Date: Thu, 25 Nov 1999 23:10:35 +0100 From: Gary Jennejohn <garyj@peedub.muc.de> To: Mark Knight <markk@knigma.org> Cc: freebsd-current@FreeBSD.ORG, freebsd-isdn@FreeBSD.ORG Subject: Re: Panic caused by mbuf exhaustion in i4b with AVM PCI Message-ID: <199911252210.XAA12659@peedub.muc.de> In-Reply-To: Your message of "Thu, 25 Nov 1999 21:29:29 GMT." <KokTwDA5oaP4EwCV@knigma.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Mark Knight writes: >Repeatable mbuf leak leading to crash in current of Saturday 20th >November and all snapshots I've taken prior to that since August (i.e. >when I got the ISDN card). > Are you using the code in -current or one of the releases ? Makes a difference. Can you a) increase the size of the message buffer in your config file (options MSGBUF_SIZE=81920, for example) b) turn on ALL the trace in the kernel with isdndebug c) cause the panic to happen again and get a crash dump ? It would also be good if you could run isdntrace in parallel so that there's some correlation between the kernel messages and the trace times. With the larger message buffer it's possible to see what was happening around the time of the crash (I used this method to debug a panic caused by a stack overflow). There's a way to read the msgbuf from the crash dump, but I can't remember what it is right now :( If all else fails, gdb should be able to do it. I can only guess, but it looks like the user-land process isn't told about the hangup and keeps sending packets down the line. The packets never go out (no connection), so the mbufs eventually run out. The raw interface evidently doesn't have the safety belts that the other interfaces (like ipr, isppp) have. --- Gary Jennejohn / garyj@muc.de garyj@fkr.cpqcorp.net gj@freebsd.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199911252210.XAA12659>