commands: Add support for retrying commands until a condition is reached

This can be useful for waiting for some state to change in the system.

Author: markgoddard

roles/commands/README.md

@@ -33,3 +33,9 @@ with at least one host in it - see the `cephadm` role for more details.
             - "fs new cephfs cephfs_metadata cephfs_data"
             - "orch apply mds cephfs --placement 3"
    ```
+
+* `cephadm_commands_until` A expression to evaluate to allow retrying commands. May reference the registered result variable, `cephadm_commands_result`. Default is `true` (do not use retries).
+
+* `cephadm_commands_retries`: Number of retries to use with `cephadm_commands_until`. Default is 0.
+
+* `cephadm_commands_delay`: Delay between retries with `cephadm_commands_until`. Default is 0.

roles/commands/defaults/main.yml

@@ -1,3 +1,6 @@
 ---
 cephadm_command: ceph
 cephadm_commands: []
+cephadm_commands_until: true
+cephadm_commands_retries: 0
+cephadm_commands_delay: 0

roles/commands/tasks/main.yml

@@ -9,6 +9,9 @@
   when: cephadm_commands | length > 0
   delegate_to: "{{ groups['mons'][0] }}"
   run_once: true
+  until: cephadm_commands_until
+  retries: "{{ cephadm_commands_retries }}"
+  delay: "{{ cephadm_commands_delay }}"
   vars:
     # NOTE: Without this, the delegate hosts's ansible_host variable will not
     # be respected.