Subject: Re: kern/32005: panic: lockmgr: release of unlocked lock! with fssconfig
To: None <kern-bug-people@netbsd.org, gnats-admin@netbsd.org,>
From: Pavel Cahyna <pcah8322@artax.karlin.mff.cuni.cz>
List: netbsd-bugs
Date: 11/05/2005 17:43:02
The following reply was made to PR kern/32005; it has been noted by GNATS.
From: Pavel Cahyna <pcah8322@artax.karlin.mff.cuni.cz>
To: Juergen Hannken-Illjes <hannken@eis.cs.tu-bs.de>
Cc: gnats-bugs@netbsd.org
Subject: Re: kern/32005: panic: lockmgr: release of unlocked lock! with fssconfig
Date: Sat, 5 Nov 2005 18:42:48 +0100
On Sat, Nov 05, 2005 at 06:26:05PM +0100, Juergen Hannken-Illjes wrote:
> - I suppose your program was run as "a.out > /msyncsnap" ?
No, it was run as "a.out > /nic".
> - Do you have a kernel crash dump ?
Yes, unfortunately I'm unable to get backtraces of either fssnap or emacs
processes. "proc 0x<addr>" in gdb does not work.
> - Do you have a backtrace of the "fssconfig" command ?
Unfortunately, not. Any hints about how one could dig it from the core
dump?
> Btw.: What is the purpose of this program ?
I suspected that msync() doesn't sync the metadata (indirect buffers). So
I wanted to make a snapshot immediately after msync() returns and examine
it.
(Why do I suspect this? Because ffs_fsync, besides flushing pages with
VOP_PUTPAGES, explicitely flushes indirect blocks. And msync() does not
seem to do it.)