- All nodes are BYO EC2 instances.
- Staging nodes & CI (builder) node are scaled down overnight.
- This script is run on node startup: https://github.com/vidsy/docker-cloud-startup
- Nodes scaled down (terminated) at ~9pm UTC.
- They remained in
- New nodes started at ~7am UTC as the ASG scaled back up.
- New nodes tried to redeploy stack.
- Redeploy failed as Docker Cloud tries to redeploy services that it thinks are still running on the
- Log into web app and terminate each of the
- Redeploy stack again.
"Unreachable" nodes ever get removed automatically?