Re: 2.4.22pre8 hangs too (Re: 2.4.21-jam1 solid hangs)

[Tejun Huh]

On Thu, Aug 28, 2003 at 08:40:00AM +0300, Ville Herva wrote:
> On Thu, Aug 28, 2003 at 10:13:41AM +0900, you [TeJun Huh] wrote:
> > 
> >  Your problem sounds very simlar to the problem we were suffering.
> > The problem was a spinlock deadlock inside drivers/char/random.c which
> > is used by tcp to generate random initial sequence number.  The bug
> > fix was checked into 2.4 tree on 28th July after the release of pre8
> > at 14th July.
> 
> Uhh, I tried 2.4.22pre8 a while ago (I think it was Herbert P?tzl's
> suggestion), and it locked up too. Shame that the fix didn't make it in
> it...
> 
> I'll give .22-final a spin.
>  
> > This problem can happen on UP machine if the kernel is compiled with
> > CONFIG_SMP.
> 
> This is UP box and the kernel is _not_ compiled with CONFIG_SMP.

 Then, it should be a different problem.  That deadlock wouldn't occur
with UP kernel.

> > Because the offending routine is called only every five
> > minutes and it should receive a SYN packet while it's connecting, it
> > occurs rarely, but it happens when it happens.
> 
> In my case, the lock up seems clearly related to disk io: it usually happens
> during the nightly oracle backup dump, and at some point it kept happening
> while compiling kernel. (It's random, I can no longer reproduce it by just
> compiling a kernel.)
> 
> Do you still think it could be the same one?

 No, I don't think so anymore.  I think trying kdb/kgdb would be
better.

> >  Please try 2.4.22.
> > 
> > P.S. This bug is a real headache.  We had many servers deployed and
> > they all randomly locked up about every two or four weeks.  I believe
> > people should be warned about this one.
> 
> What's really strange is that the box kept running with 2.4.20pre7 for
> almost a year without problems (with the same oracle dump jub in nightly
> cron), and then suddenly begun acting up on my the first day of my summer
> vacatnion...

 Good luck. :-)

-- 
tejun

-
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/