I would say about 90% of zenoss can be called using xmlrpc and the other
10% has to be called using REST.

I created a script that essentially keeps all of our clusters/projects
in sync with zenoss, (devices have all up2date info,
systems/groups/monitors/console info/ power info)

For instance if cbsnews has 30 devices in it's cluster (equivalent to 30
devices in the zenoss tab under cbsnews) and we added 3 more into the
cluster the syncing script runs in cron every 30 min and checks to see
if the cluster was modified and if it was it will move the new devices
into the right cluster as well as remove devices from other clusters.

This in conjunction with want Kent has being doing with Organized
graphs, we always have all the graphs up2date with the current devices
in the right clusters. With my script and Kent's Organized graphs we
have almost the exact same function as Ganglia (We just can't sort the
graphs by highest load like ganglia can)

That being said we pretty much made ganglia/mrtg/cacti/ and Mon obsolete
in our environment. :) Got to love Zenoss and it's possibilities.

In short there is a lot that can be done using XMLRPC/REST... I wish as
well as Kent does that we can do 100% of it using XMLRPC but have been
unsuccessful (some of the calls return invalid XML format and then
crashes, this is why we have to use rest for some of the calls ).

Great Job Zenoss Team.... Keep it coming!!
_______________________________________________
zenoss-users mailing list
[email protected]
http://lists.zenoss.org/mailman/listinfo/zenoss-users

Reply via email to