From owner-freebsd-questions@FreeBSD.ORG Mon Nov 15 16:11:25 2004 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0FCB316A4CE for ; Mon, 15 Nov 2004 16:11:25 +0000 (GMT) Received: from mail.broadpark.no (mail.broadpark.no [217.13.4.2]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7DC3843D46 for ; Mon, 15 Nov 2004 16:11:24 +0000 (GMT) (envelope-from henrik.w.lund@broadpark.no) Received: from [10.0.0.3] (52.80-202-129.nextgentel.com [80.202.129.52]) by mail.broadpark.no (Postfix) with ESMTP id 57FD84E79; Mon, 15 Nov 2004 17:11:23 +0100 (MET) Message-ID: <4198D52B.2030307@broadpark.no> Date: Mon, 15 Nov 2004 17:11:23 +0100 From: Henrik W Lund User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.7.3) Gecko/20041111 X-Accept-Language: no, en-us, en MIME-Version: 1.0 To: Subhro References: <4197cc18.3bccd5ef.71c0.001a@smtp.gmail.com> In-Reply-To: <4197cc18.3bccd5ef.71c0.001a@smtp.gmail.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit cc: freebsd-questions@freebsd.org Subject: Re: Panic under heavy HTTP load? X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 15 Nov 2004 16:11:25 -0000 Subhro wrote: > > > > Is the ports tree up to date? IF not please do so and run portupgrade -au. > Your package list makes me feel that the port tree had been updated but only > partially. Also try avoiding fancy stuff like explicitly asking gnutella to > link with gtk2 using build time flags. Also if you are not running really > short of hard disk space, cvsup with ports-all. The ps -aux output had been > snipped in the right. So could not make out a few processes. Last but not > the least, gnutella indeed calculates hashes in order to differentiate > between files having same names and accurately get the correct stuff from > different sources and join them up later. But succesice calls to stat() (or > any syscall as such) should not cause the kernel to freeze. If it indeed > causes a freeze because of that, the kernel is broken and needs to be fixed. > Also are there are cores left over in the filesystem? If yes just check if > there is anything relevant. You can try to post mortem the core files. In > case you are not comfortable with it, you can send them here. Some > knowledgeable soul would surely do it for you. > > Regards > S. > > Subhro Sankha Kar > Block AQ-13/1, Sector V > Salt Lake City > PIN 700091 > India I updated all the ports so that they're the latest version (as of just now), and searched the drives for .core files (there were none). I had it suggested to me that maybe it was something network-related after all due to the hoopla with finegrained locking of the network stack and drivers. My driver (wi) is supposed to run free of Giant, but I'll experiment by toggling debug.mpsafenet and turning on stuff like various WATCHDOGs and WITNESS and such, and see if there is some obscure bug somewhere. Either way, I'll take it off -questions and on to -hackers or even -stable if I need more help. Thanks for all your help so far! :-) -- Henrik W Lund