happens. is there any
way to record the fsck output to a text file if it's running from the
drive itself?


Some options are using a removable drive or netconsole.

_______________________________________________
Ext3-users mailing list
Ext3-users@redhat.com
https://www.redhat.com/mailman/listinfo/ext3-users

Jayen Ashar 08-23-2012 04:01 AM

bind mounts, chroots, and mismatching kernel with userland
 
Thanks for your assistance.

dmesg output:

[609652.499426] EXT4-fs error (device sda1):
ext4_mb_generate_buddy:739: group 329, 19 clusters in bitmap, 0 in gd
[609652.499433] Aborting journal on device sda1-8.
[609652.499536] EXT4-fs (sda1): Remounting filesystem read-only
[609652.499545] EXT4-fs error (device sda1) in
ext4_reserve_inode_write:4475: Journal has aborted
[609652.499638] EXT4-fs error (device sda1) in
ext4_reserve_inode_write:4475: Journal has aborted
[609652.499731] EXT4-fs error (device sda1) in
ext4_ext_remove_space:2637: Journal has aborted
[609652.499822] EXT4-fs error (device sda1) in
ext4_reserve_inode_write:4475: Journal has aborted
[609652.499912] EXT4-fs error (device sda1) in ext4_ext_truncate:4251:
Journal has aborted
[609652.500037] EXT4-fs error (device sda1) in
ext4_reserve_inode_write:4475: Journal has aborted
[609652.500141] EXT4-fs error (device sda1) in ext4_orphan_del:2115:
Journal has aborted
[609652.500323] EXT4-fs error (device sda1) in
ext4_reserve_inode_write:4475: Journal has aborted

fsck output attached.

mount options:
/dev/sda1 on / type ext4
(rw,noatime,errors=remount-ro,user_xattr,barrier=1,data=ordered,discard)
tmpfs on /home/jayen/pkg/64/tmp type tmpfs
(rw,nosuid,nodev,relatime,size=814512k)
/dev/sda1 on /home/jayen/pkg/64/home type ext4
(rw,noatime,errors=remount-ro,user_xattr,barrier=1,data=ordered,discard)
proc on /home/jayen/pkg/64/proc type proc (rw,nosuid,nodev,noexec,relatime)
udev on /home/jayen/pkg/64/dev type devtmpfs
(rw,relatime,size=10240k,nr_inodes=213489,mode=755 )
sysfs on /home/jayen/pkg/64/sys type sysfs (rw,nosuid,nodev,noexec,relatime)

fstab:
proc /proc proc defaults 0 0
UUID=2b04563a-22b9-4b8a-b7c3-98774b749d7e / ext4
errors=remount-ro,noatime,discard 0 1
/tmp /home/jayen/pkg/64/tmp bind defaults,bind 0 0
/home /home/jayen/pkg/64/home bind defaults,bind 0 0
/proc /home/jayen/pkg/64/proc bind defaults,bind 0 0
/dev /home/jayen/pkg/64/dev bind defaults,bind 0 0
/sys /home/jayen/pkg/64/sys bind defaults,bind 0 0

Thanks,
Jayen

On 16/08/12 22:48, Bruno Wolff III wrote:
> On Thu, Aug 16, 2012 at 17:19:27 +1000,
> Jayen <jayen@jayenashar.org> wrote:
>>
>> i don't have any error messages at the moment (because the disk becomes
>> read only), but i'll take a picture next time it happens. is there any
>> way to record the fsck output to a text file if it's running from the
>> drive itself?
>
> Some options are using a removable drive or netconsole.
fsck from util-linux 2.20.1
/dev/sda1 contains a file system with errors, check forced.
Pass 1: Checking inodes, blocks, and sizes
Inodes that were part of a corrupted orphan linked list found. Fix? yes

