Vsphere audit

[acozine]

Eventually this playbook should pull the parameters we care about from vSphere:

• how many VMs are running on each host?
• how much memory is free on each host?
• how much space is available on the various storage nodes?

Right now the playbook is not pulling the right values. Baby steps.

[acozine]

As of 7cb0b71 the playbook succeeds and we get most of the data we want, see this Tower run. Clearly the quickStats.overallMemoryUsage numbers are not the ones we want. Maybe set show_allocated on the VM list, and then total up the allocated memory for all the VMs on each host? Similar to what we're doing now for the number of VMs.

[acozine]

From the Ansible community - try converting MB to bytes and then using human_readable like this:
(item | int * 1024 * 1024) | human_readable.

[acozine]

Okay, a few math struggles later and this works well. There's a rounding-error difference between the memory usage stats in vSphere itself and the ones shown by the playbook, but it is certainly close enough.

Do we want to order the storage info from most-available to least-available? Or is alphabetical by storage node okay?

Also, how do we add a reminder not to use the 15TB storage nodes?

[acozine]

As of dd4de74 the playbook gets the data we need. Remaining work: present space available only for general purpose storage nodes, not for all storage nodes.