Date: Fri, 3 Aug 2007 00:34:51 +1000 (EST) From: Ian Smith <smithi@nimnet.asn.au> To: Zbigniew Szalbot <zbigniew@szalbot.homedns.org> Cc: freebsd-questions@freebsd.org Subject: Re: logging system load Message-ID: <Pine.BSF.3.96.1070803003111.18907C-100000@gaia.nimnet.asn.au> In-Reply-To: <20070802120018.9322D16A4DC@hub.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 2 Aug 2007 12:52:20 +0200 Zbigniew Szalbot <zbigniew@szalbot.homedns.org> wrote:
> On Thu, 2 Aug 2007 13:44:33 +0300, Nikos Vassiliadis <nvass@teledomenet.gr>
> wrote:
> > On Wednesday 25 July 2007 20:50, Momchil Ivanov wrote:
> >> На Wednesday 25 July 2007 19:38:41 Zbigniew Szalbot написа:
> >> > Dear all,
> >> >
> >> > Is there a tool similar to top which would measure system load and
> >> > write it to a file that could later be analyzed? The time when my
> >> > system is most loaded happens between 3 and 5 a.m. so a trace of the
> >> > system load would be a wonderful thing to have. I need it to tailor
> >> > some of the jobs accordingly. Any advice?
> >> >
> >> > Thanks in advance!
> >>
> >> You can make a cronjob doing "uptime >> /path/to/logfile" every minute
> >
> > Or perhaps "sysctl -n vm.loadavg" instead of uptime,
> > which is the same information, but requires less
> > scrubbing.
>
> Thanks but that wouldn't record the time, would it? With uptime it is nice
> to have the current time also recorded and I can compare logs to load by
> time.
paqi% /bin/echo `/bin/date` `/sbin/sysctl -n vm.loadavg`
Fri Aug 3 00:33:13 EST 2007 { 0.04 0.11 0.09 }
Cheers, Ian
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.1070803003111.18907C-100000>