Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!news.rmit.EDU.AU!lynx.aba.net.au!munnari.OZ.AU!spool.mu.edu!usenet.eel.ufl.edu!news.mathworks.com!enews.sgi.com!news.uoregon.edu!waikato!status.gen.nz!pinnacle.co.nz!not-for-mail
From: jonc@pinnacle.co.nz (Jonathan Chen)
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: 2.1.5R and behaviour under pressure.
Date: 4 Sep 1996 09:13:31 +1200
Organization: Pinnacle Software Ltd, Auckland, New Zealand
Lines: 28
Message-ID: <50i71r$t9@tui.pinnacle.co.nz>
NNTP-Posting-Host: tui.pinnacle.co.nz
X-Newsreader: NN version 6.5.0 #3 (NOV)
I'm running 2.1.5R on a measley '486 with 16Mb RAM with the GENERIC
kernel; and I've noticed that `sometimes' when I'm running a series
a compiles (eg building a new kernel), the compiles all start to fail
with bus-errors ("cc1 died with signal 10") if I load up the machine
while the compiles are running.
eg:
compile running on first screen
startx on the second screen
run another major app on 3rd screen
This seemed to happen with 2.1R as well. While I'm not absolutely
sure it's not a h/w error that's causing this, I can say that the
system behaves itself most of the time under light to medium load.
Once the bus-errors start to occur, all further compiles will continue
to fail, even if all the other big programs have ended. The only cure
is a reboot.. <sigh>
I know, I know, this is all probably very vague; and I don't expect
a cure anytime soon. However, I'd just like to poll other readers
whether they've seen this problem as well. Any quick suggestions
before I even *contemplate* the prospect of looking at the code for
the VM system (my first suspect)?
--
-------------------------------------------------------------------
Jonathan Chen <jonc@pinnacle.co.nz> | -insert-witty-quip-here-
-------------------------------------------------------------------