testing: lints partial fixes

This commit resolves the following lint rules:

- 'key-order[task]'
- 'jinja[spacing]'
- 'name[missing]'
- 'yaml[new-line-at-end-of-file]'
- 'yaml[trailing-spaces]'
- 'risky-shell-pipe'
- 'args[module]'

Also removes the branch constraint to run
the Ansible lint checks from any fork. This
is useful because in each contributor's fork
when pushing specific named branches the checks
will be executed, a future improvement could be
to run all the lints checks from a single environment
like `tox -e linters` instead of consuming the GH action.

Partially-solves: openshift-psap#10

.github/workflows/ansible-lint.yml

@@ -3,11 +3,8 @@ name: Run ansible-lint
 
 # Controls when the action will run.
 on:
-  # Triggers the workflow on push or pull request events but only for the main branch
   pull_request:
-    branches: [main]
   push:
-    branches: [main]
   schedule:
     - cron: '0 */8 * * *'
   # Allows you to run this workflow manually from the Actions tab

config/ansible-lint.yml

@@ -14,11 +14,8 @@ skip_list:
   - 'command-instead-of-module'
   - 'command-instead-of-shell'
   - 'deprecated-local-action'
-  - 'key-order[task]'
-  - 'jinja[spacing]'
   - 'no-free-form'
   - 'chema[meta]'
-  - 'name[missing]'
   - 'var-naming[no-reserved]'
   - 'var-naming[no-role-prefix]'
   - 'var-naming[pattern]'
@@ -29,15 +26,11 @@ skip_list:
   - 'yaml[indentation]'
   - 'yaml[key-duplicates]'
   - 'yaml[line-length]'
-  - 'yaml[new-line-at-end-of-file]'
   - 'yaml[octal-values]'
-  - 'yaml[trailing-spaces]'
   - 'yaml[truthy]'
   - 'name[template]'
   - 'name[casing]'
   - 'risky-file-permissions'
-  - 'risky-shell-pipe'
   - 'ignore-errors'
   - 'no-changed-when'
   - 'fqcn'
-  - 'args[module]'

roles/benchmarking/benchmarking_run_mlperf_ssd/tasks/main.yml

@@ -45,7 +45,8 @@
   command:
     oc create -f "{{ artifact_extra_logs_dir }}/001_pod_run-mlperf-ssd.yml"
 
-- block:
+- name: Make sure the benchmark completes
+  block:
   - name: Wait for the benchmark completion
     command:
       oc get pod/{{ benchmarking_run_mlperf_ssd_name }}

roles/cluster/cluster_capture_environment/tasks/main.yml

@@ -37,8 +37,6 @@
   command:
     git describe HEAD --long --always
   register: git_version
-  args:
-    warn: false # don't warn about using git here
 
 - name: Store ci-artifact version from Git
   copy:
@@ -50,8 +48,6 @@
   command:
     git show --no-patch
   register: git_show
-  args:
-    warn: false # don't warn about using git here
 
 - name: Store ci-artifact last git commit
   copy:

roles/cluster/cluster_create_osd/tasks/main.yml

@@ -121,7 +121,7 @@
     ocm edit machinepool
              {{ cluster_create_osd_machinepool_name }}
              --cluster={{ cluster_create_osd_cluster_name }}
-             --replicas={{ [2, cluster_create_osd_compute_nodes|int] |max }}
+             --replicas={{ [2, cluster_create_osd_compute_nodes | int] | max }}
 
 - name: Wait for the desired worker node count
   shell: |

roles/cluster/cluster_deploy_aws_efs/tasks/aws-efs.yaml

@@ -96,11 +96,11 @@
 
 - name: Populate the tags dict
   set_fact:
-    tags: "{{ tags | default({}) | combine ({ item.key : item.value }) }}"
+    tags: "{{ tags | default({}) | combine({item.key: item.value}) }}"
   with_items:
-      - { 'key': 'Name' , 'value': '{{ cluster_name_cmd.stdout }}'}
-      - { 'key': '{{ cluster_name_tag_cmd.stdout }}' , 'value': 'owned'}
-      - { 'key': 'Purpose', 'value': ''}
+      - {'key': 'Name' , 'value': '{{ cluster_name_cmd.stdout }}'}
+      - {'key': '{{ cluster_name_tag_cmd.stdout }}' , 'value': 'owned'}
+      - {'key': 'Purpose', 'value': ''}
 
