Merge pull request #13277 from jbradberry/psql-pager

Configure dbshell to use the 'less' pager

.github/workflows/ci.yml

@@ -2,6 +2,7 @@
 name: CI
 env:
   BRANCH: ${{ github.base_ref || 'devel' }}
+  LC_ALL: "C.UTF-8" # prevent ERROR: Ansible could not initialize the preferred locale: unsupported locale setting
 on:
   pull_request:
 jobs:

.github/workflows/devel_images.yml

@@ -1,5 +1,7 @@
 ---
 name: Build/Push Development Images
+env:
+  LC_ALL: "C.UTF-8" # prevent ERROR: Ansible could not initialize the preferred locale: unsupported locale setting
 on:
   push:
     branches:

.github/workflows/e2e_test.yml

@@ -1,5 +1,8 @@
 ---
 name: E2E Tests
+env:
+  LC_ALL: "C.UTF-8" # prevent ERROR: Ansible could not initialize the preferred locale: unsupported locale setting
+
 on:
   pull_request_target:
     types: [labeled]

.github/workflows/feature_branch_deletion.yml

@@ -0,0 +1,26 @@
+---
+name: Feature branch deletion cleanup
+env:
+  LC_ALL: "C.UTF-8" # prevent ERROR: Ansible could not initialize the preferred locale: unsupported locale setting
+on:
+  delete:
+    branches:
+      - feature_**
+jobs:
+  push:
+    runs-on: ubuntu-latest
+    permissions:
+      packages: write
+      contents: read
+    steps:
+      - name: Delete API Schema
+        env:
+          AWS_ACCESS_KEY: ${{ secrets.AWS_ACCESS_KEY }}
+          AWS_SECRET_KEY: ${{ secrets.AWS_SECRET_KEY }}
+          AWS_REGION: 'us-east-1'
+        run: |
+          ansible localhost -c local, -m command -a "{{ ansible_python_interpreter + ' -m pip install boto3'}}"
+          ansible localhost -c local -m aws_s3 \
+            -a "bucket=awx-public-ci-files object=${GITHUB_REF##*/}/schema.json mode=delete permission=public-read"
+
+

.github/workflows/pr_body_check.yml

@@ -13,21 +13,13 @@ jobs:
       packages: write
       contents: read
     steps:
-      - name: Write PR body to a file
-        run: |
-          cat >> pr.body << __SOME_RANDOM_PR_EOF__
-          ${{ github.event.pull_request.body }}
-          __SOME_RANDOM_PR_EOF__
-
-      - name: Display the received body for troubleshooting
-        run: cat pr.body
-
-      # We want to write these out individually just incase the options were joined on a single line
       - name: Check for each of the lines
+        env:
+          PR_BODY: ${{ github.event.pull_request.body }}
         run: |
-          grep "Bug, Docs Fix or other nominal change" pr.body > Z
-          grep "New or Enhanced Feature" pr.body > Y
-          grep "Breaking Change" pr.body > X
+          echo $PR_BODY | grep "Bug, Docs Fix or other nominal change" > Z
+          echo $PR_BODY | grep "New or Enhanced Feature" > Y
+          echo $PR_BODY | grep "Breaking Change" > X
           exit 0
         # We exit 0 and set the shell to prevent the returns from the greps from failing this step
         # See https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#exit-codes-and-error-action-preference

.github/workflows/promote.yml

@@ -1,5 +1,9 @@
 ---
 name: Promote Release
+
+env:
+  LC_ALL: "C.UTF-8" # prevent ERROR: Ansible could not initialize the preferred locale: unsupported locale setting
+
 on:
   release:
     types: [published]

.github/workflows/stage.yml

@@ -1,5 +1,9 @@
 ---
 name: Stage Release
+
+env:
+  LC_ALL: "C.UTF-8" # prevent ERROR: Ansible could not initialize the preferred locale: unsupported locale setting
+
 on:
   workflow_dispatch:
     inputs:

.github/workflows/upload_schema.yml

@@ -1,10 +1,15 @@
 ---
 name: Upload API Schema
+
+env:
+  LC_ALL: "C.UTF-8" # prevent ERROR: Ansible could not initialize the preferred locale: unsupported locale setting
+
 on:
   push:
     branches:
       - devel
       - release_**
+      - feature_**
 jobs:
   push:
     runs-on: ubuntu-latest

MANIFEST.in

@@ -12,7 +12,7 @@ recursive-include awx/plugins *.ps1
 recursive-include requirements *.txt
 recursive-include requirements *.yml
 recursive-include config *
-recursive-include docs/licenses *
+recursive-include licenses *
 recursive-exclude awx devonly.py*
 recursive-exclude awx/api/tests *
 recursive-exclude awx/main/tests *

Makefile

@@ -34,7 +34,7 @@ RECEPTOR_IMAGE ?= quay.io/ansible/receptor:devel
 SRC_ONLY_PKGS ?= cffi,pycparser,psycopg2,twilio
 # These should be upgraded in the AWX and Ansible venv before attempting
 # to install the actual requirements
-VENV_BOOTSTRAP ?= pip==21.2.4 setuptools==58.2.0 setuptools_scm[toml]==6.4.2 wheel==0.36.2
+VENV_BOOTSTRAP ?= pip==21.2.4 setuptools==65.6.3 setuptools_scm[toml]==7.0.5 wheel==0.38.4
 
 NAME ?= awx
 
@@ -85,6 +85,7 @@ clean: clean-ui clean-api clean-awxkit clean-dist
 
 clean-api:
 	rm -rf build $(NAME)-$(VERSION) *.egg-info
+	rm -rf .tox
 	find . -type f -regex ".*\.py[co]$$" -delete
 	find . -type d -name "__pycache__" -delete
 	rm -f awx/awx_test.sqlite3*
@@ -117,7 +118,7 @@ virtualenv_awx:
 		fi; \
 	fi
 
