On Nov 21, 2008, at 5:36 PM, laguest wrote:
> I have a JMX check setup that pulls down used memory. It works fine
> and shows the value in MB used. I then setup a threshold and added
> this to the graph. When I set the threshold value to 800000000 it
> show in the graph as 800.00 MB and the graph line draws at the 800MB
> section in the graph. I did this just to make sure I understood the
> values and the line did get drawn. I then set it to the real
> threshold I need which is 85% of 2GB so I change the threshold to
> 1740000000 for 1.7GB.
>
> The problem is in the graph it always says 1.0G for the threshold
> set point. I can’t get it to read the actual value.
This major loss of precision is a bug. Fortunately there is a very
easy fix for it. Open $ZENHOME/Products/ZenModel/MinMaxThreshold.py
and go to around line 299 depending on your version of Zenoss. You'll
see a line in the setPower method that looks like the following:
number = number / 1000
Change it to the following instead:
number = number / 1000.0
After you do this you should restart zenhub and zenjmx.
Ticket #4070: Legend on threshold graph points is imprecise
http://dev.zenoss.org/trac/ticket/4070
_______________________________________________
zenoss-users mailing list
[email protected]
http://lists.zenoss.org/mailman/listinfo/zenoss-users
