ArgoCD healthcheck and GrafanaDashboard about grafana-operator HOT 5 OPEN

@zelig81 we haven't come that far to have given it any thought on how the status would look like.

It would be great if you took a look at the Alert implementation and see how it's done there and see how if you think it would fit for the other CRs.

We are also planning to go towards finalizers in the other CRs which should remove the need for some of the existing statuses. It's probably a good idea to do that at the same time. But I do understand that it will make this refactor allot bigger.

So please come with a suggestion on how you think it could be done, and we will try to provide feedback as quickly as possible. The maintainers also have weekly bug scrub meetings, and if you want to join and talk about this refactoring, you are also free to do that.
You can just ping me on Kubernetes/Grafana slack and I can send you the info.

Btw, I think it sounds like a great idea to have a healtcheck in the upstream argo repo around this.

from grafana-operator.

Hi @zelig81 , that is really nice.
We are currently working on getting alert support added to the grafana-operator as can be seen here: #1420.
In it, we are using conditions.

When it gets merged we are planning to change the other CRs to match how status is managed in the alert. How exactly they will look like we haven't decided.

So in short, your issue is in our TODO, we just haven't created any issue for it. So this will be a great way for us to keep track of it. Thank you.

When those changes get merged, it would be great if we can work on custom health checks for ArgoCD and write some docs about it.

If you are interested in contributing the new status solution to the existing CRs that would be great. Else someone maintainer will take a stab at it in the future.

from grafana-operator.

I would be delighted to assist with this. Just please let me know the format of the conditions.

from grafana-operator.

In general, we just need to understand which status/condition will be relevant for which of 4 states of the resource in terms of ArgoCD (Progressing; Healthy; Degraded; Suspended(if relevant) )

It will be most nice to have all the types of CR to follow this convention (Grafana, GrafanaFolder, GrafanaDatasource, GrafanaDashboard, GrafanaAlertrule...). Although it can look a little bit strange, it will allow us to make a single ArgoCD healthcheck to rule them all :)
And to add the healthcheck straight to https://github.com/argoproj/argo-cd

from grafana-operator.

let's continue the conversation on slack

from grafana-operator.