Subject: panic: lockmgr: locking against myself
To: None <firstname.lastname@example.org>
From: Jukka Salmi <email@example.com>
Date: 05/28/2004 15:09:14
a -current (GENERIC kernel from todays sources) i386 system just crashed
while I was working on it via ssh (I don't have physical access at the
moment). The following was logged by syslog (the system message buffer
didn't contain more information):
May 28 14:16:44 himo syslogd: /dev/console: Too many open files in system
May 28 14:16:44 himo /netbsd: vnode: table is full - increase kern.maxvnodes or NVNODE
May 28 14:24:40 himo syslogd: restart
May 28 14:24:40 himo /netbsd: panic: lockmgr: locking against myself
May 28 14:24:40 himo /netbsd: NetBSD 2.0F (GENERIC) #0: Fri May 28 00:54:45 CEST 2004
May 28 14:24:40 himo /netbsd: firstname.lastname@example.org:/build/nbsdc/usr/src/sys/arch/i386/compile/GENERIC
The machine was building a stable NetBSD release as an unprivileged user
in a chroot environment when it crashed.
Could the vnode problem possibly be related to the panic?
That machine already crashed two days ago with the same lockmgr panic
message. Unfortunately I was running X then and didn't know how to drop
into ddb... Hmm, is it possible to do so at all?
$ ((RANDOM%6)) || rm -rf ~