Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 10 Jan 2002 01:57:49 -0800
From:      "Ted Mittelstaedt" <tedm@toybox.placo.com>
To:        <gnb@itga.com.au>
Cc:        <freebsd-bugs@FreeBSD.ORG>
Subject:   RE: kern/33637: Panic: vm_page_unwire: invalid wire count: 0 
Message-ID:  <002101c199bd$42f97d40$1401a8c0@tedm.placo.com>
In-Reply-To: <200201100339.OAA28995@lightning.itga.com.au>

next in thread | previous in thread | raw e-mail | index | archive | help
>-----Original Message-----
>From: gnb@itga.com.au [mailto:gnb@itga.com.au]
>Sent: Wednesday, January 09, 2002 7:39 PM
>To: Ted Mittelstaedt
>Cc: freebsd-bugs@FreeBSD.ORG
>Subject: Re: kern/33637: Panic: vm_page_unwire: invalid wire count: 0
>
>
>>  Any application is able to cause a kernel panic with just regular library
>>  calls.
>
>With respect to Ted, no non-root userland program should be able to
>panic the
>kernel.  Ever.
>
>Yes, I know about fork-bombs, and malloc-bombs, and that login.conf can stop
>them, but they should still never panic the kernel.  That they do (did?) in
>FreeBSD is still a (hard-to-fix) bug with a well-known workaround, not an
>acceptance of the fact that panics are acceptable.
>

I certainly don't advocate designing an operating system that is easy for
user apps to crash, I hope you didn't get that.  The submitter of the PR
was operating under a false assumption - that he could run any program and
assume that it was impossible for it to crash the OS.  This assumption is
preventing him from considering all the possibilities, and failure to
consider all possibilities and judge them is a fatal impediment to solving
problems.

I'm guessing that if this does turn out to really be a FreeBSD problem
that it's going to be related to the use
of the SMP kernel, and a bug in mmap.  This is why I told him to boot the
server on a uniprocessor kernel and see if the problem went away.  But he
almost immediately made excuses against this, whereupon I have no choice but
to believe that the submitter is more interested in theorizing an answer to
the problem than in actually testing and analyzing the results.  I don't have
time for
that and neither does anyone else.

It appears that in this PR that the submitter did not avail himself of the
Diablo mailing list or support community first - which I feel is the biggest
problem.  If the Diablo user community and developers look at this problem
and all decide that there's a problem with FreeBSD then great - let's look
into it.  But until then this problem could be due to a simple
misconfiguration
of his Diablo software.  News servers are, after all, somewhat tempermental
to begin with.

Ted Mittelstaedt                                       tedm@toybox.placo.com
Author of:                           The FreeBSD Corporate Networker's Guide
Book website:                          http://www.freebsd-corp-net-guide.com



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?002101c199bd$42f97d40$1401a8c0>