Date: Thu, 6 Jan 2000 23:30:02 -0800 (PST) From: Terry Kennedy <terry@tmk.com> To: freebsd-bugs@FreeBSD.org Subject: Re: kern/10872: Panic in soreceive() due to NULL mbuf pointer Message-ID: <200001070730.XAA84060@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
The following reply was made to PR kern/10872; it has been noted by GNATS. From: Terry Kennedy <terry@tmk.com> To: Bosko Milekic <bmilekic@dsuper.net> Cc: freebsd-gnats-submit@freebsd.org Subject: Re: kern/10872: Panic in soreceive() due to NULL mbuf pointer Date: Fri, 07 Jan 2000 02:13:35 -0500 (EST) > Do you think you could provide a trace of particularily the crash that > you're getting in tulip_rx_intr() ? This is the first time that I hear of > the Tulip cards puking out in conjunction with the ncr and quite > honestly, I'm really desperate trying to figure out what could cause the > inconsistency between sb_cc and sb_mb. Perhaps this could shed some light > on the issue as it would eliminate the sole blame on the fxp driver. The system hasn't crashed since I rebuilt a kernel with debug symbols (-g) enabled. The box needs to be down for a few hours and then get swamped with packets, and if it's down for that long I get loads of unhappy users, so I can't do much testing. If it does fail again, I will collect crash dumps. I can say that the tulip_rx_intr() crash was in a MCLGET macro, and that a netstat -m on the various crash dumps all showed an apparently impossible mbuf usage - something like "1700/1600 mbufs in use" - however, in other *BSD's, I've found that the assorted *stat programs sometimes get confused between the running kernel values and the ones in the crash dump. Perhaps this is enough of a hint, though. I did a quick peek at the other *BSD's to see if there was anything obvious but the various versions have diverged enough that it wasn't useful. I did find references to sbdrop panics on other *BSD's, but in old versions - the PR's were closed with comments like "this code was changed a lot since the version it was reported against - if it happens on a current kernel, let us know". Terry Kennedy http://www.tmk.com terry@tmk.com Jersey City, NJ USA +1 201 451 4554 (voice) +1 201 451 0900 (FAX) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-bugs" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200001070730.XAA84060>