From owner-freebsd-hackers Tue Nov 14 11:03:56 1995 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id LAA03537 for hackers-outgoing; Tue, 14 Nov 1995 11:03:56 -0800 Received: from Root.COM (implode.Root.COM [198.145.90.17]) by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id LAA03532 for ; Tue, 14 Nov 1995 11:03:53 -0800 Received: from corbin.Root.COM (corbin [198.145.90.50]) by Root.COM (8.6.12/8.6.5) with ESMTP id LAA29054; Tue, 14 Nov 1995 11:03:49 -0800 Received: from localhost (localhost [127.0.0.1]) by corbin.Root.COM (8.6.12/8.6.5) with SMTP id LAA00602; Tue, 14 Nov 1995 11:02:22 -0800 Message-Id: <199511141902.LAA00602@corbin.Root.COM> To: Joe Greco cc: hackers@freebsd.org Subject: Re: Odd crash after inode depletion.. In-reply-to: Your message of "Sun, 12 Nov 95 01:40:28 CST." <199511120740.BAA26010@brasil.moneng.mei.com> From: David Greenman Reply-To: davidg@Root.COM Date: Tue, 14 Nov 1995 11:02:21 -0800 Sender: owner-hackers@freebsd.org Precedence: bulk >> > I've got this message. Then system destroys my / partition. I also >> > believe that was'nt a bug. I think it is nice feature added in kernel. ;) >> >> I have a hard time believing that the two are related. I've exhausted >> inodes on our news server quite a few times (and on some of my own >> machines) without any data corruption whatsoever. >> >> Be careful making assertions like this if you're not *really sure* >> that a relationship exists. I do not see any of the investigative >> work required to prove this one way or another given here. > >I've noticed an infrequent but consistent tendency for the system to panic >with "free vnode isn't" some time after hitting inode exhaustion. With a >sample size of only a few instances, all I can establish is that I have NOT >seen a "free vnode isn't" panic without first having run out of inodes. > >(system is running 1026-SNAP, by the way, which may be significant - I never >saw this panic under 2.0.5R, but then again, I can't recall ever having run >out of inodes under 2.0.5R) For those that are seeing this problem ("free vnode isn't"), please add options DIAGNOSTIC ...to your kernel. I need to know if the vnode reference count is going negative. Also, a description of *all* filesystem types that you're using on the system would be useful. It looks like the problem may be correlated to the out-of-inodes condition, but maybe not. Since the problem seems to happen often for some people, perhaps I'll be able to get this one fixed quickly (although too late for 2.1...sorry). BTW, we do plan to provide patches to 2.1 after the release. -DG