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
In provider kubernetes, I want to test some alpha features using uptest. I am using hooks to patch DeploymentRuntimeConfig with feature flags which works fine. However, I am observing provider resource reporting unhealthy after each example preventing rollout of new config to the provider controller.
Upon further investigation, I noticed that this is because how we handle imports. It looks like, we stop / start Crossplane which causes losing the package in the cache which was loaded in the beginning of e2e flow.
Based on my offline conversations with @sergenyalcin, we do this to stop / start providers to clear the state cache in the providers to ensure we validate import functionality. We should achieve the same by patching the replicas in the respective DeploymentRuntimeConfig.
How can we reproduce it?
Run e2e with uptest and check provider health using:
❯ kubectl get providers.pkg.crossplane.io
NAME INSTALLED HEALTHY PACKAGE AGE
provider-kubernetes True False provider-kubernetes-v0.14.0-rc0.60.gdb2a37c.gz 59m
What environment did it happen in?
Uptest Version: v0.9.0
The text was updated successfully, but these errors were encountered:
What happened?
In provider kubernetes, I want to test some alpha features using uptest. I am using hooks to patch
DeploymentRuntimeConfig
with feature flags which works fine. However, I am observing provider resource reporting unhealthy after each example preventing rollout of new config to the provider controller.Upon further investigation, I noticed that this is because how we handle imports. It looks like, we stop / start Crossplane which causes losing the package in the cache which was loaded in the beginning of e2e flow.
Based on my offline conversations with @sergenyalcin, we do this to stop / start providers to clear the state cache in the providers to ensure we validate import functionality. We should achieve the same by patching the replicas in the respective
DeploymentRuntimeConfig
.How can we reproduce it?
Run e2e with uptest and check provider health using:
What environment did it happen in?
v0.9.0
The text was updated successfully, but these errors were encountered: