Writeback is still screwed in 3.7.1

Writeback is still screwed in 3.7.1

[Jan Engelhardt]

I thought these were fixed, but I just got one on 3.7.1 on x86_64.

[648141.344931] INFO: task sync:10395 blocked for more than 480 seconds.
[648141.344935] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[648141.344937] sync            D ffff88074c0bc760     0 10395   9926 0x00000000
[648141.344942]  ffff880813dbbdc8 0000000000000086 ffff88074c0bc380 ffff880813dbbfd8
[648141.344947]  ffff880813dbbfd8 0000000000012ac0 ffff880817d30340 ffff88074c0bc380
[648141.344951]  ffff8802449d46c8 7fffffffffffffff 7fffffffffffffff ffff880813dbbeb0
[648141.344955] Call Trace:
[648141.344969]  [<ffffffff813ebc8b>] schedule_timeout+0x1c/0xf7
[648141.344975]  [<ffffffff813ec714>] wait_for_common+0x9a/0x115
[648141.344982]  [<ffffffff8112ee60>] sync_inodes_sb+0xaa/0x18c
[648141.344988]  [<ffffffff811125ee>] iterate_supers+0x63/0xb4
[648141.344993]  [<ffffffff81131fdc>] sys_sync+0x2e/0x81
[648141.344999]  [<ffffffff813f39ed>] system_call_fastpath+0x1a/0x1f
[648141.345025]  [<00007f85ffdef117>] 0x7f85ffdef116

How can I check how much there is left to sync?
/proc/meminfo says all has been written out:

Dirty:               424 kB
Writeback:             0 kB

So if not that, what really is blocking processes?

Re: Writeback is still screwed in 3.7.1

[Sedat Dilek]

On Mon, Jan 14, 2013 at 1:10 AM, Jan Engelhardt <jengelh@inai.de> wrote:
>
> I thought these were fixed, but I just got one on 3.7.1 on x86_64.
>
> [648141.344931] INFO: task sync:10395 blocked for more than 480 seconds.
> [648141.344935] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> [648141.344937] sync            D ffff88074c0bc760     0 10395   9926 0x00000000
> [648141.344942]  ffff880813dbbdc8 0000000000000086 ffff88074c0bc380 ffff880813dbbfd8
> [648141.344947]  ffff880813dbbfd8 0000000000012ac0 ffff880817d30340 ffff88074c0bc380
> [648141.344951]  ffff8802449d46c8 7fffffffffffffff 7fffffffffffffff ffff880813dbbeb0
> [648141.344955] Call Trace:
> [648141.344969]  [<ffffffff813ebc8b>] schedule_timeout+0x1c/0xf7
> [648141.344975]  [<ffffffff813ec714>] wait_for_common+0x9a/0x115
> [648141.344982]  [<ffffffff8112ee60>] sync_inodes_sb+0xaa/0x18c
> [648141.344988]  [<ffffffff811125ee>] iterate_supers+0x63/0xb4
> [648141.344993]  [<ffffffff81131fdc>] sys_sync+0x2e/0x81
> [648141.344999]  [<ffffffff813f39ed>] system_call_fastpath+0x1a/0x1f
> [648141.345025]  [<00007f85ffdef117>] 0x7f85ffdef116
>
> How can I check how much there is left to sync?
> /proc/meminfo says all has been written out:
>
> Dirty:               424 kB
> Writeback:             0 kB
>
> So if not that, what really is blocking processes?

Hi Jan,

Can you test w/ Linux v3.7.2?

Not sure but this sounds like a fix for your problem:

commit 9359dd031b11100a3532ef7bfc04f2fc7f23c33b
"mm: fix calculation of dirtyable memory"

It is also good to have a look at queued-up patches for -stable
kernels (in [2] for Linux v3.7.y).
( Just a general recommendation ).

Hope this helps you!

Regards,
- Sedat -

[1] http://git.kernel.org/?p=linux/kernel/git/stable/linux-stable.git;a=commitdiff;h=9359dd031b11100a3532ef7bfc04f2fc7f23c33b
[2] http://git.kernel.org/?p=linux/kernel/git/stable/stable-queue.git;a=blob;f=queue-3.7/series

> --
> To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Writeback is still screwed in 3.7.1

[Jan Engelhardt]

On Monday 2013-01-14 09:50, Sedat Dilek wrote:
>On Mon, Jan 14, 2013 at 1:10 AM, Jan Engelhardt <jengelh@inai.de> wrote:
>>
>> I thought these were fixed, but I just got one on 3.7.1 on x86_64.
>>
>> [648141.344931] INFO: task sync:10395 blocked for more than 480 seconds.
>> [648141.344969]  [<ffffffff813ebc8b>] schedule_timeout+0x1c/0xf7
>> [648141.344975]  [<ffffffff813ec714>] wait_for_common+0x9a/0x115
>> [648141.344982]  [<ffffffff8112ee60>] sync_inodes_sb+0xaa/0x18c
>> [648141.344988]  [<ffffffff811125ee>] iterate_supers+0x63/0xb4
>> [648141.344993]  [<ffffffff81131fdc>] sys_sync+0x2e/0x81
>> [648141.344999]  [<ffffffff813f39ed>] system_call_fastpath+0x1a/0x1f
>> [648141.345025]  [<00007f85ffdef117>] 0x7f85ffdef116
>>
>Can you test w/ Linux v3.7.2?
>
>Not sure but this sounds like a fix for your problem:
>commit 9359dd031b11100a3532ef7bfc04f2fc7f23c33b
>"mm: fix calculation of dirtyable memory"

It seems to be resolved in 3.7.2 and subsequent releases, thank you.