/var/log/mistral/ceph-install-workflow.log:
2018-01-25 13:55:45,889 p=18529 u=mistral | PLAY [confirm whether user really meant to switch from non-containerized to containerized ceph daemons] ***
2018-01-25 13:55:45,900 p=18529 u=mistral | TASK [exit playbook, if user did not mean to switch from non-containerized to containerized daemons?] ***
2018-01-25 13:55:45,931 p=18529 u=mistral | fatal: [localhost]: FAILED! => {"changed": false, "msg": "\"Exiting switch-from-non-containerized-to-containerized-ceph-daemons.yml playbook,\n cluster did not switch from non-containerized to containerized ceph daemons.\n To switch from non-containerized to containerized ceph daemons, either say 'yes' on the prompt or\n or use `-e ireallymeanit=yes` on the command line when\n invoking the playbook\"\n"}
2018-01-25 13:55:45,931 p=18529 u=mistral | PLAY RECAP *********************************************************************
This seems to be a regression introduced by the fix to 1736707.
Even though 'ireallymeanit: yes' is in ceph-base.yaml [0], according to Mistral's last task for ceph-ansible, the Heat to Mistral call has it but the playbook was called without it [1]. How parameters are passed from Mistral to ceph-ansible was changed recently [2] so it seems this type of parameter didn't handle that change.
[0] https:/ /review. openstack. org/#/c/ 494246/ 1/docker/ services/ ceph-ansible/ ceph-base. yaml paste.openstack .org/show/ 653502/ /review. openstack. org/#/c/ 528755/ 1/workbooks/ ceph-ansible. yaml
[1] http://
[2] https:/