I would say your problem is a defective disk.
I have Seagte 1GB disk with AHA-1542 and it works like a charm on an old 486
M/B. The only issue I have is when compiling (OpenBSD 2.7), I have to kill
off as many processes as possible before building a new kernel or compiling
any other software other wise communications with the HDD come to a complete
stop. The network interfaces are still up and running however anything that
wnats to go and access the disk simply hangs.
I believe this is an issue with the C-Compiler than the OS.
Anyway, you might want to low-level format your HDD from the SCSI-HBA
utilities and then re-install your OS, then again if you can get your hands
on a good working SCSI HDD you will probably be OK.
Cheers,
Larry.
----- Original Message -----
From: "Darik Horn" <darik@rocksteady.bluenexus.com>
To: <misc@openbsd.org>
Sent: Tuesday, February 20, 2001 8:23 AM
Subject: Re: 2.8 crashes occasionally
>
> > My experimental OpenBSD firewall occasionally crashes. It reports
> > loads of stuff like 'sd0 (at aha0:0:0) timeout ' until the kernel
>
> I've had problems with the aha driver and/or host adapter too.
>
> My machine has not crashed since I exchanged the SCSI disk for an
> IDE disk. (The hardware was so old that it was not worth worrying
> about.)
>
> []
>
>
> On Mon, Feb 19, 2001 at 07:34:09PM +0000, Jan Muenther wrote:
> > Hello everybody,
> >
> > okay, this is probably hardware-related, but since I'm not 100%
> > sure as to whether that is correct, I might just ask someone
> > else's opinion. Most info is from my (normally leaky) memory so
> > please don't trout-slap me for unprecise descriptions, I've got a
> > bad conscience anyway.
> >
> > My experimental OpenBSD firewall occasionally crashes. It reports
> > loads of stuff like 'sd0 (at aha0:0:0) timeout ' until the kernel
> > panics and the ddb prompt appears. I did a trace and issued a
> > 'boot dump', which seemed to successfully write the dump to disc.
> > There's no relevant entry in /var/log/messages.
> >
> > The _simple_ question is: Am I right in guessing the disk has
> > probably seen better days? Though it doesn't report any errors
> > when I fsck any partition on it, the error messages sound like
> > either the disk (some 1,2GB Seagate SCSI disk) or the SCSI host
> > adaptor (Adaptec 1542) are faulty - or am I mistaken here?
> >
> > I originally planned to use this machine to show my boss BSD's
> > able to turn older machines into actually useful equipment.
> > Should I better drop the box or do you think I encounter a
> > non-hardware problem here?
> >
> > The kernel is pretty much the generic kernel, only difference is
> > I applied ipf 3.4.16 to it - recompiled and booted without any
> > troubles. I can of course provide a dmesg and the dump I made if
> > someone is actually interested... :-%
> > Since the hardware is not exactly exotic and the drivers should
> > be long tested, I'd say the whole thing is due to c***y
> > hardware... or not?
> >
> > Cheers, Jan
> >
> > --
> > Radio HUNDERT,6 Medien GmbH Berlin
> > - EDV -
> > j.muenther@radio.hundert6.de
> >
> >
>
>
>
|