Ian Delahorne
Ian Delahorne
@perenesenko thanks for confirming that you can reproduce. Yes, from what I observe the Target Group weight is being managed correctly and without delay (that is, during the period traffic...
> @iandelahorne I guess it's happening because for the **Ping-pong feature** we're using **readinessGates** mechanism to make pod **ready**. And it take some time for the AWS controller to update...
Having the circuitbreaker enabled on deployments would be amazing, this would lower a lot of confusion around failed/stuck ECS deploys
@spinnakerbot remove-label stale
@spinnakerbot remove-label stale
This seems to be due to how creating task definitions are handled differently if they're from an artifact (https://github.com/spinnaker/clouddriver/blob/10bba2d2e18506c3eebfd2e466cf0ee1d8a4c71e/clouddriver-ecs/src/main/java/com/netflix/spinnaker/clouddriver/ecs/deploy/ops/CreateServerGroupAtomicOperation.java#L421-L441) vs from the wizard (https://github.com/spinnaker/clouddriver/blob/10bba2d2e18506c3eebfd2e466cf0ee1d8a4c71e/clouddriver-ecs/src/main/java/com/netflix/spinnaker/clouddriver/ecs/deploy/ops/CreateServerGroupAtomicOperation.java#L307)
@spinnakerbot remove-label stale
@spinnakerbot remove-label stale
@spinnakerbot remove-label stale
@spinnakerbot remove-label stale