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
While upgrading a helm3 deployment I was receiving timeout errors with a notification that the timeout was 300ns. This was triggered by a pre-upgrade hook Job resource. Since the job takes a few seconds to run it was causing the keel-managed helm upgrade to fail due to missing the timeout. Diving into the source I found this line:
However, ints are taken to represent nanoseconds (apparently). Based on other comments in the code it seems like this value was intended to be five minutes.
keel chart version: 0.9.8
container image: keelhq/keel:0.17.0-rc1
The text was updated successfully, but these errors were encountered:
While upgrading a helm3 deployment I was receiving timeout errors with a notification that the timeout was 300ns. This was triggered by a pre-upgrade hook Job resource. Since the job takes a few seconds to run it was causing the keel-managed helm upgrade to fail due to missing the timeout. Diving into the source I found this line:
https://github.com/keel-hq/keel/blob/master/provider/helm3/implementer.go#L22
This value gets implicitly cast into a
time.Duration
as that's the type of helm3's Upgrade.Duration field: https://pkg.go.dev/helm.sh/helm/v3/pkg/action#UpgradeHowever, ints are taken to represent nanoseconds (apparently). Based on other comments in the code it seems like this value was intended to be five minutes.
keel chart version: 0.9.8
container image: keelhq/keel:0.17.0-rc1
The text was updated successfully, but these errors were encountered: