*BSD News Article 50145


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.uwa.edu.au!DIALix!haywire.DIALix.COM!not-for-mail
From: peter@haywire.DIALix.COM (Peter Wemm)
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: Strange Crash
Date: 5 Sep 1995 19:51:08 +0800
Organization: DIALix Services, Perth, Australia.
Lines: 30
Sender: peter@haywire.DIALix.COM
Message-ID: <42hdjc$6uk$1@haywire.DIALix.COM>
References: <DDywqs.Jo9@ritz.mordor.com> <41q60m$p3u@reason.cdrom.com>
NNTP-Posting-Host: peter@haywire.dialix.com

"Jordan K. Hubbard" <jkh@FreeBSD.org> writes:

>ritz@ritz.mordor.com (Chris Mauritz) wrote:
>>Anyone seen this before?
>>
>>Aug 26 15:08:20 sauron /kernel: bt0: Try to abort
>>Aug 26 15:08:39 sauron /kernel: bt0: Try to abort
>>Aug 26 15:08:41 sauron /kernel: bt0: Abort Operation has timed out
>>Aug 26 15:08:42 sauron /kernel: biodone: buffer already done

>Yeah, I have..  I used to have a drive that would, very occasionally, just
>decide that it was having an attack of existential angst and needed to spin
>down for awhile.  When this would happen, FreeBSD would notice its last
>operation go astray, try to abort it, time out on the abort (because the
>device was offline, not stuck) and then proceed to go to pieces over it.

>Bad error checking (or more simply a lack thereof) in the SCSI driver
>is the real culprit.  It's better to have your hardware work, of course,
>but FreeBSD still shouldn't have kittens when it stops in recoverable
>ways..  Bug stompers in the SCSI code are always welcome! :)

Hey!! That's what happened to me!!!  Only problem was that the system
was totally trashed almost beyond recovery as a result...

From my experience, I'd rate the potential seriousness of this bug
at about 18 out of 10...

>						Jordan

-Peter