hotloop: Optimize retry metrics tracking

Moved retry metrics collection from conditional `set_fact` tasks into
the Python modules themselves using `ansible_facts`. This eliminates
the performance cost of evaluating and skipping tasks for every stage.

Changes:
- Modules now update `hotloop_retry_metrics` fact directly via `ansible_facts`
- Removed conditional "Track retry metrics" tasks from all stage types
- Initialize fact at role start for clean metrics per execution
- Removed redundant 'type' field from metrics structure
- Metrics summary now always displays (shows "0" when no retries)

Assisted-By: Claude Code/claude-4.5-sonnet
Signed-off-by: Harald Jensås <hjensas@redhat.com>

Author: hjensas

roles/hotloop/library/hotloop_oc_apply_file.py

@@ -53,6 +53,22 @@
       - The timeout for the oc apply command
     type: int
     default: 60
+  stage_name:
+    description:
+      - The name of the stage for retry metrics tracking
+    type: str
+    required: false
+  resource_identifier:
+    description:
+      - The resource identifier for retry metrics (e.g., original manifest path from config)
+    type: str
+    required: false
+  hotloop_retry_metrics:
+    description:
+      - Current list of retry metrics to append to
+    type: list
+    required: false
+    default: []
 
 author:
     - Harald Jensås <hjensas@redhat.com>
@@ -216,6 +232,31 @@ def save_failed_manifest(file, rc, outs, errs, timeout):
     return failed_base
 
 
+def add_retry_metrics_fact(
+    result, current_metrics, stage_name, resource_identifier, retry_count, retry_time
+):
+    """Add retry metrics to ansible_facts in the result.
+
+    :param result: The module result dictionary to update.
+    :param current_metrics: Current list of retry metrics.
+    :param stage_name: The name of the stage.
+    :param resource_identifier: The resource identifier (e.g., manifest file path).
+    :param retry_count: Number of retries that occurred.
+    :param retry_time: Total time spent in retries.
+    """
+    result["ansible_facts"] = {
+        "hotloop_retry_metrics": current_metrics
+        + [
+            {
+                "stage": stage_name,
+                "file": resource_identifier,
+                "retry_count": retry_count,
+                "retry_time": retry_time,
+            }
+        ]
+    }
+
+
 def no_diff(file):
     """Check if the file is different from the previously applied version.
 
@@ -247,6 +288,9 @@ def run_module():
 
     file = module.params["file"]
     timeout = module.params["timeout"]
+    stage_name = module.params.get("stage_name")
+    resource_identifier = module.params.get("resource_identifier", file)
+    hotloop_retry_metrics = module.params.get("hotloop_retry_metrics", [])
 
     try:
 
@@ -292,6 +336,15 @@ def run_module():
                 msg += " (WARNING: {count} retries after {time}s due to transient errors)".format(
                     count=retry_count, time=retry_time
                 )
+                # Update ansible_facts with retry metrics
+                add_retry_metrics_fact(
+                    result,
+                    hotloop_retry_metrics,
+                    stage_name,
+                    resource_identifier,
+                    retry_count,
+                    retry_time,
+                )
             result["msg"] = msg
             result["success"] = True
             result["changed"] = True

roles/hotloop/library/hotloop_oc_apply_kustomize.py

@@ -50,6 +50,22 @@
       - The timeout for the oc apply command
     type: int
     default: 60
+  stage_name:
+    description:
+      - The name of the stage for retry metrics tracking
+    type: str
+    required: false
+  resource_identifier:
+    description:
+      - The resource identifier for retry metrics (e.g., original directory path from config)
+    type: str
+    required: false
+  hotloop_retry_metrics:
+    description:
+      - Current list of retry metrics to append to
+    type: list
+    required: false
+    default: []
 
 author:
     - Harald Jensås <hjensas@redhat.com>
@@ -173,6 +189,31 @@ def apply_kustomize(directory, timeout=60):
     return rc, outs, errs, out_lines, err_lines
 
 
+def add_retry_metrics_fact(
+    result, current_metrics, stage_name, directory, retry_count, retry_time
+):
+    """Add retry metrics to ansible_facts in the result.
+
+    :param result: The module result dictionary to update.
+    :param current_metrics: Current list of retry metrics.
+    :param stage_name: The name of the stage.
+    :param directory: The kustomize directory path.
+    :param retry_count: Number of retries that occurred.
+    :param retry_time: Total time spent in retries.
+    """
+    result["ansible_facts"] = {
+        "hotloop_retry_metrics": current_metrics
+        + [
+            {
+                "stage": stage_name,
+                "directory": directory,
+                "retry_count": retry_count,
+                "retry_time": retry_time,
+            }
+        ]
+    }
+
+
 def validate_directory(directory):
     """Validate the directory parameter.
 