-## Install third-party requirements needed for AWX's environment. 
+## Install third-party requirements needed for AWX's environment.
 # this does not use system site packages intentionally
 requirements_awx: virtualenv_awx
 	if [[ "$(PIP_OPTIONS)" == *"--no-index"* ]]; then \
@@ -181,7 +182,7 @@ collectstatic:
 	@if [ "$(VENV_BASE)" ]; then \
 		. $(VENV_BASE)/awx/bin/activate; \
 	fi; \
-	mkdir -p awx/public/static && $(PYTHON) manage.py collectstatic --clear --noinput > /dev/null 2>&1
+	$(PYTHON) manage.py collectstatic --clear --noinput > /dev/null 2>&1
 
 DEV_RELOAD_COMMAND ?= supervisorctl restart tower-processes:*
 
@@ -377,6 +378,8 @@ clean-ui:
 	rm -rf awx/ui/build
 	rm -rf awx/ui/src/locales/_build
 	rm -rf $(UI_BUILD_FLAG_FILE)
+        # the collectstatic command doesn't like it if this dir doesn't exist.
+	mkdir -p awx/ui/build/static
 
 awx/ui/node_modules:
 	NODE_OPTIONS=--max-old-space-size=6144 $(NPM_BIN) --prefix awx/ui --loglevel warn --force ci
@@ -386,16 +389,14 @@ $(UI_BUILD_FLAG_FILE):
 	$(PYTHON) tools/scripts/compilemessages.py
 	$(NPM_BIN) --prefix awx/ui --loglevel warn run compile-strings
 	$(NPM_BIN) --prefix awx/ui --loglevel warn run build
-	mkdir -p awx/public/static/css
-	mkdir -p awx/public/static/js
-	mkdir -p awx/public/static/media
-	cp -r awx/ui/build/static/css/* awx/public/static/css
-	cp -r awx/ui/build/static/js/* awx/public/static/js
-	cp -r awx/ui/build/static/media/* awx/public/static/media
+	mkdir -p /var/lib/awx/public/static/css
+	mkdir -p /var/lib/awx/public/static/js
+	mkdir -p /var/lib/awx/public/static/media
+	cp -r awx/ui/build/static/css/* /var/lib/awx/public/static/css
+	cp -r awx/ui/build/static/js/* /var/lib/awx/public/static/js
+	cp -r awx/ui/build/static/media/* /var/lib/awx/public/static/media
 	touch $@
 
-
-
 ui-release: $(UI_BUILD_FLAG_FILE)
 
 ui-devel: awx/ui/node_modules
@@ -451,8 +452,9 @@ awx/projects:
 COMPOSE_UP_OPTS ?=
 COMPOSE_OPTS ?=
 CONTROL_PLANE_NODE_COUNT ?= 1
-EXECUTION_NODE_COUNT ?= 2
+EXECUTION_NODE_COUNT ?= 0
 MINIKUBE_CONTAINER_GROUP ?= false
+MINIKUBE_SETUP ?= false # if false, run minikube separately
 EXTRA_SOURCES_ANSIBLE_OPTS ?=
 
 ifneq ($(ADMIN_PASSWORD),)
@@ -461,7 +463,7 @@ endif
 
 docker-compose-sources: .git/hooks/pre-commit
 	@if [ $(MINIKUBE_CONTAINER_GROUP) = true ]; then\
-	    ansible-playbook -i tools/docker-compose/inventory tools/docker-compose-minikube/deploy.yml; \
+	    ansible-playbook -i tools/docker-compose/inventory -e minikube_setup=$(MINIKUBE_SETUP) tools/docker-compose-minikube/deploy.yml; \
 	fi;
 
 	ansible-playbook -i tools/docker-compose/inventory tools/docker-compose/ansible/sources.yml \
@@ -591,7 +593,6 @@ pot: $(UI_BUILD_FLAG_FILE)
 po: $(UI_BUILD_FLAG_FILE)
 	$(NPM_BIN) --prefix awx/ui --loglevel warn run extract-strings -- --clean
 
-LANG = "en_us"
 ## generate API django .pot .po
 messages:
 	@if [ "$(VENV_BASE)" ]; then \
@@ -635,4 +636,4 @@ help/generate:
 		} \
 	} \
 	{ lastLine = $$0 }' $(MAKEFILE_LIST) | sort -u
-	@printf "\n"
+	@printf "\n"

awx/api/serializers.py

@@ -113,7 +113,7 @@ from awx.main.utils import (
 )
 from awx.main.utils.filters import SmartFilter
 from awx.main.utils.named_url_graph import reset_counters
-from awx.main.scheduler.task_manager_models import TaskManagerInstanceGroups, TaskManagerInstances
+from awx.main.scheduler.task_manager_models import TaskManagerModels
 from awx.main.redact import UriCleaner, REPLACE_STR
 
 from awx.main.validators import vars_validate_or_raise
@@ -2221,6 +2221,15 @@ class InventorySourceUpdateSerializer(InventorySourceSerializer):
     class Meta:
         fields = ('can_update',)
 
+    def validate(self, attrs):
+        project = self.instance.source_project
+        if project:
+            failed_reason = project.get_reason_if_failed()
+            if failed_reason:
+                raise serializers.ValidationError(failed_reason)
+
+        return super(InventorySourceUpdateSerializer, self).validate(attrs)
+
 
 class InventoryUpdateSerializer(UnifiedJobSerializer, InventorySourceOptionsSerializer):
 