Inode 1318990 was part of the orphaned inode list. FIXED.
Inode 1319060 was part of the orphaned inode list. FIXED.
Inode 2369258 was part of the orphaned inode list. FIXED.
Inode 2369315 was part of the orphaned inode list. FIXED.
Inode 2369316 was part of the orphaned inode list. FIXED.
Inode 2369318 was part of the orphaned inode list. FIXED.
Inode 2369319 was part of the orphaned inode list. FIXED.
Inode 2369322 was part of the orphaned inode list. FIXED.
Inode 2369323 was part of the orphaned inode list. FIXED.
Inode 2369325 was part of the orphaned inode list. FIXED.
Inode 2369327 was part of the orphaned inode list. FIXED.
Inode 2369328 was part of the orphaned inode list. FIXED.
Inode 2369332 was part of the orphaned inode list. FIXED.
Inode 2369333 was part of the orphaned inode list. FIXED.
Inode 2369334 was part of the orphaned inode list. FIXED.
Inode 2369336 was part of the orphaned inode list. FIXED.
Inode 2369337 was part of the orphaned inode list. FIXED.
Inode 2369338 was part of the orphaned inode list. FIXED.
Inode 2369339 was part of the orphaned inode list. FIXED.
Inode 2369341 was part of the orphaned inode list. FIXED.
Inode 2369342 was part of the orphaned inode list. FIXED.
Inode 2369344 was part of the orphaned inode list. FIXED.
Inode 2369345 was part of the orphaned inode list. FIXED.
Inode 2369346 was part of the orphaned inode list. FIXED.
Inode 2369348 was part of the orphaned inode list. FIXED.
Inode 2369349 was part of the orphaned inode list. FIXED.
Inode 2369350 was part of the orphaned inode list. FIXED.
Inode 2369352 was part of the orphaned inode list. FIXED.
Inode 2369353 was part of the orphaned inode list. FIXED.
Inode 2369354 was part of the orphaned inode list. FIXED.
Inode 2369355 was part of the orphaned inode list. FIXED.
Inode 2369356 was part of the orphaned inode list. FIXED.
Inode 2369360 was part of the orphaned inode list. FIXED.
Inode 2369361 was part of the orphaned inode list. FIXED.
Inode 2369363 was part of the orphaned inode list. FIXED.
Inode 2369364 was part of the orphaned inode list. FIXED.
Inode 2369365 was part of the orphaned inode list. FIXED.
Inode 2369366 was part of the orphaned inode list. FIXED.
Deleted inode 2375966 has zero dtime. Fix? yes

Inode 3321328 was part of the orphaned inode list. FIXED.
Inode 3321333 was part of the orphaned inode list. FIXED.
Inode 3321335 was part of the orphaned inode list. FIXED.
Inode 3321337 was part of the orphaned inode list. FIXED.
Inode 3321339 was part of the orphaned inode list. FIXED.
Inode 3321343 was part of the orphaned inode list. FIXED.
Inode 3321348 was part of the orphaned inode list. FIXED.
Inode 3321349 was part of the orphaned inode list. FIXED.
Inode 3321351 was part of the orphaned inode list. FIXED.
Inode 3321355 was part of the orphaned inode list. FIXED.
Inode 3321357 was part of the orphaned inode list. FIXED.
Inode 3321365 was part of the orphaned inode list. FIXED.
Inode 3321367 was part of the orphaned inode list. FIXED.
Inode 3321412 was part of the orphaned inode list. FIXED.
Inode 3321533 was part of the orphaned inode list. FIXED.
Inode 3321554 was part of the orphaned inode list. FIXED.
Inode 3321631 was part of the orphaned inode list. FIXED.
Inode 3323447 was part of the orphaned inode list. FIXED.
Inode 3323449 was part of the orphaned inode list. FIXED.
Inode 3323767 was part of the orphaned inode list. FIXED.
Inode 3323768 was part of the orphaned inode list. FIXED.
Inode 3323769 was part of the orphaned inode list. FIXED.
Inode 3323770 was part of the orphaned inode list. FIXED.
Inode 3323771 was part of the orphaned inode list. FIXED.
Inode 3323772 was part of the orphaned inode list. FIXED.
Inode 3323773 was part of the orphaned inode list. FIXED.
Inode 3323774 was part of the orphaned inode list. FIXED.
Inode 3323775 was part of the orphaned inode list. FIXED.
Inode 3434204 was part of the orphaned inode list. FIXED.
Inode 3434206 was part of the orphaned inode list. FIXED.
Inode 3434207 was part of the orphaned inode list. FIXED.
Inode 3434208 was part of the orphaned inode list. FIXED.
Inode 3434213 was part of the orphaned inode list. FIXED.
Inode 3434215 was part of the orphaned inode list. FIXED.
Inode 3434216 was part of the orphaned inode list. FIXED.
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Block bitmap differences: -(5668365--5668381) -(5695376--5695384) -9473937 -(10440704--10443488) -(10633207--10633215) -(10645232--10645233) -(10645235--10645240) -(10645304--10645348) -(10645350--10645363) -(10645365--10645368) -(10645370--10645383) -(10645385--10645394) -(10645409--10645416) -(10645418--10645421) -(10645423--10645429) -(10645449--10645463) -(10645466--10645471) -(10677389--10677407) +10789507 +10789593 +10789805 +10789856 +10789878 +10789907 +10790324 +10790510 +10790563 +10790889 +10790907 +10791394 +10791752 +10791809 +10791815 +10792378 +10792445 +10792467 +10792822 -(10826700--10827300) -(13606912--13608256) -(13608911--13609803) -(13610048--13610111) -(13610392--13611486) -(13613249--13613496) -(13619584--13620296) -(13620549--13622325) -(13622884--13623206) -(13623208--13623293) -(13629362--13629414) -(13647169--13647344) -(13647857--13648994) -(13823110--13823195) -(13826002--13826011) -(13826016--13826057) -(13826064--13826078) -(13828432--13828874) -(14559179--14559343) -(14559494--14559904) -(14560501--14560577) -(14560636--14560777)
Fix? yes

