Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 19 Jul 2004 23:14:44 -0500
From:      Eric Anderson <anderson@centtech.com>
To:        John Birrell <jb@cimlogic.com.au>
Cc:        freebsd-current@freebsd.org
Subject:   Re: HEADS UP [Re: thread+preemption stability improvement]
Message-ID:  <40FC9C34.7030700@centtech.com>
In-Reply-To: <20040720040153.GA21144@freebsd3.cimlogic.com.au>
References:  <Pine.NEB.3.96L.1040719115338.38770K-100000@fledge.watson.org> <40FC3090.9060305@freebsd.org> <40FC92C8.50604@mcsi.pp.ru> <20040720040153.GA21144@freebsd3.cimlogic.com.au>

next in thread | previous in thread | raw e-mail | index | archive | help
John Birrell wrote:

>On Tue, Jul 20, 2004 at 07:34:32AM +0400, Maxim Maximov wrote:
>  
>
>>	I think it is needed :( Things actually get _much_ better, now I've 
>>been able to use my big IMAP folders, but still after applying the patch 
>>and working about an hour mozilla freezes with the same sympthoms (cpu 
>>fan gets spinning faster and faster as if cpu temperature is raising). 
>>Are there any ways I can help track this down further?
>>	Preemption is enabled, of course. My kernel's config is latest 
>>	GENERIC with these additions:
>>    
>>
>
>Even with PREEMPTION disabled, current freezes for me with a threaded application
>with lots of sendfiles in progress. I haven't figured out the set of conditions
>that actually trigger the freeze other than it never freezes if I have no
>libpthread application running (and loaded to some degree).
>
>I wonder if anyone has had current under load and if so, in what situations.
>It looks like this code base is a long way away from being ready for a stable
>branch. Grumble. The gap between 4 and 5 is too big. Not happy. At some point
>people need to concentrate on stability. I would concede API changes for 
>stability any day. Every day.
>  
>

I have used it under load - compiling, NFS serving, SQL serving, etc.  
I'm sure many others have too.  -current has never (that I recall) been 
suggested to be used in production environments, and it is prone to 
having issues like this, which is why it is important that we DO stress 
it, test it, and then try to help debug it by giving the most useful 
information about the problem to those that can help.  If it isn't 
working well for you right now, cvsup back to about 06/01/04 and see if 
it's any better..

Eric


-- 
------------------------------------------------------------------
Eric Anderson     Sr. Systems Administrator    Centaur Technology
Talk sense to a fool and he calls you foolish.
------------------------------------------------------------------



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?40FC9C34.7030700>