@@ -4272,17 +4281,10 @@ class JobLaunchSerializer(BaseSerializer):
         # Basic validation - cannot run a playbook without a playbook
         if not template.project:
             errors['project'] = _("A project is required to run a job.")
-        elif template.project.status in ('error', 'failed'):
-            errors['playbook'] = _("Missing a revision to run due to failed project update.")
-
-            latest_update = template.project.project_updates.last()
-            if latest_update is not None and latest_update.failed:
-                failed_validation_tasks = latest_update.project_update_events.filter(
-                    event='runner_on_failed',
-                    play="Perform project signature/checksum verification",
-                )
-                if failed_validation_tasks:
-                    errors['playbook'] = _("Last project update failed due to signature validation failure.")
+        else:
+            failure_reason = template.project.get_reason_if_failed()
+            if failure_reason:
+                errors['playbook'] = failure_reason
 
         # cannot run a playbook without an inventory
         if template.inventory and template.inventory.pending_deletion is True:
@@ -4952,7 +4954,7 @@ class InstanceSerializer(BaseSerializer):
             res['install_bundle'] = self.reverse('api:instance_install_bundle', kwargs={'pk': obj.pk})
         res['peers'] = self.reverse('api:instance_peers_list', kwargs={"pk": obj.pk})
         if self.context['request'].user.is_superuser or self.context['request'].user.is_system_auditor:
-            if obj.node_type != 'hop':
+            if obj.node_type == 'execution':
                 res['health_check'] = self.reverse('api:instance_health_check', kwargs={'pk': obj.pk})
         return res
 
@@ -5038,12 +5040,10 @@ class InstanceHealthCheckSerializer(BaseSerializer):
 class InstanceGroupSerializer(BaseSerializer):
 
     show_capabilities = ['edit', 'delete']
-
+    capacity = serializers.SerializerMethodField()
     consumed_capacity = serializers.SerializerMethodField()
     percent_capacity_remaining = serializers.SerializerMethodField()
-    jobs_running = serializers.IntegerField(
-        help_text=_('Count of jobs in the running or waiting state that ' 'are targeted for this instance group'), read_only=True
-    )
+    jobs_running = serializers.SerializerMethodField()
     jobs_total = serializers.IntegerField(help_text=_('Count of all jobs that target this instance group'), read_only=True)
     instances = serializers.SerializerMethodField()
     is_container_group = serializers.BooleanField(
@@ -5069,6 +5069,22 @@ class InstanceGroupSerializer(BaseSerializer):
         label=_('Policy Instance Minimum'),
         help_text=_("Static minimum number of Instances that will be automatically assign to " "this group when new instances come online."),
     )
+    max_concurrent_jobs = serializers.IntegerField(
+        default=0,
+        min_value=0,
+        required=False,
+        initial=0,
+        label=_('Max Concurrent Jobs'),
+        help_text=_("Maximum number of concurrent jobs to run on a group. When set to zero, no maximum is enforced."),
+    )
+    max_forks = serializers.IntegerField(
+        default=0,
+        min_value=0,
+        required=False,
+        initial=0,
+        label=_('Max Forks'),
+        help_text=_("Maximum number of forks to execute concurrently on a group. When set to zero, no maximum is enforced."),
+    )
     policy_instance_list = serializers.ListField(
         child=serializers.CharField(),
         required=False,
@@ -5090,6 +5106,8 @@ class InstanceGroupSerializer(BaseSerializer):
             "consumed_capacity",
             "percent_capacity_remaining",
             "jobs_running",
+            "max_concurrent_jobs",
+            "max_forks",
             "jobs_total",
             "instances",
             "is_container_group",
@@ -5171,28 +5189,39 @@ class InstanceGroupSerializer(BaseSerializer):
         # Store capacity values (globally computed) in the context
         if 'task_manager_igs' not in self.context:
             instance_groups_queryset = None
-            jobs_qs = UnifiedJob.objects.filter(status__in=('running', 'waiting'))
             if self.parent:  # Is ListView:
                 instance_groups_queryset = self.parent.instance
 
-            instances = TaskManagerInstances(jobs_qs)
-            instance_groups = TaskManagerInstanceGroups(instances_by_hostname=instances, instance_groups_queryset=instance_groups_queryset)
+            tm_models = TaskManagerModels.init_with_consumed_capacity(
+                instance_fields=['uuid', 'version', 'capacity', 'cpu', 'memory', 'managed_by_policy', 'enabled'],
+                instance_groups_queryset=instance_groups_queryset,
+            )
 
-            self.context['task_manager_igs'] = instance_groups
+            self.context['task_manager_igs'] = tm_models.instance_groups
         return self.context['task_manager_igs']
 
     def get_consumed_capacity(self, obj):
         ig_mgr = self.get_ig_mgr()
         return ig_mgr.get_consumed_capacity(obj.name)
 
-    def get_percent_capacity_remaining(self, obj):
-        if not obj.capacity:
-            return 0.0
+    def get_capacity(self, obj):
         ig_mgr = self.get_ig_mgr()
-        return float("{0:.2f}".format((float(ig_mgr.get_remaining_capacity(obj.name)) / (float(obj.capacity))) * 100))
+        return ig_mgr.get_capacity(obj.name)
+
+    def get_percent_capacity_remaining(self, obj):
+        capacity = self.get_capacity(obj)
+        if not capacity:
+            return 0.0
+        consumed_capacity = self.get_consumed_capacity(obj)
+        return float("{0:.2f}".format(((float(capacity) - float(consumed_capacity)) / (float(capacity))) * 100))
 
     def get_instances(self, obj):
-        return obj.instances.count()
+        ig_mgr = self.get_ig_mgr()
+        return len(ig_mgr.get_instances(obj.name))
+
+    def get_jobs_running(self, obj):
+        ig_mgr = self.get_ig_mgr()
+        return ig_mgr.get_jobs_running(obj.name)
 
 
 class ActivityStreamSerializer(BaseSerializer):

