Re: call trace: WARNING: at /build/linux-uwVqDp/linux-4.16.16/fs/btrfs/ctree.h:1565 btrfs_update_device

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

[-- Attachment #1.1: Type: text/plain, Size: 1366 bytes --]



On 2018年06月28日 22:17, Christoph Anton Mitterer wrote:
> On Thu, 2018-06-28 at 22:09 +0800, Qu Wenruo wrote:
>>> [   72.168662] WARNING: CPU: 0 PID: 242 at /build/linux-
>>> uwVqDp/linux-4.16.16/fs/btrfs/ctree.h:1565
>>> btrfs_update_device+0x1b2/0x1c0It
>> looks like it's the old WARN_ON() for unaligned device size.
>> Would you please verify if it is the case?
> 
> # blockdev --getsize64 /dev/sdb2 /dev/sda2
> 999131127296
> 999131127296
> 
> 
> Since getsize64 returns bytes and not sectors, I suppose it would need
> to be aligned to 1024 by the least?
> 
> 999131127296 / 1024 = 975713991,5
> 
> So it's not.

So it's the case.

> 
> 
>> If so, "btrfs rescue fix-device-size" should handle it pretty well.
> 
> I guess this needs to be done with the fs unmounted?

Yep.

> Anything to consider since I have RAID1 (except from running it on both
> devices)?

Nothing special. Btrfs-progs will handle it pretty well.

> 
> 
> Also, it's a bit strange that this error occurred never before (though
> the btrfs-restore manpage says the kernel would check for this since
> 4.11).

Because the WARN_ON() is newly added.

> 
> It would further be nice if btrfs-check would warn about this.

Yep, latest will warn about it, and --repair can also fix it too.

Thanks,
Qu

> 
> 
> Thanks,
> Chris.
> 


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]