 - name: Get the SecurityGroup content
   amazon.aws.ec2_group_info:

roles/cluster/cluster_deploy_ldap/tasks/main.yml

@@ -145,6 +145,7 @@
 
   - name: Get the cluster ID
     shell:
+      set -o pipefail;
       ocm describe cluster "{{ cluster_deploy_ldap_cluster_name }}" --json | jq .id -r
     register: cluster_id_cmd
 

roles/cluster/cluster_deploy_operator/tasks/main.yml

@@ -231,7 +231,8 @@
 - name: Instantiate the Subscription
   command: oc apply -f "{{ artifact_extra_logs_dir }}/src/002_sub.yml"
 
-- block:
+- name: Make sure the InstallPlan is deployed
+  block:
   - name: Find the operator InstallPlan
     command:
       oc get InstallPlan
@@ -276,32 +277,34 @@
     fail: msg="ClusterServiceVersion install not successful ({{ operator_csv_phase.stdout }})"
     when: operator_csv_phase.stdout != "Succeeded"
 
+  rescue:
+  - name: Capture the Catalog Operator logs (debug)
+    shell:
+      oc logs deployment.apps/catalog-operator
+        -n openshift-operator-lifecycle-manager
+        > {{ artifact_extra_logs_dir }}/catalog_operator.log
+    failed_when: false
+
+  - name: Indicate where the Catalog-operator logs have been saved
+    debug: msg="The logs of Catalog Operator have been saved in {{ artifact_extra_logs_dir }}/catalog_operator.log"
+
+  - name: Failed because the operator could not be installed from the CatalogSource
+    fail: msg="Failed because the operator could not be installed from the CatalogSource"
+
   always:
   - name: Store the YAML of the operator CSV that was installed (debug)
     shell:
       oc get ClusterServiceVersion/{{ operator_csv_name }}
          -oyaml
           -n "{{ cluster_deploy_operator_namespace }}"
           > {{ artifact_extra_logs_dir }}/operator_csv.yml
+
   - name: Store the YAML of the subscription
     shell:
       oc get -f "{{ artifact_extra_logs_dir }}/src/002_sub.yml"
          -oyaml
          -n "{{ cluster_deploy_operator_namespace }}"
          > {{ artifact_extra_logs_dir }}/operator_sub.yml
-  rescue:
-  - name: Capture the Catalog Operator logs (debug)
-    shell:
-      oc logs deployment.apps/catalog-operator
-         -n openshift-operator-lifecycle-manager
-         > {{ artifact_extra_logs_dir }}/catalog_operator.log
-    failed_when: false
-
-  - name: Indicate where the Catalog-operator logs have been saved
-    debug: msg="The logs of Catalog Operator have been saved in {{ artifact_extra_logs_dir }}/catalog_operator.log"
-
-  - name: Failed because the operator could not be installed from the CatalogSource
-    fail: msg="Failed because the operator could not be installed from the CatalogSource"
 