Free blocks count wrong for group #172 (0, counted=17).
Fix? yes

Free blocks count wrong for group #173 (2418, counted=2427).
Fix? yes

Free blocks count wrong for group #289 (0, counted=1).
Fix? yes

Free blocks count wrong for group #318 (32, counted=2817).
Fix? yes

Free blocks count wrong for group #324 (0, counted=144).
Fix? yes

Free blocks count wrong for group #325 (780, counted=799).
Fix? yes

Free blocks count wrong for group #330 (186, counted=787).
Fix? yes

Free blocks count wrong for group #415 (16, counted=6613).
Fix? yes

Free blocks count wrong for group #416 (16, counted=1330).
Fix? yes

Free blocks count wrong for group #421 (0, counted=153).
Fix? yes

Free blocks count wrong for group #422 (0, counted=443).
Fix? yes

Free blocks count wrong for group #444 (0, counted=795).
Fix? yes

Free blocks count wrong (788752, counted=801630).
Fix? yes

Inode bitmap differences: -1318990 -1319060 -2369258 -(2369315--2369316) -(2369318--2369319) -(2369322--2369323) -2369325 -(2369327--2369328) -(2369332--2369334) -(2369336--2369339) -(2369341--2369342) -(2369344--2369346) -(2369348--2369350) -(2369352--2369356) -(2369360--2369361) -(2369363--2369366) -2375966 -3321328 -3321333 -3321335 -3321337 -3321339 -3321343 -(3321348--3321349) -3321351 -3321355 -3321357 -3321365 -3321367 -3321412 -3321533 -3321554 -3321631 -3323447 -3323449 -(3323767--3323775) -3434204 -(3434206--3434208) -3434213 -(3434215--3434216)
Fix? yes

Free inodes count wrong for group #161 (1494, counted=1496).
Fix? yes

Free inodes count wrong for group #289 (1, counted=37).
Fix? yes

Free inodes count wrong for group #290 (0, counted=1).
Fix? yes

Free inodes count wrong for group #405 (149, counted=177).
Fix? yes

Free inodes count wrong for group #419 (0, counted=7).
Fix? yes

Free inodes count wrong (3190789, counted=3190863).
Fix? yes


/dev/sda1: ***** FILE SYSTEM WAS MODIFIED *****
/dev/sda1: ***** REBOOT LINUX *****
/dev/sda1: 716721/3907584 files (0.2% non-contiguous), 14827599/15629229 blocks
_______________________________________________
Ext3-users mailing list
Ext3-users@redhat.com
https://www.redhat.com/mailman/listinfo/ext3-users

Christian Kujau 09-04-2012 11:37 AM

bind mounts, chroots, and mismatching kernel with userland
 
On Thu, 23 Aug 2012 at 14:01, Jayen Ashar wrote:
> [609652.499426] EXT4-fs error (device sda1):
> ext4_mb_generate_buddy:739: group 329, 19 clusters in bitmap, 0 in gd
> [609652.499433] Aborting journal on device sda1-8.

Sorry for the late reply. Is this still happening? Searching the interwebs
for "Aborting journal on device..." reveals a lot of hardware related
issues. Were there any messages before that fs error?

Christian.

_______________________________________________
Ext3-users mailing list
Ext3-users@redhat.com
https://www.redhat.com/mailman/listinfo/ext3-users

Jayen Ashar 09-04-2012 11:04 PM

bind mounts, chroots, and mismatching kernel with userland
 
No, there were no messages before that, but I agree now that it is a
hardware error. It started immediately after I started using the
chroot, so I assumed it was related. However, in the last two weeks,
I've reformatted, stopped using the chroot, removed the bind mounts
and the 64-bit kernel, and it is still happening.

Sorry for jumping to the wrong conclusion.

--Jayen

On Tue, Sep 4, 2012 at 9:37 PM, Christian Kujau <lists@nerdbynature.de> wrote:
> On Thu, 23 Aug 2012 at 14:01, Jayen Ashar wrote:
>> [609652.499426] EXT4-fs error (device sda1):
>> ext4_mb_generate_buddy:739: group 329, 19 clusters in bitmap, 0 in gd
>> [609652.499433] Aborting journal on device sda1-8.
>
> Sorry for the late reply. Is this still happening? Searching the interwebs
> for "Aborting journal on device..." reveals a lot of hardware related
> issues. Were there any messages before that fs error?
>
> Christian.

_______________________________________________
Ext3-users mailing list
Ext3-users@redhat.com
https://www.redhat.com/mailman/listinfo/ext3-users


All times are GMT. The time now is 09:41 PM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.