[Request] Health/AA decorator for target nodes

[tthvo]

Describe the feature

Health check decorators for target nodes. This could be the result of automatic analysis report.

Any other information?

We can add it in 2 of the remaining quadrant in target nodes.

[andrewazores]

This is an interesting idea but there are some things to consider.

1. This means that either we must automatically start a recording on all the targets, or we only add this badge if there are active recording candidates on each node
2. If there are recordings then we must either generate a new report on view, or use a cached result. How long is the cache valid? Does the user need to approve generation, if so how? If this just happens automatically and there are a large number of reports to generate, it could require a lot of compute time and be expensive in real currency terms.
3. How do we summarize the report into a single badge form? Do all analysis rules have the same weight into the final badge?

[tthvo]

Maybe, there could be an action to create Automated Analysis on the target/group and some settings for refresh intervals or just a button to refresh (+ some expire warning).

For point 3, I am not so sure of anyway to combine results into a single one but is it possible? Or we have to define ourselves?

[roberttoyonaga]

What if the badge represents an active recording (if one exists), otherwise the most recent recording? Otherwise, no badge at all.

Would it be reasonable to say any "red" automated analysis results = red badge, otherwise any "yellow" results = yellow badge, otherwise green badge?

Like you mentioned, generating a lot of reports might be a problem at scale though.

[maxcao13]

Would it be reasonable to say any "red" automated analysis results = red badge, otherwise any "yellow" results = yellow badge, otherwise green badge?

Makes sense, this is what already happens for the automated analysis card.

Generating reports on hundreds of targets at the same time definitely won't scale. What about setting some checkbox/switch/etc. config in the topology view UI to say, "automatically start a report on this target perdioically" so it's an opt in feature for targets/pods/deployments that you specifically want to have at-a-glance monitoring for?

[andrewazores]

That sounds like it could be an enhancement to Automated Rules which we can also expose some UI for within Topology.