helm-controller: logs from Helm actions no longer seem to be captured

[dkulchinsky]

Hey folks 👋🏼

We've recently upgraded Flux from v0.35.0 to v0.41.2 and now observe that when a HelmRelease fails to reconcile due to timed out waiting for the condition we no longer see the log actions which indicate which wait action has timed out.

Component versions:

helm-controller: v0.31.2
kustomize-controller: v0.35.1
notification-controller: v0.33.0
source-controller: v0.36.1

The Last Helm logs event we see in v0.35.0:

Last Helm logs:

creating 4 resource(s)
beginning wait for 4 resources with timeout of 5m0s
Deployment is not ready: thanos/thanos. 0 out of 1 expected pods are ready

which is also consistent with thetroubleshooting docs.

However, the same Last Helm logs event in v0.41.2 now looks like this:

      Last Helm logs:

      resetting values to the chart's original version
      performing update for thanos
      creating upgraded release for thanos
      waiting for release thanos resources (created: 0 updated: 23  deleted: 5)
      warning: Upgrade "thanos" failed: timed out waiting for the condition

From a cursory review of the recent helm-controller and helm changes, I can't see anything that should have this impact, but something has changed it seems.

The previous output was extremely helpful and made troubleshooting a breeze, with the new output especially on the more complex releases, it is quite a pain to figure out what exactly has failed.