Path: usenet.ee.pdx.edu!cs.uoregon.edu!news.uoregon.edu!usenet.eel.ufl.edu!MathWorks.Com!zombie.ncsc.mil!news.duke.edu!news-feed-1.peachnet.edu!gatech!rutgers!utcsri!newsflash.concordia.ca!canopus.cc.umanitoba.ca!cs.umanitoba.ca!gedetil From: gedetil@cs.umanitoba.ca (Gilbert E. Detillieux) Newsgroups: comp.sys.sun.admin Subject: Re: How to test SIMMS? Date: 15 Aug 1994 20:41:48 GMT Organization: Computer Science, University of Manitoba, Winnipeg, Canada Lines: 27 Distribution: world Message-ID: <32ojuc$bjn@canopus.cc.umanitoba.ca> References: <776212393snz@ahardy.demon.co.uk> <1994Aug10.154844.16882@aqm.com> NNTP-Posting-Host: platinum.cs.umanitoba.ca In article <1994Aug10.154844.16882@aqm.com>, melvin@rd.aqm.com (Melvin Evans) writes: > I know of two options... > > 1st choice: as root, run /usr/diag/sundiag/sundiag. It was designed to run > under sunview, and will allow you to test all of your hardware. > > 2nd choice: from the boot monitor prompt ">", in "old-mode", you can run > a test called "test-memory". Apparently, neither of these is very thorough, as they just do read tests. There's a memory testing program available, written by Phil Blanchfield , that does a better job. I don't know if it's available on the net via anon-ftp (try an archie search), but you could get a copy from me. In any case, I've found a program that seems to be even better at exercising memory, and bringing out those intermittent parity errors. I've run the Crack program (a public domain password cracker), using a very large dictionary, and managed to get parity errors out of a flaky SIMM, even when the memtst program didn't discover a problem. Go figure. :) -- Gilbert E. Detillieux Internet: Dept. of Computer Science or University of Manitoba Phone: (204)474-8161 (or Intercom 32) Winnipeg, MB, CANADA R3T 2N2 Fax: (204)269-9178 "The earth is like a tiny grain of sand, only much, much heavier."