Flushing like mad

Flushing like mad

[Morgan Smith]

Good morning.

I have a machine with a reiser4 formatted drive of redundant data on it.
The data is accessed and modified daily as if it were a production
machine except that nobody cares if the data is lost.

I have noticed these warnings in the daily logs:

Jul 14 06:37:08 logan kernel: reiser4[ktxnmgrd:sdb1:r(31883)]: commit_current_atom (fs/reiser4/txnmgr.c:1049)[nikita-3176]:

Jul 14 06:37:08 logan kernel: WARNING: Flushing like mad: 16384


This is even after a file system check with fsck.reiser4 1.0.8 which
reported 0 issues. I don't seem to have any performance issues or data
corruptions though I'd like to provide what information I can regarding
this if it's of any concern. I believe the drive was formatted with the
following options:

create=reg40,hash=tea_hash,fibration=lexic_fibre,formatting=smart


-- Morgan

Re: Flushing like mad

[Edward Shishkin]

Hello.

Yes, it can happen, especially under high workload.
Also you may observe the warning: "search by key:
too many iterations", which has the similar nature.

If those warnings are sporadic, then simply ignore them.
We are not ready to get rid of them completely for now,
as in some cases it indicates data corruption (in this
case the warnings look like flood).

Thanks for report,
Edward.


On 07/18/2013 08:03 PM, Morgan Smith wrote:
> Good morning.
>
> I have a machine with a reiser4 formatted drive of redundant data on it.
> The data is accessed and modified daily as if it were a production
> machine except that nobody cares if the data is lost.
>
> I have noticed these warnings in the daily logs:
>
> Jul 14 06:37:08 logan kernel: reiser4[ktxnmgrd:sdb1:r(31883)]: commit_current_atom (fs/reiser4/txnmgr.c:1049)[nikita-3176]:
>
> Jul 14 06:37:08 logan kernel: WARNING: Flushing like mad: 16384
>
>
> This is even after a file system check with fsck.reiser4 1.0.8 which
> reported 0 issues. I don't seem to have any performance issues or data
> corruptions though I'd like to provide what information I can regarding
> this if it's of any concern. I believe the drive was formatted with the
> following options:
>
> create=reg40,hash=tea_hash,fibration=lexic_fibre,formatting=smart
>
>
> -- Morgan
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html