Re: kernel BUG at fs/btrfs/inode.c:149!

[Josef Bacik <josef@redhat.com>]

On 05/16/2011 11:01 AM, Whirm wrote:
> On Monday 16 May 2011 16:11:19 Josef Bacik wrote:
>> On 05/16/2011 05:17 AM, whirm@gmx.com wrote:
>>> On Friday 13 May 2011 20:52:22 Josef Bacik wrote:
>>>> On 05/13/2011 01:19 PM, whirm@gmx.com wrote:
>>>>> On Thursday 05 May 2011 20:57:17 Josef Bacik wrote:
>>>>> [..]
>>>>>
>>>>>> It doesn't look like that bit had my debugging output.  Thanks,
>>>>>>
>>>>>> Josef
>>>>>
>>>>> Looks like the last message I sent didn't make to the list. Here's the
>>>>> link to
>>>>
>>>>> the debug log:
>>>> So unfortunately I don't know how we ended up with duplicate entries in
>>>> the free space cache, but I can make it so we discard the cache if this
>>>> happens.  Please try the patch I just sent to the list
>>>>
>>>> [PATCH] Btrfs: check for duplicate entries in the free space cache
>>>>
>>>> this will make your fs able to be mounted at the very least.  I'll try
>>>> and figure out how this sort of thing happens.  If you manage to make it
>>>> happen on purpose let me know how you did it so I can figure out what
>>>> I'm doing wrong.  Thanks,
>>>
>>> I was able to mount it readonly and copy all the content from it without
>>> problems (with vanilla linus' tree). The only out of ordinary thing I did
>>> before the error coming up, was to defrag the filesystem (find ~/ -xdev
>>> -type f -print0 | xargs -0 sudo ./btrfs filesystem defragment -c) with
>>> btrfs-tools from the latest git. After this I think I rebooted once the
>>> same day and everything was working ok. The day after in the morning I
>>> powered up the laptop and I wasn't able to mount the home volume.
>>> I'm on a 64 bit Debian SID with custom kernel. Using btrfs on home and
>>> root and ext4 in 3 or 4 other FS, everything is on lvm in luks.
>>>
>>> The only thing I think of I can try is to recreate the filesystem, copy
>>> the same data on it, defrag and try to umount/remount several times...
>>> (although the data will not be fragmented so we may not hit this error)
>>> What do you say?
>>
>> Sorry I'm having problems following what you are saying.  You mean this
>> is how you got into this current situation, or that this is what is
>> currently happening to you.  If it's the first one then cool, if you can
>> try and make it happen again I would be grateful.  If it's the second
>> one then my patch didn't work and I need to try and figure out whats
>> going wrong.  Thanks,
>>
>> Josef
>
>
> Sorry yes, I meant this is how I managed to get the corrupted filesystem.
>
> Ill try to break it again.

Oh ok perfect, yeah I will try and do the same sort of things and see if 
I can get it to happen as well.  Thanks,

Josef