awx/api/templates/api/job_template_launch.md

@@ -1,5 +1,5 @@
 Launch a Job Template:
-
+{% ifmeth GET %}
 Make a GET request to this resource to determine if the job_template can be
 launched and whether any passwords are required to launch the job_template.
 The response will include the following fields:
@@ -29,8 +29,8 @@ The response will include the following fields:
 * `inventory_needed_to_start`: Flag indicating the presence of an inventory
   associated with the job template.  If not then one should be supplied when
   launching the job (boolean, read-only)
-
-Make a POST request to this resource to launch the job_template. If any
+{% endifmeth %}
+{% ifmeth POST %}Make a POST request to this resource to launch the job_template. If any
 passwords, inventory, or extra variables (extra_vars) are required, they must
 be passed via POST data, with extra_vars given as a YAML or JSON string and
 escaped parentheses. If the `inventory_needed_to_start` is `True` then the
@@ -41,3 +41,4 @@ are not provided, a 400 status code will be returned.  If the job cannot be
 launched, a 405 status code will be returned. If the provided credential or
 inventory are not allowed to be used by the user, then a 403 status code will
 be returned.
+{% endifmeth %}

awx/api/views/__init__.py

@@ -5,6 +5,7 @@
 import dateutil
 import functools
 import html
+import itertools
 import logging
 import re
 import requests
@@ -20,9 +21,10 @@ from urllib3.exceptions import ConnectTimeoutError
 # Django
 from django.conf import settings
 from django.core.exceptions import FieldError, ObjectDoesNotExist
-from django.db.models import Q, Sum
+from django.db.models import Q, Sum, Count
 from django.db import IntegrityError, ProgrammingError, transaction, connection
 from django.db.models.fields.related import ManyToManyField, ForeignKey
+from django.db.models.functions import Trunc
 from django.shortcuts import get_object_or_404
 from django.utils.safestring import mark_safe
 from django.utils.timezone import now
@@ -47,9 +49,6 @@ from rest_framework import status
 from rest_framework_yaml.parsers import YAMLParser
 from rest_framework_yaml.renderers import YAMLRenderer
 
-# QSStats
-import qsstats
-
 # ANSIConv
 import ansiconv
 
@@ -283,30 +282,50 @@ class DashboardJobsGraphView(APIView):
             success_query = success_query.filter(instance_of=models.ProjectUpdate)
             failed_query = failed_query.filter(instance_of=models.ProjectUpdate)
 
-        success_qss = qsstats.QuerySetStats(success_query, 'finished')
-        failed_qss = qsstats.QuerySetStats(failed_query, 'finished')
-
-        start_date = now()
+        end = now()
+        interval = 'day'
         if period == 'month':
-            end_date = start_date - dateutil.relativedelta.relativedelta(months=1)
-            interval = 'days'
+            start = end - dateutil.relativedelta.relativedelta(months=1)
         elif period == 'two_weeks':
-            end_date = start_date - dateutil.relativedelta.relativedelta(weeks=2)
-            interval = 'days'
+            start = end - dateutil.relativedelta.relativedelta(weeks=2)
         elif period == 'week':
-            end_date = start_date - dateutil.relativedelta.relativedelta(weeks=1)
-            interval = 'days'
+            start = end - dateutil.relativedelta.relativedelta(weeks=1)
         elif period == 'day':
-            end_date = start_date - dateutil.relativedelta.relativedelta(days=1)
-            interval = 'hours'
+            start = end - dateutil.relativedelta.relativedelta(days=1)
+            interval = 'hour'
         else:
             return Response({'error': _('Unknown period "%s"') % str(period)}, status=status.HTTP_400_BAD_REQUEST)
 
         dashboard_data = {"jobs": {"successful": [], "failed": []}}
-        for element in success_qss.time_series(end_date, start_date, interval=interval):
-            dashboard_data['jobs']['successful'].append([time.mktime(element[0].timetuple()), element[1]])
-        for element in failed_qss.time_series(end_date, start_date, interval=interval):
-            dashboard_data['jobs']['failed'].append([time.mktime(element[0].timetuple()), element[1]])
+
+        succ_list = dashboard_data['jobs']['successful']
+        fail_list = dashboard_data['jobs']['failed']
+
+        qs_s = (
+            success_query.filter(finished__range=(start, end))
+            .annotate(d=Trunc('finished', interval, tzinfo=end.tzinfo))
+            .order_by()
+            .values('d')
+            .annotate(agg=Count('id', distinct=True))
+        )
+        data_s = {item['d']: item['agg'] for item in qs_s}
+        qs_f = (
+            failed_query.filter(finished__range=(start, end))
+            .annotate(d=Trunc('finished', interval, tzinfo=end.tzinfo))
+            .order_by()
+            .values('d')
+            .annotate(agg=Count('id', distinct=True))
+        )
+        data_f = {item['d']: item['agg'] for item in qs_f}
+
+        start_date = start.replace(hour=0, minute=0, second=0, microsecond=0)
+        for d in itertools.count():
+            date = start_date + dateutil.relativedelta.relativedelta(days=d)
+            if date > end:
+                break
+            succ_list.append([time.mktime(date.timetuple()), data_s.get(date, 0)])
+            fail_list.append([time.mktime(date.timetuple()), data_f.get(date, 0)])
+
         return Response(dashboard_data)
 
 
@@ -392,8 +411,8 @@ class InstanceHealthCheck(GenericAPIView):
     permission_classes = (IsSystemAdminOrAuditor,)
 
     def get_queryset(self):
+        return super().get_queryset().filter(node_type='execution')
         # FIXME: For now, we don't have a good way of checking the health of a hop node.
-        return super().get_queryset().exclude(node_type='hop')
 
     def get(self, request, *args, **kwargs):
         obj = self.get_object()
@@ -413,9 +432,10 @@ class InstanceHealthCheck(GenericAPIView):
 
             execution_node_health_check.apply_async([obj.hostname])
         else:
-            from awx.main.tasks.system import cluster_node_health_check
-
-            cluster_node_health_check.apply_async([obj.hostname], queue=obj.hostname)
+            return Response(
+                {"error": f"Cannot run a health check on instances of type {obj.node_type}.  Health checks can only be run on execution nodes."},
+                status=status.HTTP_400_BAD_REQUEST,
+            )
         return Response({'msg': f"Health check is running for {obj.hostname}."}, status=status.HTTP_200_OK)
 
 
@@ -2220,6 +2240,8 @@ class InventorySourceUpdateView(RetrieveAPIView):
 
     def post(self, request, *args, **kwargs):
         obj = self.get_object()
+        serializer = self.get_serializer(instance=obj, data=request.data)
+        serializer.is_valid(raise_exception=True)
         if obj.can_update:
             update = obj.update()
             if not update:

awx/locale/es/LC_MESSAGES/django.po

@@ -6237,4 +6237,5 @@ msgstr "%s se está actualizando."
 
 #: awx/ui/urls.py:24
 msgid "This page will refresh when complete."
-msgstr "Esta página se actualizará cuando se complete."
+msgstr "Esta página se actualizará cuando se complete."
+

awx/locale/fr/LC_MESSAGES/django.po

@@ -721,7 +721,7 @@ msgstr "DTSTART valide obligatoire dans rrule. La valeur doit commencer par : DT
 #: awx/api/serializers.py:4657
 msgid ""
 "DTSTART cannot be a naive datetime.  Specify ;TZINFO= or YYYYMMDDTHHMMSSZZ."
-msgstr "DTSTART ne peut correspondre à une DateHeure naïve. Spécifier ;TZINFO= ou YYYYMMDDTHHMMSSZZ."
+msgstr "DTSTART ne peut correspondre à une date-heure naïve. Spécifier ;TZINFO= ou YYYYMMDDTHHMMSSZZ."
 
 #: awx/api/serializers.py:4659
 msgid "Multiple DTSTART is not supported."
@@ -6239,4 +6239,5 @@ msgstr "%s est en cours de mise à niveau."
 
 #: awx/ui/urls.py:24
 msgid "This page will refresh when complete."
-msgstr "Cette page sera rafraîchie une fois terminée."
+msgstr "Cette page sera rafraîchie une fois terminée."
+

awx/locale/nl/LC_MESSAGES/django.po

@@ -6237,4 +6237,5 @@ msgstr "Er wordt momenteel een upgrade van%s geïnstalleerd."
 
 #: awx/ui/urls.py:24
 msgid "This page will refresh when complete."
-msgstr "Deze pagina wordt vernieuwd als hij klaar is."
+msgstr "Deze pagina wordt vernieuwd als hij klaar is."
+

awx/main/analytics/broadcast_websocket.py

@@ -1,7 +1,6 @@
 import datetime
 import asyncio
 import logging
-import aioredis
 import redis
 import re
 
@@ -82,7 +81,7 @@ class BroadcastWebsocketStatsManager:
 
     async def run_loop(self):
         try:
-            redis_conn = await aioredis.create_redis_pool(settings.BROKER_URL)
+            redis_conn = await redis.asyncio.create_redis_pool(settings.BROKER_URL)
             while True:
                 stats_data_str = ''.join(stat.serialize() for stat in self._stats.values())
                 await redis_conn.set(self._redis_key, stats_data_str)

awx/main/analytics/collectors.py

@@ -16,7 +16,7 @@ from awx.conf.license import get_license
 from awx.main.utils import get_awx_version, camelcase_to_underscore, datetime_hook
 from awx.main import models
 from awx.main.analytics import register
-from awx.main.scheduler.task_manager_models import TaskManagerInstances
+from awx.main.scheduler.task_manager_models import TaskManagerModels
 
 """
 This module is used to define metrics collected by awx.main.analytics.gather()
@@ -237,9 +237,8 @@ def projects_by_scm_type(since, **kwargs):
 def instance_info(since, include_hostnames=False, **kwargs):
     info = {}
     # Use same method that the TaskManager does to compute consumed capacity without querying all running jobs for each Instance
-    active_tasks = models.UnifiedJob.objects.filter(status__in=['running', 'waiting']).only('task_impact', 'controller_node', 'execution_node')
-    tm_instances = TaskManagerInstances(active_tasks, instance_fields=['uuid', 'version', 'capacity', 'cpu', 'memory', 'managed_by_policy', 'enabled'])
-    for tm_instance in tm_instances.instances_by_hostname.values():
+    tm_models = TaskManagerModels.init_with_consumed_capacity(instance_fields=['uuid', 'version', 'capacity', 'cpu', 'memory', 'managed_by_policy', 'enabled'])
+    for tm_instance in tm_models.instances.instances_by_hostname.values():
         instance = tm_instance.obj
         instance_info = {
             'uuid': instance.uuid,
@@ -251,6 +250,7 @@ def instance_info(since, include_hostnames=False, **kwargs):
             'enabled': instance.enabled,
             'consumed_capacity': tm_instance.consumed_capacity,
             'remaining_capacity': instance.capacity - tm_instance.consumed_capacity,
+            'node_type': instance.node_type,
         }
         if include_hostnames is True:
             instance_info['hostname'] = instance.hostname

