Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 10 Jul 2001 17:56:10 GMT
From:      Larry Rosenman <ler@airmail.net>
To:        Tor.Egge@fast.no
Cc:        neteng@tide.iadfw.net, freebsd-bugs@FreeBSD.ORG
Subject:   re: kern/26224: VFS Panic/SMP/CFLOW(HEAVY network)/Heavy NFS
Message-ID:  <20010710.17561000@ler-freebie.iadfw.net>
In-Reply-To: <200107101714.TAA11209@midten.fast.no>
References:  <200107101320.f6ADK1152861@freefall.freebsd.org> <200107101714.TAA11209@midten.fast.no>

next in thread | previous in thread | raw e-mail | index | archive | help

Thomas Moestl looked at the dump, and found an offending piece of code.

He said he was looking into a fix.

LER

>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<

On 7/10/01, 12:14:14 PM, Tor.Egge@fast.no wrote regarding re: kern/26224=
:=20
VFS Panic/SMP/CFLOW(HEAVY network)/Heavy NFS:


> The panic message and the stack backtrace indicates that a process has=

> gone to sleep while holding a vnode interlock.  When pagedaemon tries
> to obtain the vnode interlock (in vput() called from vm_pageout()) the=

> machine will either hang (if the vnode interlock was held by the other=

> CPU) or panic (if the vnode interlock was held by the same CPU).

> Use
>       ps axlww -N kernel.debug.76 -M vmcore.76

> to list the processes on the system when it died.

> Use gdb to list the kernel stack of each process.  The offending
> process probably has an uncommon WCHAN string.

> - Tor Egge

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?20010710.17561000>