Re: 2.6.36 io bring the system to its knees

[Aidar Kultayev]

http://picasaweb.google.com/aidar.eiei/LinuxIo#5533068249408411698

I will look into latencytop output and will figure out a usage pattern
that is most annoying with regards to IO.
Will try to see what leads to that & if possible to make a screenshot
of what is going on.
The thing is, I don't think the program that captures the screenshots
does it in a meaningful way, because at the moment the system is
brought to its knees, I don't think that this particular program
(KSnapshot) can get away from being affected. I mean it might take a
snapshot which is not representative enough.


thanks, Aidar

On Thu, Oct 28, 2010 at 5:48 PM, Pekka Enberg <penberg@xxxxxxxxxx> wrote:
> On Thu, Oct 28, 2010 at 2:33 PM, Aidar Kultayev <the.aidar@xxxxxxxxx> wrote:
>> if it wasn't picasa, it would have been something else. I mean if I
>> kill picasa ( later on it was done indexing new pics anyway ), it
>> would have been for virtualbox to thrash the io. So, nope, getting rid
>> of picasa doesn't help either. In general the systems responsiveness
>> or sluggishness is dominated by those io operations going on - the DD
>> & CP & probably VBOX issuing whole bunch of its load for IO.
>
> Do you still see high latencies in vfs_lseek() and vfs_fsync()? I'm
> not a VFS expert but looking at your latencytop output, it seems that
> fsync grabs ->i_mutex which blocks vfs_llseek(), for example. I'm not
> sure why that causes high latencies though it's a mutex we're holding.
>
> On Thu, Oct 28, 2010 at 2:33 PM, Aidar Kultayev <the.aidar@xxxxxxxxx> wrote:
>> Another way I see these delays, is when I leave system overnight, with
>> ktorrent & juk(stopped) in the background. It takes some time for
>> WM(kwin) to work out ALT+TAB the very next morning. But this might be
>> because the WM(kwin & its code) has been swapped out, because of long
>> period of not using it.
>
> Yeah, that's probably paging overhead.
>
> P.S. Can you please upload latencytop output somewhere and post an URL
> to it so other people can also see it?
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/