How to handle a dead Docker daemon?

We’re using beta6 of Docker for AWS (and therefore Docker 1.12.2-rc1), and this is the second time I’ve seen the Docker daemon apparently die on the manager:

~ $ docker ps
Cannot connect to the Docker daemon. Is the docker daemon running on this host?

Last time this happened, I ended up having to recreate the Swarm :frowning: What’s the best way to recover from this situation, given that we don’t have root access to the manager? (or do we?)

Well, I made some progress on this. The Docker daemon is dead because the manager has run out of disk space… I assume this is because one of my containers has gone a bit bonkers and filled up all the space, but it’s a bit tricky without having the docker daemon running or root access. Any suggestions on how to get out of this hole much appreciated…