awx/main/analytics/metrics.py

@@ -57,6 +57,7 @@ def metrics():
         [
             'hostname',
             'instance_uuid',
+            'node_type',
         ],
         registry=REGISTRY,
     )
@@ -84,6 +85,7 @@ def metrics():
         [
             'hostname',
             'instance_uuid',
+            'node_type',
         ],
         registry=REGISTRY,
     )
@@ -111,6 +113,7 @@ def metrics():
         [
             'hostname',
             'instance_uuid',
+            'node_type',
         ],
         registry=REGISTRY,
     )
@@ -120,6 +123,7 @@ def metrics():
         [
             'hostname',
             'instance_uuid',
+            'node_type',
         ],
         registry=REGISTRY,
     )
@@ -180,12 +184,13 @@ def metrics():
     instance_data = instance_info(None, include_hostnames=True)
     for uuid, info in instance_data.items():
         hostname = info['hostname']
-        INSTANCE_CAPACITY.labels(hostname=hostname, instance_uuid=uuid).set(instance_data[uuid]['capacity'])
+        node_type = info['node_type']
+        INSTANCE_CAPACITY.labels(hostname=hostname, instance_uuid=uuid, node_type=node_type).set(instance_data[uuid]['capacity'])
         INSTANCE_CPU.labels(hostname=hostname, instance_uuid=uuid).set(instance_data[uuid]['cpu'])
         INSTANCE_MEMORY.labels(hostname=hostname, instance_uuid=uuid).set(instance_data[uuid]['memory'])
-        INSTANCE_CONSUMED_CAPACITY.labels(hostname=hostname, instance_uuid=uuid).set(instance_data[uuid]['consumed_capacity'])
-        INSTANCE_REMAINING_CAPACITY.labels(hostname=hostname, instance_uuid=uuid).set(instance_data[uuid]['remaining_capacity'])
-        INSTANCE_INFO.labels(hostname=hostname, instance_uuid=uuid).info(
+        INSTANCE_CONSUMED_CAPACITY.labels(hostname=hostname, instance_uuid=uuid, node_type=node_type).set(instance_data[uuid]['consumed_capacity'])
+        INSTANCE_REMAINING_CAPACITY.labels(hostname=hostname, instance_uuid=uuid, node_type=node_type).set(instance_data[uuid]['remaining_capacity'])
+        INSTANCE_INFO.labels(hostname=hostname, instance_uuid=uuid, node_type=node_type).info(
             {
                 'enabled': str(instance_data[uuid]['enabled']),
                 'managed_by_policy': str(instance_data[uuid]['managed_by_policy']),

awx/main/analytics/subsystem_metrics.py

@@ -5,7 +5,9 @@ import logging
 
 from django.conf import settings
 from django.apps import apps
+
 from awx.main.consumers import emit_channel_notification
+from awx.main.utils import is_testing
 
 root_key = 'awx_metrics'
 logger = logging.getLogger('awx.main.analytics')
@@ -163,7 +165,7 @@ class Metrics:
         Instance = apps.get_model('main', 'Instance')
         if instance_name:
             self.instance_name = instance_name
-        elif settings.IS_TESTING():
+        elif is_testing():
             self.instance_name = "awx_testing"
         else:
             self.instance_name = Instance.objects.my_hostname()

awx/main/credential_plugins/conjur.py

@@ -1,6 +1,5 @@
 from .plugin import CredentialPlugin, CertFiles, raise_for_status
 
-import base64
 from urllib.parse import urljoin, quote
 
 from django.utils.translation import gettext_lazy as _
@@ -61,7 +60,7 @@ def conjur_backend(**kwargs):
     cacert = kwargs.get('cacert', None)
 
     auth_kwargs = {
-        'headers': {'Content-Type': 'text/plain'},
+        'headers': {'Content-Type': 'text/plain', 'Accept-Encoding': 'base64'},
         'data': api_key,
         'allow_redirects': False,
     }
@@ -69,9 +68,9 @@ def conjur_backend(**kwargs):
     with CertFiles(cacert) as cert:
         # https://www.conjur.org/api.html#authentication-authenticate-post
         auth_kwargs['verify'] = cert
-        resp = requests.post(urljoin(url, '/'.join(['authn', account, username, 'authenticate'])), **auth_kwargs)
+        resp = requests.post(urljoin(url, '/'.join(['api', 'authn', account, username, 'authenticate'])), **auth_kwargs)
     raise_for_status(resp)
-    token = base64.b64encode(resp.content).decode('utf-8')
+    token = resp.content.decode('utf-8')
 
     lookup_kwargs = {
         'headers': {'Authorization': 'Token token="{}"'.format(token)},
@@ -79,9 +78,10 @@ def conjur_backend(**kwargs):
     }
 
     # https://www.conjur.org/api.html#secrets-retrieve-a-secret-get
-    path = urljoin(url, '/'.join(['secrets', account, 'variable', secret_path]))
+    path = urljoin(url, '/'.join(['api', 'secrets', account, 'variable', secret_path]))
     if version:
-        path = '?'.join([path, version])
+        ver = "version={}".format(version)
+        path = '?'.join([path, ver])
 
     with CertFiles(cacert) as cert:
         lookup_kwargs['verify'] = cert
@@ -90,4 +90,4 @@ def conjur_backend(**kwargs):
     return resp.text
 
 
-conjur_plugin = CredentialPlugin('CyberArk Conjur Secret Lookup', inputs=conjur_inputs, backend=conjur_backend)
+conjur_plugin = CredentialPlugin('CyberArk Conjur Secrets Manager Lookup', inputs=conjur_inputs, backend=conjur_backend)

awx/main/dispatch/pool.py

@@ -466,7 +466,7 @@ class AutoscalePool(WorkerPool):
                 task_name = 'unknown'
                 if isinstance(body, dict):
                     task_name = body.get('task')
-                logger.warn(f'Workers maxed, queuing {task_name}, load: {sum(len(w.managed_tasks) for w in self.workers)} / {len(self.workers)}')
+                logger.warning(f'Workers maxed, queuing {task_name}, load: {sum(len(w.managed_tasks) for w in self.workers)} / {len(self.workers)}')
                 return super(AutoscalePool, self).write(preferred_queue, body)
         except Exception:
             for conn in connections.all():

awx/main/dispatch/publish.py

@@ -1,14 +1,13 @@
 import inspect
 import logging
-import sys
 import json
 import time
 from uuid import uuid4
 
-from django.conf import settings
 from django_guid import get_guid
 
 from . import pg_bus_conn
+from awx.main.utils import is_testing
 
 logger = logging.getLogger('awx.main.dispatch')
 
