-
Notifications
You must be signed in to change notification settings - Fork 7
Conversation
Looks good so far 😀 thanks for your Pull Request 🎉 Regarding your points:
The rbac role is getting generated and updated via kubebuilder. Should we integrate something to the makefile to also update the rbac role in the helm chart? (sed or something like that to generate the helm variant from the kubebuilder generated one?)
I hope that I'll be able to finish the process by next week so we can publish images to quay.io.
What do we have for options here? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think we should split out the grafana dashboard to either a separate location or another PR.
Not everyone is using the integreatly.org/v1alpha1
CRs and the dashboard would be way more reusable then I think.
Added first version 0.1.0 of the cluster-api-state-metrics helm chart. Signed-off-by: Mehdi Yedes <[email protected]>
9a5f20d
to
29a8f6b
Compare
Agree, would be better to leave it out of this PR for now and then do it in another iteration. |
Yes, that would make sense. However, we should keep in mind that we need to publish a new helm chart release whenever the role(s) are changed. Of course, that would be ideally automated but we just need to keep that in mind.
Awesome!
I think Quay already supports helm chart repos(never worked with it though). But simply using github releases might also be an option for now. We could use chart-releaser, but I am not sure if it's possible to use it for this repo, since it's not dedicated to only storing helm charts. |
29a8f6b
to
cda7366
Compare
I think publishing a chart release should be fine using chart releaser's github action: https://github.com/helm/chart-releaser-action :-) but let's do this in a follow-up PR then to limit scope here.
What would you prefer as chart repo: github release or quay.io? I think github release would be better?! |
I agree. I'll try to update my PR in the upcoming few days. |
99d9e69
to
93d7254
Compare
I have updated the PR now. The failing checks should be resolved and the role templates for the helm chart are now generated from the ones under |
Hi @mehyedes sorry for the late reply. We are currently working on a proposal to contribute CASM to github.com/kubernetes-sigs/cluster-api . So I see some options here:
What do you think? |
Hey @chrischdi, thanks for your feedback. I think it depends on how the contribution of CASM into cluster-api would be. If CASM will be integrated into the cluster-api codebase, I believe there would be no need for a separate helm chart at all in that case. AFAIK, there is no "official" helm chart for cluster-api for now, but if there was one, then that's where I would add the CASM config. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sounds fair to me 👍 Just didn't want to raise the expectation that it will be integrated to the core repo automatically :-) So we have the same opinion here 😄 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
One tiny nit, rest looks really fine to me. Thanks for your contribution! 🙂🎉
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
/lgtm
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you a lot for the contribution.
Signed-off-by: Mehdi Yedes <[email protected]>
fixed indentation in deployment template removed grafanaDashboard template updated chart README Signed-off-by: Mehdi Yedes <[email protected]>
- Added script to generate role templates for the helm chart. The script is executed with `make manifests` - Added missing SPDX headers Signed-off-by: Mehdi Yedes <[email protected]>
Added helm-lint to the lint target Signed-off-by: Mehdi Yedes <[email protected]>
374ba9f
to
89e42cb
Compare
Use YAML instead JSON Co-authored-by: Tobias Giese <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Awesome. Thank you very much.
One last nit, the Readme seems not to be happy 🤔 |
Updated make targets docs
Thank you very much for your contribution 👍 If I remember correctly: the first external contribution for a Mercedes Benz repo on GitHub 🎉 |
This adds a helm chart for
cluster-api-state-metrics
Set default image repo (related to Find a proper destination registry #3)Add additional CI steps for the helm chartWill be done in a separate PRSet helm repo for publishing the chartWill be done in a separate PRIt would be great if you provide some feedback on how to proceed with the points above 😁
In the chart, there are also serviceMonitor and GrafanaDashboard templates. We thought they might be useful for anyone out there using the prometheus and grafana operators additionally.
Solves: #9