INFO: Started provisioning ECS Slave docker from aws-ecs with 1 executors. Remaining excess workload: 0
Sep 26, 2017 12:32:48 PM hudson.slaves.NodeProvisioner$2 run
WARNING: Unexpected exception encountered while provisioning agent ECS Slave docker
com.amazonaws.services.ecs.model.ClusterNotFoundException: Cluster not
found. (Service: AmazonECS; Status Code: 400; Error Code:
ClusterNotFoundException; Request ID:
f7bc84e2-a2a5-11e7-801f-1bbc4489a697)
The error makes no sense as the plugin does autodetect the cluster and fills it in the dropdown in the Manage Jenkins interface (so there's no manual input error possible...).
Does anyone have any idea what I should check/debug/investigate?
I can provide more info about the setup, just tell me what would help.
I was thinking about a workaround of using a "regular" Docker plugin, but I'm not super confident about it after seeing this:
https://github.com/jenkinsci/docker-plugin/issues/235#issuecomment-147975445
Is anyone using one of the Docker plugins for agents in production, and if so, are you happy with it/them? When I say production, I mean something preferably with tens if not hundreds of jobs configured :)