 - name: Deploy the operator CustomResource from its ClusterServiceVersion
   include_tasks: deploy_cr.yml

roles/cluster/cluster_ensure_machineset/tasks/main.yml

@@ -1,6 +1,7 @@
 ---
 - name: "Check if the cluster already has a {{ machineset_instance_type }} machineset"
   shell:
+    set -o pipefail;
     oc get machineset -n openshift-machine-api
     {% if machineset_name | length > 0 %}
       -ojson | jq '.items[] | select(.spec.template.spec.providerSpec.value.instanceType=="{{ machineset_instance_type }}" and .metadata.name=="{{ machineset_name }}") | .metadata.name' -r

roles/cluster/cluster_fill_workernodes/tasks/main.yml

@@ -1,5 +1,6 @@
 - name: Get the list of the worker nodes
   shell:
+    set -o pipefail;
     oc get nodes
        -l{{ cluster_fill_workernodes_label_selector }}
        -oname

roles/cluster/cluster_set_project_annotation/tasks/cluster_set.yaml

@@ -55,8 +55,8 @@
     echo "- deleting the project  $test_project_name ..."
     oc delete ns "$test_project_name" >/dev/null
     echo "--> project annotation value: $project_annotation_value"
-    echo "==> expected value:           "'{{ cluster_set_project_annotation_value or "null"}}'
-    [[ "$project_annotation_value" == '{{ cluster_set_project_annotation_value or "null"}}' ]]
+    echo "==> expected value:           "'{{ cluster_set_project_annotation_value or "null" }}'
+    [[ "$project_annotation_value" == '{{ cluster_set_project_annotation_value or "null" }}' ]]
   retries: 120
   delay: 5
   register: wait_project_template_active

roles/cluster/cluster_set_scale/tasks/main.yml

@@ -48,8 +48,8 @@
   register: oc_get_machinesets
   failed_when: not oc_get_machinesets.stdout
 
-- when: current_replicas_sum != scale
-  name: Change all {{ machineset_instance_type }} machinesets replicas to have sum {{ scale }}
+- name: Change all {{ machineset_instance_type }} machinesets replicas to have sum {{ scale }}
+  when: current_replicas_sum != scale
   block:
   - name: Do not downscale any machinesets other than the first one, unless the user used force
     block:
@@ -80,7 +80,8 @@
       oc patch machineset -n openshift-machine-api {{ first_machineset }}
       --patch '{"spec": {"replicas": {{ scale }} }}' --type merge
 
-- block:
+- name: Make sure the machinesets are ready
+  block:
   - name: Wait for all machinesets with type {{ machineset_instance_type }} to be ready
     # This is done by verifying that at the availableReplicas
 
@@ -92,6 +93,7 @@
     #    for more information.
     # 3. Perform some extra formatting for nicer logging
     shell: >-
+      set -o pipefail;
       oc get machinesets -n openshift-machine-api \
       {% if machineset_name | length > 0 %}
              "{{ machineset_name }}" -ojson \
@@ -111,6 +113,10 @@
     retries: 120
     delay: 30
 
+  rescue:
+  - name: Fail because the cluster machineset creation failed
+    fail: msg="Failing because cluster machineset creation failed"
+
   always:
   # info about the 'machines'
   - name: Capture the description of the machines
@@ -150,7 +156,4 @@
     failed_when: false
     loop: "{{ oc_get_machinesets.stdout_lines }}"
 
-  rescue:
 
-  - name: Fail because the cluster machineset creation failed
-    fail: msg="Failing because cluster machineset creation failed"

roles/codeflare/codeflare_cleanup_appwrappers/tasks/main.yml

@@ -1,10 +1,12 @@
 - name: List all the AppWrappers in the namespace
   shell: |
+    set -o pipefail;
     oc get appwrappers -n {{ codeflare_cleanup_appwrappers_namespace }} -ojson | jq '.items[] | .metadata.name + " ==> "+ .status.state' -r > "{{ artifact_extra_logs_dir }}/appwrappers.status"
     oc get appwrappers -n {{ codeflare_cleanup_appwrappers_namespace }} -oyaml > "{{ artifact_extra_logs_dir }}/appwrappers.yaml"
 
 - name: Count the AppWrappers in the namespace
   shell:
+    set -o pipefail;
     oc get appwrappers -n {{ codeflare_cleanup_appwrappers_namespace }} -oname | wc -l
 
