fix: supporting unauthenticated version endpoint

[prometherion]

Closes #315.

[prometherion]

@sybnex I tested this personally and seems fixing what you're looking for.

$: curl -k https://127.0.0.1:9001/version
{
  "major": "1",
  "minor": "25",
  "gitVersion": "v1.25.3",
  "gitCommit": "434bfd82814af038ad94d62ebe59b133fcb50506",
  "gitTreeState": "clean",
  "buildDate": "2022-10-25T19:35:11Z",
  "goVersion": "go1.19.2",
  "compiler": "gc",
  "platform": "linux/amd64"
}

For the /version endpoint we're skipping the error in case of retrieval of authentication data, bypassing the user impersonation and decorating the request with the bearer token which would be required to perform the necessary user impersonation.

I wanted to give it a try also if we support a protected endpoint, and it seems to do the trick:

$: curl -k -H 'Authorization: Bearer <REDACTED>' https://127.0.0.1:9001/version
{
  "major": "1",
  "minor": "25",
  "gitVersion": "v1.25.3",
  "gitCommit": "434bfd82814af038ad94d62ebe59b133fcb50506",
  "gitTreeState": "clean",
  "buildDate": "2022-10-25T19:35:11Z",
  "goVersion": "go1.19.2",
  "compiler": "gc",
  "platform": "linux/amd64"
}

This would be required for those Kubernetes distributions which are protecting the version endpoint, thus a valid token, and a user along with groups, must be enforced.

It would be great if you could give it a try, ensuring it works as expected in your environment.

[sybnex]

Hi,

thanks for your big effort! The part, you developed, with the version works as designed!

Unfortunately it didn't solved the problem. I increased all debug logs to the max but couldn't find any hint what's the problem with the dashboard. It is still not starting and hangs in a CrashLoopBackOff.

I even tried to create an own tenant for the dashboard with his serviceaccount as owner but wasn't successfully.
Last thing would be sniffing the network ...

Is there any change to increase the logs of the capsule-proxy by more than 10?

[prometherion]

Unless sharing more details, such the reason for the crashloopbackoff, dashboard logs, and further details, I cannot help you.

I would suggest moving the discussion to the issue itself, since this place must be used mostly to address changes and technical discussions.

[oliverbaehler]

Stale, Reopen on activity