Re: 'btrfs rescue' command (recommended by btrfs check) fails on old BTRFS RAID1 on (currently) openSUSE Leap 15.3

[Qu Wenruo <quwenruo.btrfs@gmx.com>]

On 2022/4/24 17:10, Johannes Kastl wrote:
> Hi Qu,
>
> On 24.04.22 at 01:07 Qu Wenruo wrote:
>
>> No need to run btrfs check on each device.
>>
>> Btrfs check will assemble the array automatically (just like kernel),
>> and check the fs on all involved devices.
>> Thus no need to run the same check on all devices.
>
> OK, good to know. That saves half the time :-)
>
>>> The output of the check is below. The TL;DR was that I should run 'btrfs
>>> rescue fix-device-size' to fix a "minor" issue.
>>>
>>> Unfortunately, running this command fails:
>>>
>>>> root dumbo:/root # btrfs rescue fix-device-size /dev/sdc1
>>>> Unable to find block group for 0
>>>> Unable to find block group for 0
>>>> Unable to find block group for 0
>>
>> This is an unique error message, which can only be triggered when
>> btrfs-progs failed to find a block group with enough free space.
>
> So would resizing the filesystem (to 8GiB) workaround this "limitation",
> so afterwards it could properly fix the device size?

I'm not yet sure if it's a bug in progs causing false ENOSPC, or really
there isn't many space left.

For the former case, no matter how much free space you have, it won't help.

For the latter case, it would definitely help.

Thanks,
Qu

>
>>>> btrfs unable to find ref byte nr 2959295381504 parent 0 root 3  owner
>>>> 1 offset 0
>>>> transaction.c:168: btrfs_commit_transaction: BUG_ON `ret` triggered,
>>>> value -5
>>
>> So at least no damage done to the good and innocent (but a little old)
>> fs.
>
> Puuuh, nice to hear that. :-)
>
>>> So, my question is what I should do:
>>>
>>> Do I need to run another command to fix this issue?
>>
>> Not really.
>>
>> But if you want to really remove the warning, please update btrfs-progs
>> first, to the latest stable version (v5.16.2), and try again.
>
> I'll have a look if I can easily install a newer version of btrfsprogs
> on this machine.
>
>> The involved progs, v4.19 is a little old, and IIRC we had some ENOSPC
>> related fixed in progs, thus if above problem a bug caused false ENOSPC,
>> it should be fixed now.
>
> If I can install a newer version, I'll let you know if the bug disappears.
>
>> You can ignore it for now.
>> It's not a big deal and kernel can handle it without problem.
>
> That's good.
>
>>> Should I copy all of the data to another disk, and create a new BTRFS
>>> RAID1 from scratch? (Which of course I would like to avoid, if
>>> possible...)
>>
>> Definitely no.
>
> Perfect.
>
> Thanks for your reply! Have a nice day.
>
> Kind Regards,
> Johannes
>