-
Notifications
You must be signed in to change notification settings - Fork 3.3k
Description
Describe the bug
Containerapp with kind = functionapp.
Scale triggers are automatically handled by the function host.
When issuing command:
az containerapp update \
--name somename \
--resource-group somerg \
--image someimageregistry:commithash
all existing scale rules are removed. This causes an issue, as the azure function host checks a hash on blob storage, and if the hash is the same, it doesn't perform a synctriggers, to re-create the rules.
I effectively end up with a non-functioning containerapp every time I deploy.
The only workaround I have found it to go and manually delete the blob that contains the hash, and restart the containerapp.
Related command
az containerapp update \
--name somename \
--resource-group somerg \
--image someimageregistry:commithash
Errors
No errors, output shows rules removed.
Issue script & Debug output
"scale": ***
"cooldownPeriod": 300,
"maxReplicas": 3,
"minReplicas": 0,
"pollingInterval": 30,
"rules": null
***,
Expected behavior
Expect existing scale rules to not be deleted on az containerapp update.
Environment Summary
azure-cli available on github runner (ubuntu-latest)
Additional context
No response