 - name: Create a configmap for the beginning of the test timestamp

roles/codeflare/codeflare_generate_mcad_load/tasks/main.yml

@@ -43,6 +43,7 @@
 
 - name: Create the namespace if it does not exist
   shell:
+    set -o pipefail;
     oc create ns "{{ codeflare_generate_mcad_load_namespace }}" -oyaml --dry-run=client | tee "{{ artifact_extra_logs_dir }}/src/namespace.yaml" | oc apply -f-
 
 - name: Create a configmap for the beginning of the test timestamp

roles/entitlement/entitlement_deploy/tasks/main.yml

@@ -31,29 +31,29 @@
 - name: "Deploy RHSM from file '{{ entitlement_rhsm }}'"
   shell:
     set -o pipefail;
-    cat "{{ entitlement_mc_rhsm }}" 
-    | python3 "{{ entitlement_py_apply }}" BASE64_ENCODED_RHSM_FILE @"{{ entitlement_rhsm }}" 
+    cat "{{ entitlement_mc_rhsm }}"
+    | python3 "{{ entitlement_py_apply }}" BASE64_ENCODED_RHSM_FILE @"{{ entitlement_rhsm }}"
     | python3 "{{ entitlement_py_apply }}" MACHINE_CONFIG_ROLE "{{ machine_config_role }}"
     | oc apply -f-
 
 - name: "Deploy the pem and key-pem from file '{{ entitlement_pem }}'"
   shell:
     set -o pipefail;
-    cat "{{ entitlement_mc_pem }}" 
+    cat "{{ entitlement_mc_pem }}"
     | python3 "{{ entitlement_py_apply }}" BASE64_ENCODED_PEM_FILE @"{{ entitlement_pem }}"
     | python3 "{{ entitlement_py_apply }}" MACHINE_CONFIG_ROLE "{{ machine_config_role }}"
     | oc apply -f-
 
 - name: "Deploy the repo CA from file '{{ entitlement_mc_rhsm_ca }}' if requested"
+  when: entitlement_repo_ca | default('', true) | trim != ''
   block:
   - name: Compute the md5sum of the CA file (debug)
     command: md5sum '{{ entitlement_repo_ca }}'
 
   - name: "Deploy the repo CA from file '{{ entitlement_mc_rhsm_ca }}'"
     shell:
       set -o pipefail;
-      cat "{{ entitlement_mc_rhsm_ca }}" 
+      cat "{{ entitlement_mc_rhsm_ca }}"
       | python3 "{{ entitlement_py_apply }}" BASE64_ENCODED_RHSM_CA_FILE @"{{ entitlement_repo_ca }}"
       | python3 "{{ entitlement_py_apply }}" MACHINE_CONFIG_ROLE "{{ machine_config_role }}"
       | oc apply -f-
-  when: entitlement_repo_ca | default('', true) | trim != ''

roles/entitlement/entitlement_test_in_cluster/tasks/main.yml

@@ -26,7 +26,8 @@
 - name: Delete the entitlement tester Pod if it exists
   command: oc delete  --ignore-not-found=true -f "{{ entitlement_tester_pod }}"
 
-- block:
+- name: Make sure the entitlement Pod is created
+  block:
   - name: Create the entitlement tester Pod
     command: oc create -f "{{ entitlement_tester_pod }}"
 

roles/entitlement/entitlement_test_wait_deployment/tasks/main.yml

@@ -9,7 +9,8 @@
     entitlement_retries: "{{ entitlement_nb_wait_retries }}"
   when: entitlement_test_and_wait == 'yes'
 
-- block:
+- name: Make sure the entitlement Pod is deployed
+  block:
   - name: Wait for the entitlement Pod to succeed
     shell: |
       set -o errexit;

roles/gpu_operator/gpu_operator_deploy_from_operatorhub/tasks/deploy_clusterpolicy.yml

