From owner-freebsd-hackers Wed Feb 19 17:21:09 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA03862 for hackers-outgoing; Wed, 19 Feb 1997 17:21:09 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id RAA03844 for ; Wed, 19 Feb 1997 17:20:34 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA11479; Wed, 19 Feb 1997 20:20:02 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Wed, 19 Feb 1997 20:20 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id TAA02340; Wed, 19 Feb 1997 19:48:43 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id TAA29513; Wed, 19 Feb 1997 19:53:39 -0500 (EST) Date: Wed, 19 Feb 1997 19:53:39 -0500 (EST) From: Thomas David Rivers Message-Id: <199702200053.TAA29513@lakes.water.net> To: ponds!root.com!dg, ponds!freefall.cdrom.com!freebsd-hackers Subject: Re: another dup alloc panic. Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > >This morning's panic was inode #177593. (Certainly doesn't > >seem to be one that would fall on any inodes/group boundary) > >I'll clear that one and hope for the best. > > > >However, when I did the fsck on that file system; two files were > >associated with that inode - so, it probably was, in fact, already > >allocated... Maybe the allocation structure is sorta messed up and > >it's picking up random bits somewhere... I'm beginning to believe > >some randomness (along the lines of an uninitilized variable) is > >part of the problem here... Although this was personal mail - I'm taking the liberty of reflecting it to freebsd-hackers so it will go into the archives... > > It's really weird that you're seeing these panics so often. Yep - it sure is; especially since I seem to be the only one seeing it. Just F.Y.I. - when I came home this afternoon this particular machine had panic'd again (that's twice in one day) - this time with inode #77681. I seem to have fallen into that state where the panic's just come on-and-on. A machine will run for a month or so, and then suddenly start getting this panics daily (or, as today, even more often.) > Is there > something special you're doing to make those filesystems? Well - on the machine where I reproduce the problem; I'm doing a newfs and a mkdir... nothing too special there. On the "production" machine that's seeing this; it's my news/mail gateway... running an old version of Cnews. > I've not seen > this on wcarchive (19 disk drives, all very busy), and I haven't seen it > on a local news server (which gets a full news feed and delivers news > to dozens of other sites). ...it just seems really strange that this > problem is hitting you so often when I've *never* seen it happen. Yes - that's been the paradox ever since it started.... > Obviously, there is a bug somewhere since phk was able to reproduce > the problem. ...but just out of curiosity, what do you have the DMA > rate set to on your 1542B, and what is your ISA bus speed set to on > your 386 motherboard? Ah - you're assuming I only see this on one machine. The machine that's got the reliable reproduction is a 386 with a 1542B. The machine that's getting the daily panics is a 386 with an IDE controller and a relatively new MAXTOR IDE drive... > It would also be interesting if the problem was > affected by adding more memory. Is it possible to upgrade the machine > to 16MB? Unfortunately; both machines are maxed out (out of 30-pin SIMM slots.) One machine has 8 meg; the other has 12 meg. [Also, remember, my reliable reproduction is caused during the base installation of FreeBSD, so not too much else is going on :-) ] But - not to damper this at all; I'm thrilled to examine just about any avenue you care to consider... and if you have any ideas to try I'd be happy to undertake them!!! _Any_ Ideas :-) > > -DG - Dave Rivers -