Description of problem:
I was running a test upgrade of OSP12/Ceph 2 to OSP13/Ceph 3. I hit an issue at the Ceph Storage upgrade stage. The ceph-ansible playbook fails at the following task:

TASK [ceph-client : kill a dummy container that created pool(s)/key(s)] ********
Wednesday 13 June 2018  14:32:19 +1000 (0:00:00.028)       0:05:17.843 ******** 
fatal: [192.0.2.107]: FAILED! => {"changed": false, "cmd": ["docker", "rm", "-f", "ceph-create-keys"], "delta": "0:00:00.019216", "end": "2018-06-13 04:32:19.560333", "msg": "non-zero return code", "rc": 1, "start": "2018-06-13 04:32:19.541117", "stderr": "Error response from daemon: No such container: ceph-create-keys", "stderr_lines": ["Error response from daemon: No such container: ceph-create-keys"], "stdout": "", "stdout_lines": []}

I think the container already gets removed due to the change introduced with this BZ:

https://bugzilla.redhat.com/show_bug.cgi?id=1568157

So when the command runs, it fails because the container has already been removed due to the introduced --rm option in the previous task.

Have tested without the "kill a dummy container" task and it seems to work.

Version-Release number of selected component (if applicable):
3.1.0-0.1.rc8.el7cp

How reproducible:
Always

Steps to Reproduce:
1. Upgrade an OSP12 with Ceph 2 to OSP13 with Ceph 3
2. Get to the Ceph upgrade stage
3. Wait until the Ceph upgrade step fails
4. Check the output of the mistral workflow execution and spot the failed step (see above)

Actual results:
My face -> :(

Expected results:
My face -> :)

Additional info: