Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!rutgers!cs.utexas.edu!tut.cis.ohio-state.edu!bloom-beacon!husc6!rice!sun-spots-request
From: apctrc!zjat02@uunet.uu.net (Jon A. Tankersley)
Newsgroups: comp.sys.sun
Subject: Re: swapping over nfs
Keywords: SunOS
Message-ID: <4108@kalliope.rice.edu>
Date: 24 Jun 89 06:11:08 GMT
Sender: usenet@rice.edu
Organization: Sun-Spots
Lines: 14
Approved: Sun-Spots@rice.edu
X-Sun-Spots-Digest: Volume 8, Issue 49, message 2 of 11

One thing I noticed is the fudge factor is kinda large for the swap
partitions (6%).  This is due to the default settings for the partition
with respect to inodes, etc.  Since /export/swap will usually have less
than 50 (or even 20) files and really doesn't need lost+found, wouldn't it
be better to change the newfs options(mkfs options) and trim the fat down
to 1% overhead?  I've been meaning to try this, as I am also running with
111% /export/swap, but haven't tried it yet.

Has anybody figured out why something tries to use /export/swap on the
server for more than nfs swap?  I get 'file system full' messages every
once in a while.  Dunno what vmunix is doing...
-tank-
#include <std/disclaimer.h>		/* nobody knows the trouble I .... */
tank@apctrc.trc.amoco.com    ..!uunet!apctrc!tank