From a1443bb8271deb8b134c5cd6660ab85cf2f5ed39 Mon Sep 17 00:00:00 2001
From: Emilien Macchi <emilien@redhat.com>
Date: Tue, 9 Jun 2020 12:32:37 -0400
Subject: [PATCH] Relax facts gathering plays on the overcloud

Tolerate failures on all the plays that run on the overcloud and which
potentially run with facts gathering only (no other tasks).

Since they can't apply a strategy, we want to let them fail without
error. In a later play, we have our tripleo strategies which will figure
out if we have reached the maximum of failures that was configured with
MaxFailPercentage.

Change-Id: I43652972494a9ed1f2b8b483fe19e685cbe4da1b
(cherry picked from commit 107b9c4c903924784a9829a405bbdd05b3cb1931)
---
 common/deploy-steps-playbooks-common.yaml | 21 ++++++++++++++++-----
 common/deploy-steps.j2                    |  1 -
 2 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/common/deploy-steps-playbooks-common.yaml b/common/deploy-steps-playbooks-common.yaml
index 5ec1350a7c..e977afcb83 100644
--- a/common/deploy-steps-playbooks-common.yaml
+++ b/common/deploy-steps-playbooks-common.yaml
@@ -1,3 +1,10 @@
+---
+# If a new play with potential facts gathering is added, make sure we set:
+# any_errors_fatal: false
+# ignore_unreachable: true
+# So we don't break the MaxFailPercentage feature since these plays can't use
+# Ansible strategies.
+
 - hosts: all
   name: Clear cached facts
   tasks:
@@ -7,7 +14,7 @@
   name: Gather facts from undercloud
   gather_facts: yes
   any_errors_fatal: true
-  ignore_unreachable: "{{ scale_ignore_unreachable | default(false) }}"
+  ignore_unreachable: false
   become: false
   tags:
     - facts
@@ -17,7 +24,11 @@
   gather_facts: yes
   # False because https://github.com/ansible/ansible/issues/70663
   any_errors_fatal: false
-  ignore_unreachable: "{{ scale_ignore_unreachable | default(false) }}"
+  # If an overcloud node is down, we will let MaxFailPercentage
+  # figuring out if the deployment can continue. For the facts gathering tasks,
+  # we will simply ignore unreachable nodes and errors, and let the
+  # Ansible reports the failure in the next plays.
+  ignore_unreachable: true
   tags:
     - facts
 
@@ -25,7 +36,7 @@
   name: Load global variables
   gather_facts: "{{ gather_facts | default(false) }}"
   any_errors_fatal: false
-  ignore_unreachable: "{{ scale_ignore_unreachable | default(false) }}"
+  ignore_unreachable: true
   tasks:
     - include_vars: global_vars.yaml
       no_log: true
@@ -38,7 +49,7 @@
   become: true
   gather_facts: false
   any_errors_fatal: false
-  ignore_unreachable: "{{ scale_ignore_unreachable | default(false) }}"
+  ignore_unreachable: true
   tasks:
     - name: Ensure /var/lib/config-data context
       shell: |-
@@ -77,7 +88,7 @@
   name: Set all_nodes data as group_vars for overcloud
   gather_facts: "{{ gather_facts | default(false) }}"
   any_errors_fatal: false
-  ignore_unreachable: "{{ scale_ignore_unreachable | default(false) }}"
+  ignore_unreachable: true
   tasks:
     - name: Set all_nodes data as group_vars for overcloud
       include_vars: "{{ playbook_dir }}/group_vars/overcloud.json"
diff --git a/common/deploy-steps.j2 b/common/deploy-steps.j2
index e3ba115136..e53f249d2f 100644
--- a/common/deploy-steps.j2
+++ b/common/deploy-steps.j2
@@ -1144,7 +1144,6 @@ outputs:
               vars:
                 deploy_source_host: "DEPLOY_SOURCE_HOST"
                 deploy_target_host: "DEPLOY_TARGET_HOST"
-                scale_ignore_unreachable: true
 
             - hosts: DEPLOY_TARGET_HOST
               name: Scaling