Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!news.cs.su.oz.au!metro!metro!munnari.OZ.AU!news.ecn.uoknor.edu!news.wildstar.net!serv.hinet.net!news-peer.nctu.edu.tw!news.nctu.edu.tw!spring.edu.tw!news.maxwell.syr.edu!news.mathworks.com!news1.best.com!news.alamar-usa.com!not-for-mail From: Eric Sorenson (Taos Mtn) <eric@alamar-usa.com> Newsgroups: comp.unix.bsd.bsdi.misc Subject: Re: Why is my load so high? Date: 24 Apr 1997 00:30:05 GMT Organization: Alamar Electronics, USA Lines: 23 Message-ID: <5jm9id$1va$1@alamar4.alamar-usa.com> References: <5iek0e$ho$1@madmax.keyway.net> <5ii3to$52n@nuhou.aloha.net> <5ipj46$hvc@pulp.ucs.ualberta.ca> NNTP-Posting-Host: alamar-usa.com X-Disclaimer: Neither Taos nor Alamar speak for me; I return the favour. X-Newsreader: TIN [UNIX 1.3 unoff BETA 970327; i386 BSD/OS 2.1] Xref: euryale.cc.adfa.oz.au comp.unix.bsd.bsdi.misc:6741 Marc Slemko informed comp.unix.bsd.bsdi.misc thusly: > This means that if you have a few dozen idle servers, they will all be > woken up when a connection comes in; all but one will then be put back > to sleep. The cost isn't overly high to do this, but it can falsely > increase the load average to obscene numbers (250 once under FreeBSD > with good interactive response). This isn't really what I've seen -- the behaviour is that the load as reported by 'top' goes up to around 1.0 on a completely idle system (one root console login) and hovers right there indefinitely. A SIGHUP sent to the main http process will bring the average back down slowly until it's at more normal levels, around 0.06 or so. No disk activity, no processes on the cpu (well, except for 'top' when it's actually polling). I've seen this from early versions of Apache and BSDI's from 2.0 through 3.0. *shrug* Doesn't matter too much, I was just scanning this group from work and thought I'd clarify. -- Eric Sorenson / eric at alamar dash usa dot com Taos Mountain @ Alamar Electronics USA, Inc