Fix node scaling

The playbook fails when removing unreachable nodes from deployment with
`openstack overcloud delete node`. Some `ignore_unreachable: true` are
missing. Also, one cannot use `any_errors_fatal: true` when ignoring
unreachable nodes, otherwise the playbook execution will stop after the
current task.

Change-Id: Ibcb84e58bac1975490df281c0de950cdf74337b2

common/deploy-steps-playbooks-common.yaml

@@ -1,7 +1,7 @@
 - hosts: "{{ deploy_source_host }}"
   name: Gather facts from undercloud
   gather_facts: yes
-  any_errors_fatal: yes
+  any_errors_fatal: "{{ not ignore_unreachable | default(true) }}"
   ignore_unreachable: "{{ ignore_unreachable | default(false) }}"
   become: false
   tags:
@@ -10,7 +10,7 @@
 - hosts: "{{ deploy_target_host }}"
   name: Gather facts from overcloud
   gather_facts: yes
-  any_errors_fatal: yes
+  any_errors_fatal: "{{ not ignore_unreachable | default(true) }}"
   ignore_unreachable: "{{ ignore_unreachable | default(false) }}"
   tags:
     - facts
@@ -18,7 +18,8 @@
 - hosts: all
   name: Load global variables
   gather_facts: "{{ gather_facts | default(false) }}"
-  any_errors_fatal: yes
+  any_errors_fatal: "{{ not ignore_unreachable | default(true) }}"
+  ignore_unreachable: "{{ ignore_unreachable | default(false) }}"
   tasks:
     - include_vars: global_vars.yaml
       no_log: true

common/deploy-steps.j2

@@ -1151,7 +1151,7 @@ outputs:
               #                  Without up-to-date facts, we're creating a potential failure
               #                  scenario.
               gather_facts: true
-              any_errors_fatal: yes
+              ignore_unreachable: true
               become: false
               vars:
                 bootstrap_server_id: BOOTSTRAP_SERVER_ID