Hi there,

Despite setting all my datapoints as DERIVE instead of COUNTER and also an RRD 
Min of 0, I still get a graph where it will spike periodically, similar to the 
picture found in http://forums.zenoss.com/viewtopic.php?t=5145.  

This datasource is a command that I run via ssh on the remote machine, which 
outputs Nagios-style statistics.  The command counts the number of occurrences 
of the string "error" in a logfile.  When the logfile gets rolled, however, the 
command returns a count of 0, which is when this spike occurs.

I read the docs and followed the instructions for avoiding this statistical 
anomaly (DERIVE + RRD Min=0) but it still occurs.  I also tried setting an RRD 
Max of 100000 (as the spike values are always > 1M), but the datapoints still 
get graphed, causing my threshold alert to often be set off inadvertently.  I 
guess a spinoff question is whether the RRDMax applies to the stat returned or 
the processed data point (e.g. the derivative value instead of the original 
value) though this doesn't really matter in this case anyway.

Any help would be much appreciated - thanks!

Clarence




-------------------- m2f --------------------

Read this topic online here:
http://forums.zenoss.com/viewtopic.php?p=32565#32565

-------------------- m2f --------------------



_______________________________________________
zenoss-users mailing list
[email protected]
http://lists.zenoss.org/mailman/listinfo/zenoss-users

Reply via email to