Cloning Alert Definitions and Enabling in VCF Operations Policies

In our last post we discussed adjusting Alert Definition thresholds in VCF Operations Polices, but what if we want to change Alert Definition Wait Cycles? This isn't doable in Policies directly, but rather we have to Clone the existing Alert Definitions, adjust the Wait Cyles, and enable them in our Policy. Here's how.

For reference, all screenshots here are taking from VCF Operations 9. Go to Infrastructure Operations - Configurations - Policies - Policy Definition and click ADD.

Give it a name, description, and have it Inherit from the vSphere Solution's Default Policy. It is a best practice to create your own custom Policy like this, leaving the out-of-the-box vSphere Solution's Default Policy unchanged. Click CREATE POLICY.

We now have our Policy. Going into Alerts and Symptoms and searching for latency related VM Alert Definitions, you'll notice we have two of them.

As indicated, both of these Alert Definitions are just copies of those inherited from the vSphere Solution's Default Policy. Selecting VM experiencing disk read latency you see details.

As discussed in our previous blog, this is where you can make threshold changes that are local to this Policy, but you can't adjust the Wait Cycle on these Alert Definitions. In order to do that, you'll need to clone the existing Alert Definition, make your changes, then active the new Alert Definition in this new Policy.

Go to Infrastructure Operations - Configurations - Alert Definitions and search for the Alert Definition you want to change.

Click the three dots next to the Alert Definition and select Clone.

Give it a name, description, and adjust your Wait Cycle in Advanced Settings.

I wanted this Alert Definition to wait 12 cycles (assuming a 5m collection interval) before generating an Alert. I also adjusted the Cancel Cycle down to 2. Click NEXT.

Here I could adjust any thresholds if I'd like. Click NEXT.

Here we could add Recommendations and/or create new Recommendations if we'd like. Click NEXT.

Step 4 is where you activate this new Alert Definition, in my case I'd like to active it in the new Policy. Click CREATE.

Back in Policies, you can now see our new Alert.

But you'll notice it's still not Active.

In order to activate the Policy it needs a Group or Object assigned to it.

Select Groups and Objects.

I'd like to Activate this new Alert for all VMs, so I've selected the All VMs Custom Group, but you could select whatever you'd like. Click SAVE.

Back in Policies you'll notice our new Policy (including our new Alert Definition) is now Active.

Looking at the VM Alerts related to latency in this Policy we now shows this.

As indicated our new Alert is Activated, but you'll also notice the original VM experiencing disk read latency is also still Activated (Inherited from the vSphere Solution's Default Policy). You'll want to Deactivate this, otherwise you'll get duplicate Alerts.

We have now successfully cloned our original Alert Definition and Activated it on all VMs, all without changing the vSphere Solution's Default Policy. You can do the same for any out-of-the-box Alert Definition you want, enjoy!