[prev in list] [next in list] [prev in thread] [next in thread] 

List:       freebsd-hackers
Subject:    ncr timeouts
From:       Christoph Kukulies <kuku () gilberto ! physik ! RWTH-Aachen ! DE>
Date:       1997-05-31 9:43:00
[Download RAW message or body]


cvsup.de.freebsd.org (blues.physik.rwth-aachen.de) had some problems
over the past which I'm hoping to get cured. Uptime wasn't very high
(a couple of days only) and I don't know which circumstance
causes it to get amnesic by ncr timeouts.

It has two ncr PCI controllers (cheapo) and today I swapped them against
SC200 (ASUS). Still not sure if this will cure it. The symptom is as
follows:

ncr1: SCSI phase error fixup: CCB already queued (0xf0d55200)
ncr1: timeout f0d55200 (skip)
ncr1: timeout f0d55200 (skip)
ncr1: timeout f0d55200 (skip)
ncr1: timeout f0d55200 (skip)
ncr1: timeout f0d55200 (skip)
ncr1: timeout f0d55200 (skip)

The system is dead in this situation, i.e. cannot access it's  disks.

The two ncr controllers form a CCD devices (two quantum 3 GB disks = 6GB)
This ccd device is NFS exported to ftp.de.freebsd.org and I suspect
that it happens when ftp.de.freebsd.org runs it's nightly ls -lR >ls-lR
file which might also be a NFS stress test and maybe it's the
combination of NFS, CCD and the two controllers. Or might it be just a
hard drive problem? Well, the hardware was brand new when I bought it
5 months ago. So at present I put my hope in the maybe somewhat
better (ASUS) controllers. But maybe Stefan could comment this
also.

--
Chris Christoph P. U. Kukulies kuku@gil.physik.rwth-aachen.de

[prev in list] [next in list] [prev in thread] [next in thread]