@@ -235,6 +276,9 @@ def run_module():
 
     directory = module.params["directory"]
     timeout = module.params["timeout"]
+    stage_name = module.params.get("stage_name")
+    resource_identifier = module.params.get("resource_identifier", directory)
+    hotloop_retry_metrics = module.params.get("hotloop_retry_metrics", [])
 
     try:
         # Validate directory parameter
@@ -273,6 +317,15 @@ def run_module():
             msg = f"Kustomize directory {directory} applied"
             if retry_count > 0:
                 msg += f" (WARNING: {retry_count} retries after {retry_time}s due to transient errors)"
+                # Update ansible_facts with retry metrics
+                add_retry_metrics_fact(
+                    result,
+                    hotloop_retry_metrics,
+                    stage_name,
+                    resource_identifier,
+                    retry_count,
+                    retry_time,
+                )
             result["msg"] = msg
             result["success"] = True
             result["changed"] = True

roles/hotloop/tasks/kustomize.yml

@@ -75,18 +75,6 @@
   hotloop_oc_apply_kustomize:
     directory: "{{ _kustomize_apply_dir }}"
     timeout: "{{ item.kustomize.timeout | default(60) }}"
-  register: kustomize_result
-
-- name: "Stage: {{ item.name }} :: Track retry metrics"
-  ansible.builtin.set_fact:
-    hotloop_retry_metrics: >-
-      {{
-        hotloop_retry_metrics | default([]) + [{
-          'stage': item.name,
-          'type': 'kustomize',
-          'directory': item.kustomize.directory,
-          'retry_count': kustomize_result.retry_count | default(0),
-          'retry_time': kustomize_result.retry_time | default(0)
-        }]
-      }}
-  when: kustomize_result.retry_count | default(0) | int > 0
+    stage_name: "{{ item.name }}"
+    resource_identifier: "{{ item.kustomize.directory }}"
+    hotloop_retry_metrics: "{{ hotloop_retry_metrics }}"

roles/hotloop/tasks/main.yml

@@ -14,6 +14,10 @@
 # License for the specific language governing permissions and limitations
 # under the License.
 
+- name: Initialize retry metrics
+  ansible.builtin.set_fact:
+    hotloop_retry_metrics: []
+
 - name: Assert config is defined
   ansible.builtin.assert:
     that:
@@ -114,7 +118,6 @@
     label: "{{ item.name }}"
 
 - name: Display retry metrics summary
-  when: hotloop_retry_metrics is defined
   ansible.builtin.include_tasks: retry_metrics.yml
 
 - name: Remove temporary hotloop work directory

roles/hotloop/tasks/retry_metrics.yml

@@ -18,19 +18,19 @@
   vars:
     _total_retries: >-
       {{
-        hotloop_retry_metrics
+        hotloop_retry_metrics | default([])
         | map(attribute='retry_count')
         | map('int')
         | sum
       }}
     _total_retry_time: >-
       {{
-        hotloop_retry_metrics
+        hotloop_retry_metrics | default([])
         | map(attribute='retry_time')
         | map('int')
         | sum
       }}
-    _stages_with_retries: "{{ hotloop_retry_metrics | length }}"
+    _stages_with_retries: "{{ hotloop_retry_metrics | default([]) | length }}"
   ansible.builtin.debug:
     msg: |
 
@@ -42,9 +42,8 @@
       Total Retry Attempts:      {{ _total_retries }}
       Total Retry Time:          {{ _total_retry_time }}s
 
-      {% for metric in hotloop_retry_metrics %}
+      {% for metric in hotloop_retry_metrics | default([]) %}
       ┌─ Stage: {{ metric.stage }}
-      │  Type:     {{ metric.type }}
       │  Resource: {{ metric.file | default(metric.directory | default('N/A')) }}
       │  Retries:  {{ metric.retry_count }}
       │  Time:     {{ metric.retry_time }}s

roles/hotloop/tasks/static_manifest.yml

@@ -83,18 +83,6 @@
           item.manifest | ansible.builtin.basename
         ] | ansible.builtin.path_join
       }}
-  register: _manifest_apply_result
-
-- name: "Stage: {{ item.name }} :: Track retry metrics"
-  ansible.builtin.set_fact:
-    hotloop_retry_metrics: >-
-      {{
-        hotloop_retry_metrics | default([]) + [{
-          'stage': item.name,
-          'type': 'manifest',
-          'file': item.manifest,
-          'retry_count': _manifest_apply_result.retry_count | default(0),
-          'retry_time': _manifest_apply_result.retry_time | default(0)
-        }]
-      }}
-  when: _manifest_apply_result.retry_count | default(0) | int > 0
+    stage_name: "{{ item.name }}"
+    resource_identifier: "{{ item.manifest }}"
+    hotloop_retry_metrics: "{{ hotloop_retry_metrics }}"

roles/hotloop/tasks/template_manifest.yml

@@ -88,18 +88,6 @@
           item.j2_manifest | ansible.builtin.basename | ansible.builtin.splitext | first
         ] | ansible.builtin.path_join
       }}
-  register: _j2_manifest_apply_result
-
-- name: "Stage: {{ item.name }} :: Track retry metrics"
-  ansible.builtin.set_fact:
-    hotloop_retry_metrics: >-
-      {{
-        hotloop_retry_metrics | default([]) + [{
-          'stage': item.name,
-          'type': 'j2_manifest',
-          'file': item.j2_manifest,
-          'retry_count': _j2_manifest_apply_result.retry_count | default(0),
-          'retry_time': _j2_manifest_apply_result.retry_time | default(0)
-        }]
-      }}
-  when: _j2_manifest_apply_result.retry_count | default(0) | int > 0
+    stage_name: "{{ item.name }}"
+    resource_identifier: "{{ item.j2_manifest }}"
+    hotloop_retry_metrics: "{{ hotloop_retry_metrics }}"