[mrtg] Re: reducing interval below 5 minutes

[PAUL WILLIAMSON]

Yeah, make the refresh 300 seconds or more.

Or switch to RRDTool.  MRETG won't (unless this has changed in the 
new version) without some serious hacking, display a graph 
with less than a 5 minute resolution.

If you poll more than every 5 minutes, the resolution of the 
graph would be much more in detail (of course), but the number 
of points in the day (currently 400, or 33 hours, 20 minutes 
worth), week, month, and year would need to be adjusted as well. 
Each of these sets has exactly 400 points:

5 minute for daily
30 minute average for weekly
2 hour average for monthly 
1 day average for yearly

with the possibility of 2 more points representing the high 
points of in and out for each interval if so configured.

By monitoring more often than 5 minutes, what would you 
like to do with that information?  If you went up to, say, 
800 points (to keep close to the same amount of information) 
your log files would effectively double in size.  If you 
stayed with 400 points, you would end up with about half 
the historical information.  Plus, if you go to 800 data 
points, the graphs would be super wide.

Get RRDTool and some other collector (like Cricket, et al)
to finer than 5 minute intervals.  That directive should 
be restricted (and it appears that it is) to intervals 
GREATER than 300 seconds, not less.

This whole thing doesn't even address the fact that once you 
decrease your interval to 2 minutes, you will (at a minimum) 
double the error correction that MRTG has to do because of 
the shorter interval and tighter tolerances for exactly 
120 seconds versus 300 seconds.  Being off by a second 
in a 300 second interval is only half as bad as if you are 
off b a second in a 120 second interval.

Paul
>>> "Carl Friedberg" <friedberg at exs.esb.com> 05/14/02 18:11 PM >>>

Hi,

In a recent "newbie" thread, a user wanted to know how to make mrtg run more frequently than the default 5 minute interval, something I've often wondered as well.

I am running mrtg 2.9.18pre10 (ActiveState Perl 5.6.1, Windows 2000 sp2) now, and am installing pre11. I decided to try these 2 keywords:

Interval: 2
Refresh:120

I reran the indexmaker script (have not tried running this through mrtg yet), and indexmaker complains that

ERROR: CFG Error in "refresh", line 8: 120 should be 300 seconds or more

the line used to run indexmaker was:

perl indexmaker mrtg.cfg --output mrtgindex.html

and indeed, running mrtg gives this error:

ERROR: CFG Error in "refresh", line 8: 120 should be 300 seconds or more

Any suggestions?

Carl Friedberg
friedberg at esb.com

--
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

--
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