Return to BSD News archive
Newsgroups: comp.unix.bsd.netbsd.misc Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!simtel!zombie.ncsc.mil!news.mathworks.com!newsfeed.internetmci.com!news.msfc.nasa.gov!elroy.jpl.nasa.gov!usc!news.cerf.net!daver!dlr From: dlr@daver.bungi.com (Dave Rand) Subject: cannot fork - IAC error? Message-ID: <DH6rMo.144@daver.bungi.com> Organization: Association for the Prevention of Polar bears and Kangaroos Date: Sun, 29 Oct 1995 01:18:23 GMT Lines: 25 I'm running a fairly simple environment. NetBSD-1.0, 16M of memory, 2G of disk, with cnews and nntpd. I've configured MAXUSERS to 20 (340 processes), and have 64M of swap space. Generally, I see a load average of 0.7 to 2.1. I am getting period "cannot fork" messages from the news processes. Doesn't seem to matter which ones - newsrun, incoming batch processing, outgoing batch spooling, etc. When this happens, there are only 70-90 processes 'running' (as in ps aux | wc -l). I can continue to compile, edit, and run various programs under a non-news uid. Lots of space available in swap, according to pstat -T. I did notice, under this environment, that it is a Really Bad idea to set the sticky bit on /bin/sh. Makes you run out of swap space *real* quick. When the system goes down, with shutdown, it complains a lot about lost space in swap area (sorry - exact message lost to the mists of my mind). But this isn't the question... So with regard to this "cannot fork" message - is it the dreaded Idiot At Console (IAC) error? Or have I just set something up wrong? Thanks for your help! -- Dave Rand Internet: dlr@daver.bungi.com