[mrtg] Re: MRTG not good for data archive

[PAUL WILLIAMSON]

>>> "Jason Humes" <jhumes at acs.on.ca> 04/01/04 12:09PM >>>
>Hi
>I was just looking into writing a utility for regular MRTG which would allow
>me to enter a date and have mrtg generate a graph for that date using the
data stored in the logs. 
>


I don't thing it's possible with MRTG unless you hack rateup.  If you switch to 
RRD, any of the front ends can do it.  With 14all, just add a png= parameter 
to the end of the call to get just the graph for the period you are looking.

> The problem is that MRTG seems to change the log
>data as time goes on...I mean that the five minute values that mrtg collects
>get processed and re-entered into the log as 30 minute values and then 1
>hour values etc...as time goes on.

Good.  That means MRTG is working the way it should.  It only keeps the largest 
value (peak value) from that specific interval.  This is the strength of MRTG.
The database for a specific target never grows, but maintains a relative 
history of the data being measured.

>  Is there any way to force mrtg to keep
>ALL the five minute values and not recalculate it for the 30 minute, 1 hour,
>etc graphs.  Would switching to RRD or something help with this problem.

Yes.  That and telling cfgmaker (or however you create your RRD files) to 
either extend the existing 5 minute RRD to hold more samples, or create 
another RRA to hold the "extended" 5 minute samples (silly, but could be done).

Paul


--
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