*BSD News Article 82394


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!news.mel.connect.com.au!munnari.OZ.AU!news.ecn.uoknor.edu!feed1.news.erols.com!howland.erols.net!news.mathworks.com!fu-berlin.de!uniol!uni-erlangen.de!lrz-muenchen.de!news.informatik.uni-muenchen.de!news.muc.de!news.space.net!news.ms.tlk.com!usenet
From: Andreas Braukmann <braukman@tse-online.de>
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: SCSI: timeout-problems with AHA 2940AU (FBSD 2.1.5)
Date: Wed, 06 Nov 1996 09:19:32 +0100
Organization: TSE TeleService GmbH
Lines: 49
Message-ID: <32804A14.5D30@tse-online.de>
References: <327CD8B0.31BE@tse-online.de> <55m2nl$3fn@anorak.coverform.lan>
NNTP-Posting-Host: 194.97.69.188
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 2.02 (Win95; I)
CC: braukman@tse-online.de

Hello Brian,

Brian Somers wrote:

> : recently I installed FreeBSD 2.1.5 (Walnut Creek CD-ROM) on my new
> : system.
> :     ASUS T2P4,
> :     Intel P5-133,
> :     64 MB RAM,
> :     AHA 2940AU
> :
> :     ahc(0,0,0) Micropolis 4421, 2 GB
> :     ahc(0,1,0) IBM DORS, 2 GB
> :     ahc(0,2,0) IBM DFHS, 4,3 GB
> :     ahc(0,6,0) Toshiba 3401
> :
> :     custom kernel (without AHC_TAGENABLE [misspelled, .. I know])
> :
> : "Stressing" the io-subsystem [ transferring large amounts of data > 
> : from one disk to another; running parallel bonnie-benchmarks on 
> : different disks; copying "/usr"-Filesystem from one disk to 
> : another by
> : 'tar -c ... | (cd ... ; tar -x ...)' ] leads to massive
> : scsi-timeout errors.
> 
> Sounds like dodgy cableing/termination.

... definetely not; I should have mentioned, that this was my "first 
shot", too. I have the habit of changing scsi-cables before submitting 
a scsi-related problem-report. 

I add a few facts:
- Ultra-negotiation disabled (AHA-Bios-Utility)
- active termination (terminator already swapped)
- the error occors only (see above), if the io-subsystem has to do
  "some real work"
For example: 
-  A single "bonnie"-benchmark ( size >> 256 MB) runs like
  a charm. But two or three "bonnies" targeting different disks are
  likely to produce the errors.

But: My regular day to day work (software development; quite huge 
make-jobs) doesn't trigger the error.

TIA,
   Andreas

---
   A. Braukmann - TSE TeleService GmbH - braukman@tse-online.de