Xref: utzoo comp.sys.att:6288 comp.unix.wizards:15836
Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uflorida!ukma!rutgers!cmcl2!phri!cooper!hak
From: hak@cooper.cooper.EDU (Jeff Hakner )
Newsgroups: comp.sys.att,comp.unix.wizards
Subject: 3B2/400 HD#1 hangup?
Message-ID: <1491@cooper.cooper.EDU>
Date: 29 Apr 89 18:34:14 GMT
Organization: The Cooper Union (NY, NY)
Lines: 27


	We have four 3B2/400s.  Our newest, and most 
heavily loaded unit has been crashing a bit more 
often than I care for.  It seems as if integral hard
disk #1, which contains the /usr and /usr3 file systems,
is becoming not ready.  Shells continue to run, but logins
are not possible.  Any attempt to access /usr or /usr3 files
hangs the process, for example, df shows / and /usr2, which
are on integral HD#0, but hangs after that.  Our devices:
3B2/400 4MB.
Two 72 Meg HDs (internal)
SCSI 60Meg cart. tape
SCSI host adapter, controller, 1 external SCSI HD.
EPORTS 
3BNET
Interestingly, our older, non-SCSI 3B2s, identical in all
other respects, have never had this crash.
	At any rate, my *REAL* question is this:  I know how
to take a crash dump to floppies, but at 4MBytes main store,
this would take 15-45 minutes, which is prohibitive.  
Is there a way to make a crash dump partition on one of the
HDs, and then take the dump on that?
	Any thoughts would be greatly appreciated.
Jeff Hakner
Cooper Union Computer Center
hak@meagan.cooper.edu
hak%meagan.cooper.edu@CMCL2.NYU.EDU