[Ocfs2-users] crash on mount

[Sunil Mushran]

Did you save the earlier debugfs output?

For this case, one solution is to rerun fsck. It should fix the flag. It should. Else we have a bug. Other solution is to hand edit it. Which is riskier. So let's give fsck another try. 

On Jan 29, 2011, at 12:07 AM, Christophe Bouder <Christophe.Bouder at lip6.fr> wrote:

> many thanks for your answer,
> 
> i do it
> at he end of fsck i have :
> many :
> [INODE_ORPHANED] Inode 4207284830 was found in the orphan directory. Delete its contents and unlink it? y
> then :
> Pass 4b: Checking inodes link counts.
> [INODE_COUNT] Inode 525 has a link count of 1373 on disk but directory entry references come to 2. Update the count on disk to match? y
> [INODE_COUNT] Inode 536 has a link count of 7 on disk but directory entry references come to 2. Update the count on disk to match? y
> [INODE_COUNT] Inode 3134107580 has a link count of 0 on disk but directory entry references come to 65534. Update the count on disk to match? y
> [INODE_NOT_CONNECTED] Inode 3378844300 isn't referenced by any directory entries.  Move it to lost+found? y
> [INODE_COUNT] Inode 3378844300 has a link count of 65535 on disk but directory entry references come to 1. Update the count on disk to match? y
> Pass 5: Checking quota information.
> All passes succeeded.
> Slot 1's journal dirty flag removed
> 
> and when i remount it don't crash but :
> Jan 29 08:55:03 bigtwentyfour kernel: ocfs2: Mounting device (8,1) on (node 24, slot 0) with ordered data mode.
> Jan 29 08:55:04 bigtwentyfour kernel: OCFS2: ERROR (device sda1): ocfs2_validate_inode_block: Invalid dinode #3378844300: OCFS2_VALID_FL not set
> Jan 29 08:55:04 bigtwentyfour kernel:
> Jan 29 08:55:04 bigtwentyfour kernel: File system is now read-only due to the potential of on-disk corruption. Please run fsck.ocfs2 once the file system is unmounted.
> 
> do i need an other fsck.ocsf2 directly ?
> 
> 
> 
> 
> On Fri, 28 Jan 2011 15:04:52 -0800, Sunil Mushran wrote:
>> Dump the following:
>> 
>> debugfs.ocfs2 -R "stat " /dev/sdX
>> 
>> This will save the errant inode.
>> 
>> Then run fsck -fy. Hopefully that will fix it.
>> 
>> On 01/28/2011 02:50 PM, Christophe Bouder wrote:
>> 
>>> hi,
>>> i run debian with 2.6.37
>>> and ocfs2-tools  1.6.3-1
>>> i have  22  nodes
>>> and a fiber channel connect device of 33T.
>>> everything was ok.
>>> but suddenly it does not respond.
>>> So i switch off all the nodes
>>> then switch on only one (i have try different)
>>> and when i try to mount it crashed perpetually
>>> see the log attach.
>>> 
>>> any help welcome.
>>> thank you.
>>> 
>>> _______________________________________________
>>> Ocfs2-users mailing list
>>> Ocfs2-users at oss.oracle.com [1]
>>> http://oss.oracle.com/mailman/listinfo/ocfs2-users [2]
>> 
>> 
>> 
>> Links:
>> ------
>> [1] mailto:Ocfs2-users at oss.oracle.com
>> [2] http://oss.oracle.com/mailman/listinfo/ocfs2-users
> 
> cheers
> -- 
> Christophe Bouder
>