Date: Wed, 7 Jul 2004 19:40:44 +0200 From: Daniel Lang <dl@leo.org> To: Robert Watson <rwatson@freebsd.org> Cc: current@freebsd.org Subject: Re: panic: m_copym, length > size of mbuf chain Message-ID: <20040707174044.GC45200@atrbg11.informatik.tu-muenchen.de> In-Reply-To: <Pine.NEB.3.96L.1040707122259.37929D-100000@fledge.watson.org> References: <20040707162154.GB45200@atrbg11.informatik.tu-muenchen.de> <Pine.NEB.3.96L.1040707122259.37929D-100000@fledge.watson.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi Robert, Robert Watson wrote on Wed, Jul 07, 2004 at 12:24:59PM -0400: [..] > > As I told Bruce, I have never set debug.mpsafenet (in case you are > > especially interested in this tunable). > > Yeah, that's one of the ones I was particularly interested in :-). > > Just to try ruling out possibilities -- have you run an extensive set of > hardware diagnostics? Most server class hardware ships with a decent > diagnostics disk, and I'm sure we can find some for you in the event your > hardware didn't come with some. While it's quite possibly a software I agree. I have indeed have asked our hardware department to check the memory. They did run memtest86 several days without any error, so I am somewhat convinced, that the memory is intact. It is Dell memory, ordered just for this server, so I assume the specs match the requirements. However, it is obviously possible, that other system components could be faulty. I did follow your advice and downloaded a set of diagnostic utilities for this server from the Dell support page. I will execute them tomorrow, when I'm back at the office. > problem, tracking hardware problems using software symptoms constitutes > undesirable pain and so it wouldn't hurt to give that a spin. I remember > seing your earlier e-mails about running with WITNESS increasing the > chances of pain -- this could be a bug in WITNESS as you suggest, or it > could be that WITNESS increases the opportunities for a variety of locking > related races by increasing the cost of lock/unlock operations. True. And I have already reconsidered my suggestion, since the removal of WITNESS did not result in a stable system. Instead I just encountered tight lockups. The system was up a bit longer, though. Regardless of the diagnostics to be done tomorrow, I will inspect the latest crash, in which I was again lucky to get a crash-dump. I will file a PR with my findings and mail the number to the -current list. It was again inside of WITNESS code, this time a failed assertion. Cheers, Daniel -- IRCnet: Mr-Spock - ceterum censeo Microsoftinem esse delendam - Daniel Lang * dl@leo.org * +49 89 289 18532 * http://www.leo.org/~dl/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040707174044.GC45200>