Description

When fargate has an atlantis task running and a new deployment was issued the existing task prevents the new one from starting with this error :

Error: initializing server: starting BoltDB: timeout (a possible cause is another Atlantis instance already running

Note: EFS is being used.

• ✋ I have searched the open/closed issues and my issue is not listed.

Reproduction Code [Required]

Update terraform to redeploy the atlantis fargate task.

Expected behavior

The new tasks starts and fargate stops the old task.

Actual behavior

The new task fails with above BoltDB error which makes the new task fail, fargate retries starting a new task which also fails, etc.

Terminal Output Screenshot(s)

Error: initializing server: starting BoltDB: timeout (a possible cause is another Atlantis instance already running

Workaround

After deploying an update to this module manually stop all running tasks and wait for fargate to start the new one.