Yep, this was a change I made during Dev-Jam.

The primary reasoning behind this was because the minimum and maximum of the 
AVERAGE number does not accurately reflect the minimum and maximum of the real 
value.  One of my big-wigs noticed this first in our bandwidth graphs - we were 
trying to come up with some historic peak bandwidth usage numbers, and they 
noticed that the minimum and maximum values appeared to get skewed toward the 
average the further back you looked.  I believe another user on opennms-discuss 
mentioned this several months ago, noting that they were trying to track the 
maximum number of users logged into a device, but the maximum listed on the 
graph was clearly the maximum of the average, not the true maximum number of 
users.

It was a lot of work to put this change in (and it would probably be even more 
work to take it back out), and I do believe it is rather misleading to list the 
average's min and max on the graph rather than the real min and max, so I, for 
one, would rather not take this change back out.  However, you are correct that 
OpenNMS is currently storing daily MIN and MAX values by default, and this does 
cause unexpected behavior in the graphs.

Perhaps a better solution than backing this change out is to change the default 
RRD collection intervals for MAX and MIN so that they match the collection 
intervals of the AVERAGE values?  This would solve the graphing issues you're 
seeing without printing misleading numbers on the graphs...

-Paul

On Wed, Aug 27, 2008 at 06:17:53AM -0400, Tarus Balog wrote:
> Gang:
> 
> During Dev-Jam I believe it was, the snmp-graph.properties file was  
> changed to use the MIN/MAX values stored in the RRD file instead of  
> the minimum and maximum of the AVERAGE number. It was a lot of work  
> and I can see some reasoning behind it, but I want to proposed that it  
> be backed out.
> 
> By default OpenNMS stores the MIN and MAX daily values. So if I am  
> looking at a subset of the day on a graph, I get the MIN and MAX for  
> the whole day instead of the range I've selected. Try it, zoom in on a  
> 24 hour graph and choose a local maximum and you'll see the value  
> reported is wrong. Plus it takes a day on the new system before those  
> values are other than NaN.
> 
> It will be a lot of work to back out, as it was a lot of work to  
> create, so I don't want to attempt this unless we are in agreement it  
> should be done. Comments?
> 
> -T
> _______________________________________________________________________
> Tarus Balog, OpenNMS Maintainer             Main:   +1 919 533 0160
> The OpenNMS Group, Inc.                     Fax:    +1 503 961 7746
> Email: [EMAIL PROTECTED]                    URL: http://www.opennms.org
> PGP Key Fingerprint: 8945 8521 9771 FEC9 5481  512B FECA 11D2 FD82 B45C
> 
> 
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Please read the OpenNMS Mailing List FAQ:
> http://www.opennms.org/index.php/Mailing_List_FAQ
> 
> opennms-devel mailing list
> 
> To *unsubscribe* or change your subscription options, see the bottom of this 
> page:
> https://lists.sourceforge.net/lists/listinfo/opennms-devel
> 

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-devel mailing list

To *unsubscribe* or change your subscription options, see the bottom of this 
page:
https://lists.sourceforge.net/lists/listinfo/opennms-devel

Reply via email to