You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am using the helm-operator with one Helm chart and one entry in the watches.yaml, along with the --zap-log-level=info command line flag.
What did you expect to see?
I expected the logging output of the operator to be more or less concise, as one would typically expect from an info log level.
What did you see instead? Under which circumstances?
The operator logs a colored diff for every update in any deployed Helm chart release. In my case, the operator manages around 60 releases. These have to be updated from time to time, resulting in a stream of Helm release diffs, making the actual log output unreadable.
The verbose diff is printed in the HelmOperatorReconciler.Reconcile to standard out, when log.V(0).Enabled(), which seems to be true regardless of what parameter is given to --zap-log-level. My guess is the Enabled() check should be done for a higher (i.e., more verbose) log level.
Additional context
I also tried to pass --zap-log-level=0, --zap-log-level=1, and --zap-log-level=warning to the operator, with the same result.
The text was updated successfully, but these errors were encountered:
We have just merged #5307 on master now. It will be included in the 1.14 release, but can also be used currently with the helm-operator binary built from the master.
When running the Helm operator with the info zap log level, the Helm diffs will no longer be displayed. Feel free to reopen the issue if you have any concerns. Thanks!
Bug Report
What did you do?
I am using the
helm-operator
with one Helm chart and one entry in thewatches.yaml
, along with the--zap-log-level=info
command line flag.What did you expect to see?
I expected the logging output of the operator to be more or less concise, as one would typically expect from an
info
log level.What did you see instead? Under which circumstances?
The operator logs a colored diff for every update in any deployed Helm chart release. In my case, the operator manages around 60 releases. These have to be updated from time to time, resulting in a stream of Helm release diffs, making the actual log output unreadable.
Environment
Operator type:
/language helm
Kubernetes cluster type:
Vanilla Kubernetes
$ operator-sdk version
$ kubectl version
Possible Solution
The verbose diff is printed in the HelmOperatorReconciler.Reconcile to standard out, when
log.V(0).Enabled()
, which seems to be true regardless of what parameter is given to--zap-log-level
. My guess is theEnabled()
check should be done for a higher (i.e., more verbose) log level.Additional context
I also tried to pass
--zap-log-level=0
,--zap-log-level=1
, and--zap-log-level=warning
to the operator, with the same result.The text was updated successfully, but these errors were encountered: