This repository has been archived by the owner. It is now read-only.
Please sign in to comment.
scheduler: Stop containers before booting new ones
Before this commit, if the join returned that a container with the same IP needed to be restarted (i.e. there was a container to be killed and a container to be booted with the same IP), the boot request would fail because Docker doesn't allow two running containers to have the same IP. While this currently doesn't happen because changes to container attributes result in new IP addresses, the upcoming secret implementation will result in containers restarting if their secret value in Vault changes. When this happens, a new IP is not assigned, triggering this bug.
- Loading branch information...