Date: Sun, 13 Feb 2005 14:46:13 +0100 From: Willem Jan Withagen <wjw@withagen.nl> To: Harti Brandt <harti@freebsd.org> Cc: Borja Marcos <borjamar@sarenet.es> Subject: Re: Devilator - performance monitoring for FreeBSD Message-ID: <420F5A25.9060603@withagen.nl> In-Reply-To: <20050202124154.Q11339@beagle.kn.op.dlr.de> References: <Pine.NEB.3.96L.1050202113202.68249E-100000@fledge.watson.org> <20050202124154.Q11339@beagle.kn.op.dlr.de>
next in thread | previous in thread | raw e-mail | index | archive | help
Harti Brandt wrote: > On Wed, 2 Feb 2005, Robert Watson wrote: > > RW> > RW>On Wed, 2 Feb 2005, Borja Marcos wrote: > RW> > RW>> I'm not sure about the correct values in the process description > RW>> to get a picture as accurate as possible of the cpu usage of different > RW>> processes. I've seen that top uses p_runtime (FreeBSD 5 and FreeBSD 4), > RW>> but I'm not sure if the value would be really useful. > RW> > RW>This is very cool. :-) How are you currently extracting the information? > RW>One of the things I've wanted to do for a while is make sure all this sort > RW>of thing is exposed via snmpd so that the information can be gathered > RW>easily across a large number of hosts (say, 10,000). > > That could be a nice JUH (junior userspace hacker's) task to add a module > to bsnmp. net-snmp is able to run arbitrary external code to obtain values to be monitored, and it seem to be able to use modules (haven't used them yet). I've been using net-snmp/mrtg already for as long as I can remember to monitor load and diskspace. Processes and other things with MRTG are IMHO sort of troublesome since sample period is 5 minutes. And most processes that outlive that timespan are kernel/daemon processes. What I like about Borja's stuff is that he is able to plot more that just 2 params in 1 graph. --WjW
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?420F5A25.9060603>