@@ -10,7 +10,8 @@
   retries: 15
   delay: 30
 
-- block:
+- name: Make sure the GPU Operator ClusterPolicy is ready
+  block:
   - name: Wait for the GPU Operator ClusterPolicy CRD to appear
     command: oc get crd clusterpolicies.nvidia.com
     register: has_clusterpolicy_crd

roles/gpu_operator/gpu_operator_deploy_from_operatorhub/tasks/deploy_from_catalog.yml

@@ -113,7 +113,8 @@
 - name: Instantiate the OperatorHub subscription
   command: oc create -f "{{ artifact_extra_logs_dir }}/gpu_operator_sub.yml"
 
-- block:
+- name: Install the GPU Operator
+  block:
   - name: Find the GPU Operator OperatorHub InstallPlan
     command:
       oc get InstallPlan

roles/gpu_operator/gpu_operator_get_csv_version/tasks/main.yml

@@ -9,11 +9,15 @@
   - name: Get the gpu-operator subscription package name
     block:
     - name: Count gpu operator subscription candidates
-      shell: oc get subscription --all-namespaces -ojson | jq '[.items[] | select(.spec.name | test("gpu-operator"))] | length'
+      shell:
+        set -o pipefail;
+        oc get subscription --all-namespaces -ojson | jq '[.items[] | select(.spec.name | test("gpu-operator"))] | length'
       register: gpu_subscriptions
       failed_when: gpu_subscriptions.stdout != '1'
     - name: Read the package name from the first gpu-operator subscription
-      shell: oc get subscription -A -ojson | jq '[.items[] | select(.spec.name | test("gpu-operator"))][0].spec.name'
+      shell:
+        set -o pipefail;
+        oc get subscription -A -ojson | jq '[.items[] | select(.spec.name | test("gpu-operator"))][0].spec.name'
       register: gpu_operator_subscription_package_name
   - name: Ensure that there is a CSV for the GPU Operator
     command:

roles/gpu_operator/gpu_operator_run_gpu_burn/tasks/main.yml

@@ -91,13 +91,16 @@
   - name: Ensure that no GPU was faulty
     loop: "{{ gpu_burn_gpu_nodes.stdout_lines }}"
     shell:
+      set -o pipefail;
       oc logs pod/gpu-burn-{{ item }} -n {{ gpu_operator_run_gpu_burn_namespace }} | grep FAULTY
     register: gpu_burn_test_faulty
     failed_when: gpu_burn_test_faulty.rc == 0
 
   always:
   - name: Save the logs of the GPU burn Pods
-    shell: oc logs pod/gpu-burn-{{ item }} -n {{ gpu_operator_run_gpu_burn_namespace }} | grep -o "[^$(printf '\r')]*$"
+    shell:
+      set -o pipefail;
+      oc logs pod/gpu-burn-{{ item }} -n {{ gpu_operator_run_gpu_burn_namespace }} | grep -o "[^$(printf '\r')]*$"
     with_items: "{{ gpu_burn_gpu_nodes.stdout_lines }}"
     failed_when: false
 
@@ -124,4 +127,4 @@
   - name: Delete the src ConfigMap
     command: oc --ignore-not-found=true delete configmap gpu-burn-src -n {{ gpu_operator_run_gpu_burn_namespace }}
     failed_when: false
-    when: not gpu_operator_run_gpu_burn_keep_resources
+    when: not gpu_operator_run_gpu_burn_keep_resources

roles/gpu_operator/gpu_operator_wait_deployment/tasks/main.yml

@@ -78,7 +78,8 @@
     retries: 15
     delay: 60
 
-- block:
+- name: Wait for the nodes labeling
+  block:
   - name: Wait for the gpu-feature-discovery Pod to label the nodes
     command: oc get nodes -l nvidia.com/gpu.count -oname
     register: has_gpu_feature_discovery_labels

roles/gpu_operator/gpu_operator_wait_deployment/tasks/metrics.yml

@@ -1,5 +1,6 @@
 ---
-- block:
+- name: Create the GPU operator requirements
+  block:
   - name: Check if the GPU Operator namespace has the openshift.io/cluster-monitoring label
     shell:
       set -o pipefail;