Date: Tue, 24 Oct 1995 23:16:53 -0700 From: David Greenman <davidg@Root.COM> To: Brian Tao <taob@io.org> Cc: FREEBSD-HACKERS-L <freebsd-hackers@freebsd.org> Subject: Re: panic: free vnode isn't Message-ID: <199510250616.XAA28141@corbin.Root.COM> In-Reply-To: Your message of "Wed, 25 Oct 95 02:06:05 EDT." <Pine.BSF.3.91.951025015909.277B-100000@flinch.io.org>
next in thread | previous in thread | raw e-mail | index | archive | help
>On Tue, 24 Oct 1995, David Greenman wrote:
>>
>> >panic: free vnode isn't
>> >Debugger("panic")
>> >Stopped at _Debugger+0x2b [/sys/compile/FTPSERV/:261]: movb $0,_inDebugger 110
>> ^^^^^^^^^^^^^^^^^^^^^^^^^^^
>>
>> Where did that come from? Did you build your kernel with -g or something?
>> I wouldn't recommend doing that and it's not needed for DDB.
>
> I used "config -g FTPSERV" and then a normal "make depend &&
>make". The regular, non-debugging kernel was just a "config FTPSERV".
>The above three lines were on ttyv0 in white-on-red text (I'm using
>the pcvt console driver) with the ddb> prompt waiting below it.
Okay...but for DDB this really isn't necessary and only results in you
losing about 4-6MB of memory for all the unused gdb debugging symbols. In
fact, this in itself might cause you problems.
>> Are your using msdosfs on this machine?
>
> Nope. I switched back to the non-debugging kernel today and it
>only ran for about a day before it locked up. Again, no syslog
>messages or any indication that something went wrong (except that
>everything is frozen). We left the office at 9:20pm for dinner, and
>it died at 9:30pm (figures).
>
> Is there any other information I should be posting? It runs stock
>irc 2.8.21 (around 140 users connected to it at peak) and stock
>wu-ftpd 2.4 (around 40 users connected to it at peak). Could that be
>too much of a load on the server? The FTP server is chroot'd to a
>local directory, but anything beneath ~ftp/pub is NFS-mounted. All
>user home directories are also NFS-mounted.
Well, we run a load about ten times that on wcarchive and have failures
about once a week that are caused by two known bugs (one disk controller
related and the other a VM system problem). I've recently added a bunch of
extra debugging code to the kernel in an attempt to isolate the latter
problem...but I'm still waiting for the next crash.
I think your problem is caused by bug(s) in the NFS client code. While we
do test it relatively thoroughly, we don't pond on it in the way that you
are doing and that likely explains why the bug hasn't been seen before and
fixed. :-) Debugging this problem will almost certainly be quite difficult. It
will likely require careful analysis of a variety of data structures,
following pointers, etc, and perhaps even analyzing the NFS traffic just before
the hang/crash. It's probably caused by some kind of failure condition which
isn't properly handled in the NFS client code. Unfortunately, without a lot
more information, it's unlikely that I'll be able solve this problem.
> Should I upgrade to the latest snapshot, or fall back to 2.0.5?
That's always a good idea, perhaps SUPing -stable might be easier.
Unfortunately, we haven't fixed any problems that would directly fix the
"free vnode isn't" panic, but we have fixed other bugs - so I think you'd
benefit by upgrading.
-DG
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199510250616.XAA28141>