balance loop with raid-1 on 5.7.6-1

balance loop with raid-1 on 5.7.6-1

[Russell Coker]

My monthly cron job which runs a btrfs scrub followed by a btrfs balance has 
just gone into an infinite loop.

root@xev:~# grep -c "found 69 extents" /var/log/kern.log
322889
root@xev:~# uname -a
Linux xev 5.7.0-1-amd64 #1 SMP Debian 5.7.6-1 (2020-06-24) x86_64 GNU/Linux

Aug  2 02:57:09 xev kernel: [894810.916299] BTRFS info (device dm-1): scrub: 
finished on devid 3 with status: 0
Aug  2 02:57:09 xev kernel: [894810.947243] BTRFS info (device dm-1): balance: 
start -dusage=60 -musage=10 -susage=10
Aug  2 02:57:09 xev kernel: [894810.948707] BTRFS info (device dm-1): 
relocating block group 2734629453824 flags system|raid1
Aug  2 02:57:09 xev kernel: [894811.034399] BTRFS info (device dm-1): 
relocating block group 2659366862848 flags data|raid1
Aug  2 02:57:15 xev kernel: [894816.467989] BTRFS info (device dm-1): found 
12902 extents, stage: move data extents
Aug  2 02:57:40 xev kernel: [894841.585184] BTRFS info (device dm-1): found 
12902 extents, stage: update data pointers
Aug  2 02:57:41 xev kernel: [894842.878672] BTRFS info (device dm-1): found 69 
extents, stage: update data pointers
Aug  2 02:57:41 xev kernel: [894842.915311] BTRFS info (device dm-1): found 69 
extents, stage: update data pointers
[...]
Aug  2 11:04:01 xev kernel: [924022.697116] BTRFS info (device dm-1): found 69 
extents, stage: update data pointers
Aug  2 11:04:01 xev kernel: [924022.754418] BTRFS info (device dm-1): found 69 
extents, stage: update data pointers
Aug  2 11:04:01 xev kernel: [924022.805762] BTRFS info (device dm-1): found 69 
extents, stage: update data pointers
Aug  2 11:04:01 xev kernel: [924022.870560] BTRFS info (device dm-1): found 69 
extents, stage: update data pointers


-- 
My Main Blog         http://etbe.coker.com.au/
My Documents Blog    http://doc.coker.com.au/

Re: balance loop with raid-1 on 5.7.6-1

[Qu Wenruo]

[-- Attachment #1.1: Type: text/plain, Size: 2036 bytes --]



On 2020/8/2 上午9:07, Russell Coker wrote:
> My monthly cron job which runs a btrfs scrub followed by a btrfs balance has 
> just gone into an infinite loop.
> 
> root@xev:~# grep -c "found 69 extents" /var/log/kern.log
> 322889
> root@xev:~# uname -a
> Linux xev 5.7.0-1-amd64 #1 SMP Debian 5.7.6-1 (2020-06-24) x86_64 GNU/Linux

This is a known bug, and backport for 5.7.x only arrives at 5.7.11.

So please update your kernel to latest 5.7.x stable kernel, or try
v5.8-rc kernels.

Thanks,
Qu
> 
> Aug  2 02:57:09 xev kernel: [894810.916299] BTRFS info (device dm-1): scrub: 
> finished on devid 3 with status: 0
> Aug  2 02:57:09 xev kernel: [894810.947243] BTRFS info (device dm-1): balance: 
> start -dusage=60 -musage=10 -susage=10
> Aug  2 02:57:09 xev kernel: [894810.948707] BTRFS info (device dm-1): 
> relocating block group 2734629453824 flags system|raid1
> Aug  2 02:57:09 xev kernel: [894811.034399] BTRFS info (device dm-1): 
> relocating block group 2659366862848 flags data|raid1
> Aug  2 02:57:15 xev kernel: [894816.467989] BTRFS info (device dm-1): found 
> 12902 extents, stage: move data extents
> Aug  2 02:57:40 xev kernel: [894841.585184] BTRFS info (device dm-1): found 
> 12902 extents, stage: update data pointers
> Aug  2 02:57:41 xev kernel: [894842.878672] BTRFS info (device dm-1): found 69 
> extents, stage: update data pointers
> Aug  2 02:57:41 xev kernel: [894842.915311] BTRFS info (device dm-1): found 69 
> extents, stage: update data pointers
> [...]
> Aug  2 11:04:01 xev kernel: [924022.697116] BTRFS info (device dm-1): found 69 
> extents, stage: update data pointers
> Aug  2 11:04:01 xev kernel: [924022.754418] BTRFS info (device dm-1): found 69 
> extents, stage: update data pointers
> Aug  2 11:04:01 xev kernel: [924022.805762] BTRFS info (device dm-1): found 69 
> extents, stage: update data pointers
> Aug  2 11:04:01 xev kernel: [924022.870560] BTRFS info (device dm-1): found 69 
> extents, stage: update data pointers
> 
> 


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

Re: balance loop with raid-1 on 5.7.6-1

[Russell Coker]

On Sunday, 2 August 2020 11:19:21 AM AEST Qu Wenruo wrote:
> This is a known bug, and backport for 5.7.x only arrives at 5.7.11.
> 
> So please update your kernel to latest 5.7.x stable kernel, or try
> v5.8-rc kernels.

Thanks a lot for your advice and for replying so quickly.  I'll try a newer 
kernel.

Keep up the good work everyone!

-- 
My Main Blog         http://etbe.coker.com.au/
My Documents Blog    http://doc.coker.com.au/