Date: Thu, 14 Sep 1995 14:53:08 +0200 From: Andreas Klemm <andreas@sunny.wup.de> To: larry@seminole.iag.net Cc: jhk@freebsd.org, hackers@freebsd.org, current@freebsd.org Subject: SCSI problems with FreeBSD 2.0.5 using AHA 2940 and Quantum Grand Prix Message-ID: <9509141253.AA10430@sunny.wup.de>
next in thread | raw e-mail | index | archive | help
----------
X-Sun-Data-Type: text
X-Sun-Data-Description: text
X-Sun-Data-Name: text
X-Sun-Charset: us-ascii
X-Sun-Content-Lines: 100
Subject: Re: SCSI errors 2.0.5 FreeBSD "sd0 timed out"
Newsgroups: comp.periphs.scsi,comp.unix.bsd.freebsd.misc
In article <42o73m$6vr@news.iag.net> you wrote:
: I have an Opti based Pentium 100 Mhz motherboard with 32
: megs of RAM. The PCI controller is a 2940 and the hard
: drive is a Quantum XP series 2 gig drive. The OS is Free
: BSD 2.0.5.
: With heavy DiskIO I see:
: ahc0: target 0, lun 0, (sd0) timed out
This happens to me, too, sometimes.
: on the console. all diskIO is stopped, and the LED on
: the SCSI host adapter and hard drive are on solid
Yes !
: this happens with heavy disk IO with the configuration
: supporting 10 mhz transfers rates and sync transfers
: enabled
My scenario: Asus P/I 55TP4XE board, P90, 256k synchr. cache, AHA 2940
BIOS V1.16, Toshiba 3601 CD-Rom, Quantum Grand Prix 4GB 7200U/min.
Parity enabled on all devices, as well as 10MB synchr.
My problems started with Termination pronlems. Here how I solved
it, but problems remain ...
The system locked up more frequently under FreeBSD and Linux, when I had
the following order on the scsi bus:
2940-----------Hd-------------CD-Rom
Termin. X X
Putting the harddisk onto the end of the scsi bus I have much
less problems:
2940-----------CD-------------HD
Termin. X X
Before that I tried nearly every combination of SCSI - Termination ;-)
But still somethimes seems to break ...
I had 4 OS on the Harddisk
Slice 1 primary DOS - 250 MB DOS
Slice 2 other - 3 GB FreeBSD
Slice 3 Ext. Part - 500 MB Win NT 351 AS
Slice 4 " - 400 MB Linux Slackware 2.3 - accessable via boot disk
The FreeBSD bootmanager is _the_ boot manager on my system.
So I have to take care to use this order during installation:
DOS, FreeBSD partitioning, Windows NT, Linux, FreeBSD installation.
When I had the disk in the middle of the scsi bus, then the Linux distrib
with 1.2.12 kernel fucked up after installing about 75% of the packages.
Now it works nearly flawlessly.
But then after a few days again some drawbacks ...
The Extended partition couldn't be accessed any more. The first warnings
about a bad extended partition I received during the FreeBSD startup.
After that I installed everything new ... But I left out Linux.
So I have now DOS,NT,FreeBSD, to speed up installing everything again...
But working for a while I get strange results again....
Under FreeBSD I noticed in /usr 1 dir and one file with control characters.
When trying to boot WinNT I can't boot NT from one day to another, because
ntdetect was screwed up ... I should install this file new ...
Is the hardware so weak, that it can't drive 10MB/sec synchronously without
loosing data ????? Or is it due to driver problems ????
The only OS, where I have no crashes in the moment is DOS/Windows 3.11.
BTW: Everything is terminated properly, and I enabled Parity checking
on all devices !!
Hope, that -stable will help me out of this situation, if it should be
a driver problem.
BTW: I made a copy for thr freebsd mailinglists ... Since I'm not reading
these lists currently, please write a cc: to my e-mail
address, too
Best regards
Andreas ///
--
andreas@wup.de /\/\___ Wiechers & Partner Datentechnik GmbH
Andreas Klemm ___/\/\/ - Support Unix -
----------
X-Sun-Data-Type: default
X-Sun-Data-Description: default
X-Sun-Data-Name: .signature
X-Sun-Charset: us-ascii
X-Sun-Content-Lines: 2
andreas@wup.de /\/\___ Wiechers & Partner Datentechnik GmbH
Andreas Klemm ___/\/\/ - Support Unix -
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9509141253.AA10430>