Cannot restart docker daemon on management nodes

[nralbers]

Expected behavior

when running service docker start on a management node, I expect the docker service on that node to restart

Actual behavior

When trying to configure the management nodes to export prometheus metrics via the guide on https://docs.docker.com/config/thirdparty/prometheus/, I discovered that it is not possible to restart the docker service.

Information

• Full output of the diagnostics from "docker-diagnose" ran from one of the instance

docker-diagnose OK hostname=swarm-manager000000 session=1556540019-NctZsb6LoDwZGgC7tYieWnFu79v76Dm0 OK hostname=swarm-manager000001 session=1556540019-NctZsb6LoDwZGgC7tYieWnFu79v76Dm0 OK hostname=swarm-manager000002 session=1556540019-NctZsb6LoDwZGgC7tYieWnFu79v76Dm0 OK hostname=swarm-worker000000 session=1556540019-NctZsb6LoDwZGgC7tYieWnFu79v76Dm0 OK hostname=swarm-worker000001 session=1556540019-NctZsb6LoDwZGgC7tYieWnFu79v76Dm0 Done requesting diagnostics. Your diagnostics session ID is 1556540019-NctZsb6LoDwZGgC7tYieWnFu79v76Dm0 Please provide this session ID to the maintainer debugging your issue.

• Page URL if this is a docs issue or the name of a man page https://docs.docker.com/config/thirdparty/prometheus/ Page does not describe how to restart docker service when running swarm in azure.

Alternatively, the azure template does not provide a configuration option to enable the prometheus endpoint on the management nodes.

Steps to reproduce the behavior

1. Create /etc/docker/daemon.json as described in documentation on a management node
2. sudo service docker restart