VM Metric aggregator container

[davidvossel]

/kind enhancement

#1840 Is introducing a VMI aggregated collector into the virt-handler process. I think the aggregated collector belongs in the virt-handler Pod, but it's possible we should consider moving the aggregator to a separate container within that pod. This would isolate the virt-handler VMI launch flow, migrations, shutdowns, etc... from the aggregated metrics collection.

The pros of splitting the aggregator out is that it gives us stronger guarantees that metric collection won't impact the rest of our infrastructure.

The cons are that a new container image adds more complexity. It's yet another thing for us to ship which may or may not offer any tangible benefit.

thoughts?

[ffromani]

I think that the pros in splitting the aggregator in a separate container outweights the costs, and I volunteer for the implementation once PR#1840 is merged.

It is true we add some complexity, but OTOH once we set up properly the pod, the maintenance costs of the separate container are expected to be low (IOW, changes are expected to happen rarely)

[kubevirt-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

/lifecycle stale

[ffromani]

/remove-lifecycle stale

[ffromani]

Do we still want this? I can find some capacity to work here, but I think we will need some design
/cc @rmohr @fabiand

[kubevirt-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

/lifecycle stale

[kubevirt-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

/lifecycle rotten

[kubevirt-bot]

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

/close

[kubevirt-bot]

@kubevirt-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.