Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!think.com!linus!agate!ucbvax!bloom-beacon!bloom-picayune.mit.edu!athena.mit.edu!pschmidt
From: pschmidt@athena.mit.edu (Peter H. Schmidt)
Newsgroups: comp.sys.3b1
Subject: running out of swap space
Summary: hard, but possible to do
Message-ID: <1991Apr11.164132.15894@athena.mit.edu>
Date: 11 Apr 91 16:41:32 GMT
References: <1991Apr8.081311.18416@cbnewse.att.com> <1991Apr9.160921.11987@oswego.oswego.edu>
Sender: Peter H. Schmidt
Followup-To: comp.sys.3b1
Distribution: usa
Organization: Massachusetts Institute of Technology
Lines: 18

In article <1991Apr9.160921.11987@oswego.oswego.edu> ostroff@oswego.Oswego.EDU
(Boyd Ostroff) writes:
>[...] I'm not sure if you really gain anything by [enlarging the swap
>partition] (unless you're swapping out LOTS of processes).  Perhaps someone
>else can answer this one.....

The only way I ever ran out of swap was by running two different, unshared GNU
Emacses, both trying run RMAIL (which is a fair amount of elisp).  Since I had
need to do this then (don't ask why), I increased my swap to 10M, and haven't
run out since, not even w/TeX, gcc, g++, emacs, etc.  I have managed to run
out of *virtual* memory space, though, which is a major bummer.  4M VM is
probably the single biggest uncorrectable limitation of this machine.

REEEEgards -- Peter
--
Peter H. Schmidt	| ...mit-eddie!winter!pschmidt
3 Colonial Village, #10	| winter!pschmidt@mit-eddie.mit.edu
Arlington, MA  02174	| -- Speaking for myself.