On Wed, 14 Jul 2010, Felipe W Damasio wrote:

quoted text
> Hi,
> 
> 2010/7/14 Eric Dumazet <eric.dumazet@gmail.com>:
> >> I can, but my bosses will kick my ass if I bring down the ISP again :)
> >
> > I have no guarantee at all, even if we find the bug.
> 
> Ok :-)
> 
> >> If you think it's the only way to find the problem I'll tell them that
> >> I need to do it. In this case, please tell me what other config
> >> options/tools I can use to get as much info as possible...since I'll
> >> probably be able to test this only once more on the production
> >> environment for debugging purposes.
> >
> > You really should try to setup a lab to trigger the bug, and not doing
> > experiments on production :)
> 
> Right, I'm trying.
> 
> The thing is: The ISP is a 200Mbps network with 10,000 users. The
> first time it took around 2 minutes to trigger the bug. The second
> time it took around 17 minutes.
> 
> So I *think* it's some TCP flag with some weird content...but I can't
> find out what it is so I can trigger it on the lab.
> 
> So my only guess is to enable every possible debug flag I can think of
> to track the bug down on the production environment. Any hints here
> would be appreciated :)

Is it possible for you to mirror the production traffic to another
port, and then do a tcpdump capture to a series of files, so that
you might possibly be able to correlate the kernel crash to the
actual packets on the wire (and the Invalid Request squid errors)?
Just a suggestion.

						-Bill
--
unsubscribe notice
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/