Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 18 Oct 2006 16:01:20 -0600
From:      Scott Long <scottl@samsco.org>
To:        Tom Judge <tom@tomjudge.com>
Cc:        Jason Thomson <jason.thomson@mintel.com>, freebsd-stable@freebsd.org, Conrad Burger <conrad.burger@gmail.com>, Kris Kennaway <kris@obsecurity.org>
Subject:   Re: When will the new BCE driver in HEAD be	incorporated	into	RELENG_6?
Message-ID:  <4536A430.4070807@samsco.org>
In-Reply-To: <45367B71.40704@tomjudge.com>
References:  <d0c4d1610610180258l1fb94e96m349022fcb2d0330c@mail.gmail.com>	<45364782.8090209@samsco.org>	<45366122.6090704@mintel.com>	<20061018175257.GA97258@xor.obsecurity.org> <45367B71.40704@tomjudge.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Tom Judge wrote:
> Kris Kennaway wrote:
> 
>> On Wed, Oct 18, 2006 at 06:15:14PM +0100, Jason Thomson wrote:
>>  
>>
>>> Using the driver from HEAD* in the latest RELENG_6 didn't fix our
>>> problems.
>>>
>>> We could still trigger the Watchdog timeout when copying a local file to
>>> an NFS mounted filesystem (UDP mount, GigE speeds).
>>>
>>> It was also possible to trigger this bug with multiple simultaneous TCP
>>> streams,  but that took a little longer.
>>>
>>> Copying a local file to an NFS/UDP filesystem would trigger the bug in a
>>> few seconds.
>>>
>>> If there's anything we can do to help debug this,  please let us know.
>>>     
>>
>>
>> Per my previous mails, the (known) bce watchdogs are symptoms of
>> driver bugs which can be usefully converted into panics by enabling
>> INVARIANTS and INVARIANT_SUPPORT.  Please do so, then report what
>> happens.
>>
>> Kris
>>   
> 
> Hi Kris,
> 
> I am a colleague of Jason's,  when we were testing this patch the kernel 
> used had both options set:
> 
> option INVARIANTS
> option INVARIANT_SUPPORT
> 
> Although on several boxes we have failed to cause a kernel panic, only 
> watchdog timeouts.  However the last crash that we reproduced did 
> trigger several:
> 
> bce: need to defrag
> 
> Messages on the console before the watchdob timeout occured.
> 
> 
> Tom

This is a debugging message that I accidentally left in.  It may relate
to the problem, I'll take a look.

Scott




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4536A430.4070807>