Re: IO latency - a special case

[Trenton D. Adams]

On Sun, Apr 5, 2009 at 4:42 PM, Trenton D.
Adams<trenton.d.adams@xxxxxxxxx> wrote:
> Hi Ted,
>
> I haven't seen the problem arise for awhile now.  I think it may be
> related to editing a file in one of my editors, and not saving it.
> The editor might be putting some sort of lock on the file.  I'm
> waiting for it to occur again so that I can pin point it.
>
> For now, what about that patch thread sent to LKML called "[tip PATCH
> v7 0/9] RFC: futex: requeue pi implementation"???  Do you think that
> could have fixed my issue?  I don't have time to try it right now.
> And besides, I should try and make it so I can reproduce the problem,
> otherwise I might not know for awhile that it's been fixed, after
> applying the patch.
>
> I'll let you know more, when I have more data.

I'm back into doing some Java development, and it's happening again.
I'm running 2.6.30-rc7, and I did a SysReq+T, I hope this is helpful.
I've attached a txt.gz file, because it's 367K to my
/var/log/messages.

The important stuff, I believe, is the Java related stuff.  As a quick
refresher to this problem, it appears to be related to how my Java IDE
saves the file, and perhaps something doesn't release the futex right
away?

I've also seen these symptoms previously...

1. Problem arises, I Ctrl-C the process when I notice the system
getting really sluggish.
2. Start process again, and it locks up the machine for about an hour

Now, when i try the things below, it seems to be resolvable.

1. Problem arises, I Ctrl-C the process when I notice the system
getting really sluggish.
2. I go to my IDE, press Ctrl-S (save), or exit the IDE
3. Start process again, and it does not lock up

I have a feeling it may be a bug in the Java VM that is causing some
sort of problem with the kernel, but I have no idea how to diagnose
such a thing, given the very much a lack of how the kernel works. :P

Attachment: trace.txt.gz
Description: GNU Zip compressed data