Date: Wed, 8 Dec 1999 11:20:03 -0800 (PST) From: Bosko Milekic <bmilekic@dsuper.net> To: freebsd-bugs@FreeBSD.org Subject: Re: kern/10872: Panic in sorecieve() due to NULL mbuf pointer Message-ID: <199912081920.LAA66328@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: Bosko Milekic <bmilekic@dsuper.net> To: freebsd-gnats-submit@freebsd.org Cc: Subject: Re: kern/10872: Panic in sorecieve() due to NULL mbuf pointer Date: Wed, 8 Dec 1999 14:18:19 -0500 (EST) (C.C.ed to Bob already) !>Hi Bosko, !> !>I must confess that I wound up working around the problem by running !>2.2.8 on my amanda backup server. This, of course, was only a temporary !>solution and I planned to get back to working on the problem, but I just !>never have gotten to it. I would, however, be happy to help on solving !>this problem in any way that I can. Though I haven't tried to recreate !>the problem for along time, I have a couple of systems here (one on !>4.0-CURRENT and another on 3.3-STABLE) that I could attempt it on. !> !>Bob !> What would be interesting to find out is if you can still reproduce it on -CURRENT and 3.3-STABLE. This time, I would ask for you to try printing out the value of the mmbfree mbuf pointer as well as the mclfree mcluster pointer from the debugger. Also, if you ever end up crashing somewhere else (besides for in soreceive or sbdrop), e.g. somewhere directly from the if_fxp code, or if something has significantly changed from the first two traces that you provided, if you could get a new trace and post that along as well. This problem could be related to either a bad cast somewhere along the line which can ultimately generate a 'garbage pointer' no longer referencing the mbuf or mbuf cluster free list, but something undefined, or, even bad offsetting into an mbuf cluster (probably also due to a bad cast somewhere along the line) which _may_ lead to the free mbuf or mbuf cluster pointer being NULL. If something gets screwed up along the way like this, chances are, it'll only be noticed after the following mcluster or mbuf (whichever it is that was screwed up) gets allocated -- it could potentially be pointing to something that's not within mb_map at all! --Bosko -- Bosko Milekic <bmilekic@dsuper.net> 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?199912081920.LAA66328>