@@ -93,7 +92,7 @@ class task:
                 obj.update(**kw)
                 if callable(queue):
                     queue = queue()
-                if not settings.IS_TESTING(sys.argv):
+                if not is_testing():
                     with pg_bus_conn() as conn:
                         conn.notify(queue, json.dumps(obj))
                 return (obj, queue)

awx/main/management/commands/provision_instance.py

@@ -38,7 +38,14 @@ class Command(BaseCommand):
             (changed, instance) = Instance.objects.register(ip_address=os.environ.get('MY_POD_IP'), node_type='control', uuid=settings.SYSTEM_UUID)
             RegisterQueue(settings.DEFAULT_CONTROL_PLANE_QUEUE_NAME, 100, 0, [], is_container_group=False).register()
             RegisterQueue(
-                settings.DEFAULT_EXECUTION_QUEUE_NAME, 100, 0, [], is_container_group=True, pod_spec_override=settings.DEFAULT_EXECUTION_QUEUE_POD_SPEC_OVERRIDE
+                settings.DEFAULT_EXECUTION_QUEUE_NAME,
+                100,
+                0,
+                [],
+                is_container_group=True,
+                pod_spec_override=settings.DEFAULT_EXECUTION_QUEUE_POD_SPEC_OVERRIDE,
+                max_forks=settings.DEFAULT_EXECUTION_QUEUE_MAX_FORKS,
+                max_concurrent_jobs=settings.DEFAULT_EXECUTION_QUEUE_MAX_CONCURRENT_JOBS,
             ).register()
         else:
             (changed, instance) = Instance.objects.register(hostname=hostname, node_type=node_type, uuid=uuid)

awx/main/management/commands/register_queue.py

@@ -17,7 +17,9 @@ class InstanceNotFound(Exception):
 
 
 class RegisterQueue:
-    def __init__(self, queuename, instance_percent, inst_min, hostname_list, is_container_group=None, pod_spec_override=None):
+    def __init__(
+        self, queuename, instance_percent, inst_min, hostname_list, is_container_group=None, pod_spec_override=None, max_forks=None, max_concurrent_jobs=None
+    ):
         self.instance_not_found_err = None
         self.queuename = queuename
         self.instance_percent = instance_percent
@@ -25,6 +27,8 @@ class RegisterQueue:
         self.hostname_list = hostname_list
         self.is_container_group = is_container_group
         self.pod_spec_override = pod_spec_override
+        self.max_forks = max_forks
+        self.max_concurrent_jobs = max_concurrent_jobs
 
     def get_create_update_instance_group(self):
         created = False
@@ -45,6 +49,14 @@ class RegisterQueue:
             ig.pod_spec_override = self.pod_spec_override
             changed = True
 
+        if self.max_forks and (ig.max_forks != self.max_forks):
+            ig.max_forks = self.max_forks
+            changed = True
+
+        if self.max_concurrent_jobs and (ig.max_concurrent_jobs != self.max_concurrent_jobs):
+            ig.max_concurrent_jobs = self.max_concurrent_jobs
+            changed = True
+
         if changed:
             ig.save()
 

awx/main/migrations/0164_remove_inventorysource_update_on_project_update.py

@@ -1,24 +1,14 @@
 # Generated by Django 3.2.13 on 2022-06-21 21:29
 
 from django.db import migrations
-import logging
-
-logger = logging.getLogger("awx")
 
 
 def forwards(apps, schema_editor):
     InventorySource = apps.get_model('main', 'InventorySource')
-    sources = InventorySource.objects.filter(update_on_project_update=True)
-    for src in sources:
-        if src.update_on_launch == False:
-            src.update_on_launch = True
-            src.save(update_fields=['update_on_launch'])
-            logger.info(f"Setting update_on_launch to True for {src}")
-        proj = src.source_project
-        if proj and proj.scm_update_on_launch is False:
-            proj.scm_update_on_launch = True
-            proj.save(update_fields=['scm_update_on_launch'])
-            logger.warning(f"Setting scm_update_on_launch to True for {proj}")
+    InventorySource.objects.filter(update_on_project_update=True).update(update_on_launch=True)
+
+    Project = apps.get_model('main', 'Project')
+    Project.objects.filter(scm_inventory_sources__update_on_project_update=True).update(scm_update_on_launch=True)
 
 
 class Migration(migrations.Migration):

awx/main/migrations/0173_instancegroup_max_limits.py

@@ -0,0 +1,23 @@
+# Generated by Django 3.2.13 on 2022-10-24 18:22
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('main', '0172_prevent_instance_fallback'),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name='instancegroup',
+            name='max_concurrent_jobs',
+            field=models.IntegerField(default=0, help_text='Maximum number of concurrent jobs to run on this group. Zero means no limit.'),
+        ),
+        migrations.AddField(
+            model_name='instancegroup',
+            name='max_forks',
+            field=models.IntegerField(default=0, help_text='Max forks to execute on this group. Zero means no limit.'),
+        ),
+    ]

