Helm Charts: Support for Prometheus Operator

[alexandrev]

Currently the helm charts supports integration with Prometheus by the installation on the traditional kubernetes way or even on-premises if required, but it doesn't support the Prometheus Operator installation.

This is important because this is the default installation for many Kubernetes distributions such as Openshift or Rancher.

Regards

[mdiez-tibco]

Hi @alexandrev,

This is a fair request. We can explore that possibility. Note however that, as stated in the first README:

The Cloud Deployment Kit for TIBCO Spotfire® provides a reference and starting point for deploying TIBCO Spotfire® using containers and Helm charts on a Kubernetes cluster.

This repository contains recipes to build container images and Helm charts for certain TIBCO Spotfire® products. You can extend and customize these recipes and examples.

This means that we intend to provide a "vanilla"/generic/canonical/reference/baseline deployment. We do not intend to cover all possible deployment cases. We still want to know which configuration or integrations may be "common" so we can include them as part of this reference deployment.

Thanks

[alexandrev]

Sure, but I think it makes sense to have it ready so somebody can create a PR associated to this issue.

[mjohanss-tibco]

@alexandrev Can you explain what is missing for the current chart implementation? The pods do expose scrape endpoints and prometheus podAnnotations by default, making it possible for prometheus to scrape the spotfire pods. Do you have any additional information or references regarding your request? Thanks

[alexandrev]

Sure, Prometheus Operator doesn’t work using the annotations but using CRD such as ServiceMonitor and PrometheusRules. What is missing is that the chart should create the relevant ServiceMonitor as part of the deployment so this can be recognized by Prometheus Operator deployment. Thanks in advance,

…

On 10 Oct 2022, at 10:49, mjohanss-tibco ***@***.***> wrote: @alexandrev <https://github.com/alexandrev> Can you explain what is missing for the current chart implementation? The pods do expose scrape endpoints and prometheus podAnnotations by default, making it possible for prometheus to scrape the spotfire pods. Do you have any additional information or references regarding your request? Thanks — Reply to this email directly, view it on GitHub <#2 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAHHGYNYAIUCVHB3YDWUT7TWCPKB3ANCNFSM6AAAAAAQAJ7ZQE>. You are receiving this because you were mentioned.