SBX-Added servicemonitor and services to scrape alastria metrics correctly

[alejandro-nieto-git]

Added service monitor, services and endpoints in order to scrape metrics correctly and populate alastria t network and b network grafana dashboards. I think it should work as is but please take a look and correct me if some file should go to another folder in order for ArgoCD to pick it up correctly @groundnuty

Regards.

[groundnuty]

Why do you think that Helm templates will work here?
This is the definition of the argocd app for kube-prom stack https://github.com/DOME-Marketplace/dome-gitops/blob/main/applications_prd/monitoring/kube-prometheus-stack.yaml

[alejandro-nieto-git]

Why do you think that Helm templates will work here? This is the definition of the argocd app for kube-prom stack https://github.com/DOME-Marketplace/dome-gitops/blob/main/applications_prd/monitoring/kube-prometheus-stack.yaml

Thank you! just wanted to keep it as abstract as possible and did not realise I could not put the templates there. I substituted all the values and now it should be ready to go.

[groundnuty]

ok, but this will still not work, it's not how argo/helm works here at all :-)
In order to do what you want properly, you need to take all the files you want to put into ionos_sbx/kube-prometheus-stack and add them to the helm chart of the application you want to monitor, and then then they can be helm templates.
I can't find alastria helm chart in our gitops repo. Which app/chart it is?

On the other hand, the files you want to put in ionos_common/grafana-dashboards are ok, as this is the designated place to put dashboards.

[alejandro-nieto-git]

ok, but this will still not work, it's not how argo/helm works here at all :-) In order to do what you want properly, you need to take all the files you want to put into ionos_sbx/kube-prometheus-stack and add them to the helm chart of the application you want to monitor, and then then they can be helm templates. I can't find alastria helm chart in our gitops repo. Which app/chart it is?

On the other hand, the files you want to put in ionos_common/grafana-dashboards are ok, as this is the designated place to put dashboards.

The thing is that we don't want to monitor an internal application deployed with a helm chart since the blockchain nodes are completely decentralised, so there is no application to act as an "umbrella" to hold and fill the templates. This is aiming to monitor alastria t and b network health so therefore it needs to monitor those external nodes.

[alejandro-nieto-git]

In local I used the kube-prometheus-stack helm chart with an umbrella chart to add the templates to it and it works but I don't think it is appropriate to do it here.

[alejandro-nieto-git]

ok, but this will still not work, it's not how argo/helm works here at all :-) In order to do what you want properly, you need to take all the files you want to put into ionos_sbx/kube-prometheus-stack and add them to the helm chart of the application you want to monitor, and then then they can be helm templates. I can't find alastria helm chart in our gitops repo. Which app/chart it is?
On the other hand, the files you want to put in ionos_common/grafana-dashboards are ok, as this is the designated place to put dashboards.

The thing is that we don't want to monitor an internal application deployed with a helm chart since the blockchain nodes are completely decentralised, so there is no application to act as an "umbrella" to hold and fill the templates. This is aiming to monitor alastria t and b network health so therefore it needs to monitor those external nodes.

I suppose that adding the new needed yaml files to:

repoURL: https://github.com/DOME-Marketplace/dome-gitops
    targetRevision: feature/monitoring
    path: ionos_sbx/kube-prometheus-stack
    directory:
      include: '{kube-prometheus-stack-sso-secret.yaml,kube-prometheus-stack-discord-webhook-infrastructure-secret.yaml,standard-cluster-monitoring-recording-rules.yaml,onzack-namespace-monitoring-recording-rules.yaml}'

should work, and be the best approach but I'm open to other alternatives.

[alejandro-nieto-git]

just uploaded what I was talking about @groundnuty

[bialask-daiteap]

Please squash the commits.

[alejandro-nieto-git]

This PR was pending to be refactored to be aligned to a more standard way suggested by @groundnuty. Since this is aiming to monitor nodes of the alastria network just to see that the network is working properly directly from DOME Grafana but it is also in Alastria Grafana, and given the dates we are in we either go ahead or drop this PR for now and keep this in a separate branch. We can also go for the refactor if we want to have this feature added "by the book" but again, maybe given the dates the other approaches can be better @bialask-daiteap. Let me hear your thoughts and we will act accordingly.