*BSD News Article 94589


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.ecn.uoknor.edu!feed1.news.erols.com!newsfeeds.sol.net!uwm.edu!news.nap.net!news2.nap.net!not-for-mail
From: Paul Mazza <pmazza@naccess.net>
Newsgroups: comp.unix.bsd.bsdi.misc
Subject: SCSI channel lock-up woes
Date: Mon, 28 Apr 1997 08:56:01 -0400
Organization: North American Communications, Inc.
Lines: 48
Message-ID: <33649E61.41C67EA6@naccess.net>
NNTP-Posting-Host: sce.naccess.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 3.0 (X11; I; BSD/OS 2.0 i386)
Xref: euryale.cc.adfa.oz.au comp.unix.bsd.bsdi.misc:6772


I am experiencing a problem when attempting to back up my BSDI 2.1 box.

Here is the configuration:
	BSDI 2.1 OS with all recent patches running on a:
	Pentium P166 w/64MB DRAM and an Adaptec AHA-2940 SCSI cntrllr
	connected to a Seagate Barracuda ST32550N 2GB drive and an HP
	4mm DAT Drive (DDS-II).

Here is the problem:
	The system runs beautifully during normal use.
	When attempting to perform a backup to the DAT drive using the
	dump utility, the system gets anywhere from 18% to 70% complete
	and locks up the SCSI bus.  The disk activity light (wired into
	the Adaptec) stays lit, but there is no activity to either the
	Seagate or the DAT drive.  During this lock-up, the CPU and OS
	are still functioning but have no access to the storage media.
	A hard reset is necessary at this point.

Here is one attempt at a solution:
	In order to isolate the problem to either the Seagate or the
	DAT drive, I removed the DAT drive and attached it to another
	server on the network running BSDI also.
	I then changed the dump script to look like this:
	
		/sbin/dump 0ufB ns2:/dev/nrst0_dds 2000000 /
		/sbin/dump 0ufB ns2:/dev/nrst0_dds 2000000 /usr
		rsh ns2 "/bin/mt -f /dev/nrst0_dds rewind"

	I then began the script and the same thing happened.  In other
	words, the problem was not being caused by the DAT drive.

Please note that in both of the situations listed above, the first dump
was successful (the root fs).  Only during the "usr" dump did the SCSI
bus seize up.  It is worth noting that the root fs is quite small
whereas the "usr" fs is over 1.8 GB.

Also, I have checked bus termination and cabling and everything seems in
order.

Any help would be appreciated.

Thank You,

-- 
S. Paul Mazza, III
North American Communications, Inc
<pmazza@naccess.net>
1-800-326-9559