Re: [Bugme-new] [Bug 14518] New: I/O appears to get stuck on certain rsync backup job and system clock halts then

[Martin Steigerwald <Martin@lichtvoll.de>]

[-- Attachment #1: Type: Text/Plain, Size: 2037 bytes --]

Am Samstag 31 Oktober 2009 schrieb Andrew Morton:
> (switched to email.  Please respond via emailed reply-to-all, not via
>  the bugzilla web interface).
> 
> On Sat, 31 Oct 2009 06:59:07 GMT bugzilla-daemon@bugzilla.kernel.org 
wrote:
> > http://bugzilla.kernel.org/show_bug.cgi?id=14518
> >
> >            Summary: I/O appears to get stuck on certain rsync backup
> > job and system clock halts then
> >            Product: IO/Storage
> >            Version: 2.5
> >     Kernel Version: 2.6.31.5
> >           Platform: All
> >         OS/Version: Linux
> >               Tree: Mainline
> >             Status: NEW
> >           Severity: normal
> >           Priority: P1
> >          Component: Block Layer
> >         AssignedTo: axboe@kernel.dk
> >         ReportedBy: Martin@Lichtvoll.de
> >         Regression: Yes
> >
> >
> > I sorted this to generic block problems, might be an issue with the
> > sata_sil driver tough.
> 
> Yup, it looks more like an ata issue.
> 
> Or it might be that the errors are due to real hardware problems and
> triggered not-well-tested code paths in block/scsi/VFS/etc.
> 
> Could someone take a look please?

I think I better add one information about a special tuning I applied to 
CFS. CC'ing Ingo Molnar for feedback on that one:

shambhala:~> tail -2 /etc/sysfs.conf
# Scheduler-Tuning
kernel/debug/sched_features = NO_NEW_FAIR_SLEEPERS

I am not sure whether this could interfere with the workload I described.

I believed I had some more settings in /etc/sysfs.conf, but apparently to 
due an editing error on this one has been left.

The idea for is that coming from a scheduler tuning thread that originated 
from the initial Brain Fuck Scheduler posting by Con Koliva.

AFAIK NO_NEW_FAIR_SLEEPERS is disabled on current 2.6.32-rc kernels as 
well.

Ingo might that be a contribution to the issues seen here?

Ciao,
-- 
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de
GPG: 03B0 0D6C 0040 0710 4AFA  B82F 991B EAAC A599 84C7

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]