Date: Fri, 28 May 1999 12:22:15 +0930 From: Greg Lehey <grog@lemis.com> To: Rob Secombe <robseco@wizard.teksupport.net.au> Cc: freebsd-isp@FreeBSD.ORG Subject: Re: biodone: buffer already done Message-ID: <19990528122215.Q5509@freebie.lemis.com> In-Reply-To: <3.0.5.32.19990528111057.00b26bc0@moat-gw.teksupport.net.au>; from Rob Secombe on Fri, May 28, 1999 at 11:10:57AM %2B1000 References: <006401bea856$008f55c0$03451acb@teksupport.net.au> <Pine.BSF.3.96.990527164313.18068A-100000@almazs.pacex.net> <3.0.5.32.19990528111057.00b26bc0@moat-gw.teksupport.net.au>
next in thread | previous in thread | raw e-mail | index | archive | help
On Friday, 28 May 1999 at 11:10:57 +1000, Rob Secombe wrote: > At 16:52 27-05-99 -0700, you wrote: >> On Fri, 28 May 1999, Rob Secombe wrote: >>> We just had one of our servers spontaneously reboot. This machine is >>> running FreeBSD 2.1.7 RELEASE and has been running flawlessly for 2 years, >>> up until now. The last entry in the syslog prior to reboot was: >>> >>> /kernel biodone: buffer already done >>> >>> Could one of you kernel gurus please tell me what this means and is there >>> something I can do to prevent it happening again. >> >> I am not a kernel guru but I had exactly the same problem!! >> It happened to me when my SCSI controller card was about to hickup and >> during automatic backup session (high data transfers) my disk would >> freeze-up (red LED on all the time) and I would see the error message: >> >> /kernel biodone: buffer already done >> >> but my machine did not auto reboot. >> The problem is mostlikely in your SCSI host adopter (if you are using one >> and you get a SCSI timeout error message as well) >> Or your disk has some bad sectors and is about to give up >> Consider yourself lucky and do a backup!! >> I had to replace my SCSI controller and the disk as well which was >> degraded by the controller problem. > > Yeah, it would have been backing up SCSI --> SCSI DAT Tape at the time that > it fell over. No sign of media errors in the logs though. Sorry, I just saw this. The message is indicative of a logic problem in the SCSI code which handles exception conditions, and it's probably not the real problem, just a symptom of the problem. Were you able to get a dump? If so, that should help you further. Of course, 2.1.7 isn't exactly the newest version of the system any more, and there's a very good chance that the same behaviour won't occur on 3.2-RELEASE, since the SCSI code has been completely rewritten. I'd be interested to hear of your configuration. Greg -- See complete headers for address, home page and phone numbers finger grog@lemis.com for PGP public key To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-isp" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990528122215.Q5509>