[mrtg] Re: Config File Size

Greg.Volk at edwardjones.com Greg.Volk at edwardjones.com
Tue Jun 18 17:15:14 MEST 2002


> 
> Hi,
> 
> This might have been answered previously, but a quick search through the
> archives did not yield the results I wanted.  Anyway, is there a practical
> limit on the number of interfaces to monitor in a single config file?  For
> instance, are there significant performance drop-offs if one config file
> contains several thousand interfaces?  I believe that the data collection
> probably isn't affected too much, but would the drawing of the graphs and
> pages be affected by having to parse such a large config file?  Do people
> have experience with how large they let their config files grow before
> splitting them into different files?  I guess I'm looking for a trade-off
> between many small files with many processes and larger files with fewer
> processes.  I'm already using rrdtool to elimininate the time spent creating
> graphs and such, so I guess I'm looking for a way to optimize my current
> settings.  Any help that can be provided would be appreciated.
> 

For my configuration, I don't like to run with more than 300 targets 
per config file. I do this for several reasons:

Polling robustness - if you set the snmp timeout at 1 sec, you
should be safe with a five minute poll interval if a bunch of your 
300 targets fail to respond.

14all performance - 14all.cgi performance seems to suffer if the 
config files are too big. What is too big? I'm not sure, but as a
general rule I don't like to see configs that are more than 256kbytes.
One of my 299 target config files is 240kbytes.


--
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



More information about the mrtg mailing list