robusta VS chartmuseum

To put one more option out there, we use Hikaru (https://pypi.org/project/hikaru/) in Robusta.dev (https://github.com/robusta-dev/robusta) and have been pretty happy with it. Example code below:

  with Pod().read(name='thename', namespace='the-namespace') as p:

Open Source alternative based on Prometheus

I'm from the Robusta team (https://github.com/robusta-dev/robusta) which bursts upwards towards the end. Happy to answer any questions!

The guys at Robusta, added it into thier slack so when alerts pop of issues in k8s, chatgpt can be used to investigate and offer options to check and potentially solutions. Next phase is probably to let it also apply it :)

It's no replacement for learning how things work, but we're trying to automate common troubleshooting cases with Robusta. https://github.com/robusta-dev/robusta

https://docs.robusta.dev/master/ https://github.com/robusta-dev/robusta

This is the sort of thing we built Robusta for: https://github.com/robusta-dev/robusta

We may switch over to Chartmusuem, and something like Charts-Syncer to try to help with this, or maybe abandon the whole idea of mirroring external repositories and just keep our repository hosting internal projects. What are your thoughts on this?

Also trying to find a small self-hosted container registry (not some beast like goharbor.io) and possibly a Helm chart repository (looking into chartmuseum.com). Anyone got some recommendations?

You can use either something like https://chartmuseum.com/ or any docker registry if it has OCI support.

ChartMuseum is indeed open source and is on GitHub.

This is also basically an s3 proxy, but it specifically implements the Terraform Registry API so that things like version constraints are handled correctly. If you use Helm at all, an analogous project for charts would be https://github.com/helm/chartmuseum

I raised a PR (WIP) https://github.com/helm/chartmuseum/pull/464 but wondering if others would think this is useful or if I just have a niche usecase.

Another benefit of Helm is in it's package management. If your application requires another team's application up and running, they can publish their Helm chart to a remote repository like a ChartMuseum. You can then install their application into your Kubernetes by naming that remote chart combined with a local values file. E.g., helm install other-teams-app https://charts.mycompany.com/other-teams-app-1.2.3.tgz -f values-other-teams-app.yaml. This is convenient because it means you don't have to checkout their project and dig through it for their helm charts to get up and running - all you need to supply is your own values file.