"update in progress" stuck

[dimasusername]

During copilot svc deploy sometimes when the application fails to start (boot error, or something similar), the deployer will continue trying to restart the app for a very long time.

And to deploy the fix, I either have to wait, or go into the AWS Console, cancel the cluster update and rollback, which is troublesome.

Is there any way to stop this process from the cli and re-deploy a new version of the image? Or at least control the number of retries and the delay between them?

[huanjani]

Hello @dimasusername!
I believe your long delay is due to Circuit Breaker waiting for 10 failures.
Would you mind adding a 👍🏼 here: aws/containers-roadmap#1247 ?
Thanks!

[efekarakus]

Heya! I'll close this issue for now, once the circuit breaker behavior provides more configuration knobs we will surface it in the manifest files!