[mrtg] Re: MRTG/RRDTOOL performance hit

[Corey Smith]

What is probably happening is that you are temporarily bringing the load
average of the box to 15 as the 15 jobs kick off simultaneously.  Any
single processor box with a load average of 15 is going to be sluggish
:)

One thing you may want to do is spread your cron jobs over different
time intervals.

IE:

1,6,11,16,21, ...  4 jobs
2,7,12,17,22, ...  4 jobs
3,8,13,18,23, ...  4 jobs
etc...

- or -

Make a master config that includes the other 15 configs and run it with
the fork option as documented in:

http://people.ee.ethz.ch/~oetiker/webtools/mrtg/reference.html#forks_(unix_only)

IE:

Forks: 4

This means you would be at a load average of about 4.0 until the polling
interval is complete.

----

If your cron jobs are taking more than a minute to complete you may want
to consider using a faster processor.  If you see swap utilization
in swapinfo/top you should get more RAM.  Disk is 1000x slower than ram
:)

I'm using 2 - 400 mhz processors and 1GB of ram to monitor 800+ targets
and the box is still very responsive.  No noticeable processing time for
viewing graphs using mrtgrrd & rrdcgi.

Corey Smith

On Tue, 2003-12-02 at 10:11, M.D. DeWar wrote:
> now the box takes along time to respond to me over ssh. Even doing a
> directory change sputters and lags.
> Also when I go to the mrtg-rrdtool cgi it takes a long time to make the
> pages/graphs.

--
Unsubscribe mailto:mrtg-request at list.ee.ethz.ch?subject=unsubscribe
Archive     http://www.ee.ethz.ch/~slist/mrtg
FAQ         http://faq.mrtg.org    Homepage     http://www.mrtg.org
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi