box293 vmware plugin

[WillemDH]

Another small question for you Troy,

Can you confirm that the Datastore_Performance check of your VMware plugin checks the general performance of a specific Datastore? Or will the check only gather the disk performance of the datastore for one esx server? I'm asking this as the --host is a required parameter...

If it would only measure the peformance of one esx server - datastore combiantion, it would mean that to gather the total performance of the datastore, I would have to make a check for each esx server to each datastore, which would be kind of crazy as we have 50 esx servers and 30 datastores...

Please tell me the check measures the complete performance of the datastore.

Grtz

Willem

[Box293]

Currently it is as you describe. The plugin is obtaining performance metrics for that datastore for that specific host. Which would mean lots of service checks for lots of hosts.

I'll do some research and see if there is a way of getting a combined total. One way or another, this really shouldn't be hard to implement.

I'll add this to my to-do list. It's questions and suggestions like these that help me improve the plugin for others

PS I'll also answer your other email soon, I've just been flat out here at the conference this week zzzz

[WillemDH]

Hey,

1000 times thanks man. It just doesn't seem really useful to know the datastore performance load of just one host. And it would be a feature some other paying VMware monitoring solutions don't have (I heared). If you have the chance, maybe you could put your vmware plugin on GitHub? Then I could make a fork, take a look to try implement a solution for one of the issue in my email myself and push it to you. It would be something like 'if guestname is not found, try again with guestname without domain name'.

Grtz

Willem

[Box293]

I'm not yet convinced of putting the plugin on github, maybe one day in the future who knows.

I'll keep you posted on the plugin changes in relation to the datastore and keep you posted. For now I'll close this thread.