awx/main/models/ha.py

@@ -233,11 +233,12 @@ class Instance(HasPolicyEditsMixin, BaseModel):
         if not isinstance(vargs.get('grace_period'), int):
             vargs['grace_period'] = 60  # grace period of 60 minutes, need to set because CLI default will not take effect
         if 'exclude_strings' not in vargs and vargs.get('file_pattern'):
-            active_pks = list(
-                UnifiedJob.objects.filter(
-                    (models.Q(execution_node=self.hostname) | models.Q(controller_node=self.hostname)) & models.Q(status__in=('running', 'waiting'))
-                ).values_list('pk', flat=True)
-            )
+            active_job_qs = UnifiedJob.objects.filter(status__in=('running', 'waiting'))
+            if self.node_type == 'execution':
+                active_job_qs = active_job_qs.filter(execution_node=self.hostname)
+            else:
+                active_job_qs = active_job_qs.filter(controller_node=self.hostname)
+            active_pks = list(active_job_qs.values_list('pk', flat=True))
             if active_pks:
                 vargs['exclude_strings'] = [JOB_FOLDER_PREFIX % job_id for job_id in active_pks]
         if 'remove_images' in vargs or 'image_prune' in vargs:
@@ -378,6 +379,8 @@ class InstanceGroup(HasPolicyEditsMixin, BaseModel, RelatedJobsMixin):
             default='',
         )
     )
+    max_concurrent_jobs = models.IntegerField(default=0, help_text=_("Maximum number of concurrent jobs to run on this group. Zero means no limit."))
+    max_forks = models.IntegerField(default=0, help_text=_("Max forks to execute on this group. Zero means no limit."))
     policy_instance_percentage = models.IntegerField(default=0, help_text=_("Percentage of Instances to automatically assign to this group"))
     policy_instance_minimum = models.IntegerField(default=0, help_text=_("Static minimum number of Instances to automatically assign to this group"))
     policy_instance_list = JSONBlob(
@@ -391,6 +394,8 @@ class InstanceGroup(HasPolicyEditsMixin, BaseModel, RelatedJobsMixin):
 
     @property
     def capacity(self):
+        if self.is_container_group:
+            return self.max_forks
         return sum(inst.capacity for inst in self.instances.all())
 
     @property

awx/main/models/inventory.py

@@ -247,6 +247,19 @@ class Inventory(CommonModelNameNotUnique, ResourceMixin, RelatedJobsMixin):
         return (number, step)
 
     def get_sliced_hosts(self, host_queryset, slice_number, slice_count):
+        """
+        Returns a slice of Hosts given a slice number and total slice count, or
+        the original queryset if slicing is not requested.
+
+        NOTE: If slicing is performed, this will return a List[Host] with the
+        resulting slice. If slicing is not performed it will return the
+        original queryset (not evaluating it or forcing it to a list). This
+        puts the burden on the caller to check the resulting type. This is
+        non-ideal because it's easy to get wrong, but I think the only way
+        around it is to force the queryset which has memory implications for
+        large inventories.
+        """
+
         if slice_count > 1 and slice_number > 0:
             offset = slice_number - 1
             host_queryset = host_queryset[offset::slice_count]
@@ -554,17 +567,6 @@ class Host(CommonModelNameNotUnique, RelatedJobsMixin):
     # Use .job_host_summaries.all() to get jobs affecting this host.
     # Use .job_events.all() to get events affecting this host.
 
-    '''
-    We don't use timestamp, but we may in the future.
-    '''
-
-    def update_ansible_facts(self, module, facts, timestamp=None):
-        if module == "ansible":
-            self.ansible_facts.update(facts)
-        else:
-            self.ansible_facts[module] = facts
-        self.save()
-
     def get_effective_host_name(self):
         """
         Return the name of the host that will be used in actual ansible

awx/main/models/jobs.py

@@ -15,6 +15,7 @@ from urllib.parse import urljoin
 from django.conf import settings
 from django.core.exceptions import ValidationError
 from django.db import models
+from django.db.models.query import QuerySet
 
 # from django.core.cache import cache
 from django.utils.encoding import smart_str
@@ -43,7 +44,7 @@ from awx.main.models.notifications import (
     NotificationTemplate,
     JobNotificationMixin,
 )
-from awx.main.utils import parse_yaml_or_json, getattr_dne, NullablePromptPseudoField, polymorphic
+from awx.main.utils import parse_yaml_or_json, getattr_dne, NullablePromptPseudoField, polymorphic, log_excess_runtime
 from awx.main.fields import ImplicitRoleField, AskForField, JSONBlob, OrderedManyToManyField
 from awx.main.models.mixins import (
     ResourceMixin,
@@ -844,22 +845,35 @@ class Job(UnifiedJob, JobOptions, SurveyJobMixin, JobNotificationMixin, TaskMana
     def get_notification_friendly_name(self):
         return "Job"
 
-    def _get_inventory_hosts(self, only=['name', 'ansible_facts', 'ansible_facts_modified', 'modified', 'inventory_id']):
+    def _get_inventory_hosts(self, only=('name', 'ansible_facts', 'ansible_facts_modified', 'modified', 'inventory_id'), **filters):
+        """Return value is an iterable for the relevant hosts for this job"""
         if not self.inventory:
             return []
         host_queryset = self.inventory.hosts.only(*only)
-        return self.inventory.get_sliced_hosts(host_queryset, self.job_slice_number, self.job_slice_count)
+        if filters:
+            host_queryset = host_queryset.filter(**filters)
+        host_queryset = self.inventory.get_sliced_hosts(host_queryset, self.job_slice_number, self.job_slice_count)
+        if isinstance(host_queryset, QuerySet):
+            return host_queryset.iterator()
+        return host_queryset
 
-    def start_job_fact_cache(self, destination, modification_times, timeout=None):
+    @log_excess_runtime(logger, debug_cutoff=0.01, msg='Job {job_id} host facts prepared for {written_ct} hosts, took {delta:.3f} s', add_log_data=True)
+    def start_job_fact_cache(self, destination, log_data, timeout=None):
         self.log_lifecycle("start_job_fact_cache")
+        log_data['job_id'] = self.id
+        log_data['written_ct'] = 0
         os.makedirs(destination, mode=0o700)
-        hosts = self._get_inventory_hosts()
+
         if timeout is None:
             timeout = settings.ANSIBLE_FACT_CACHE_TIMEOUT
         if timeout > 0:
             # exclude hosts with fact data older than `settings.ANSIBLE_FACT_CACHE_TIMEOUT seconds`
             timeout = now() - datetime.timedelta(seconds=timeout)
-            hosts = hosts.filter(ansible_facts_modified__gte=timeout)
+            hosts = self._get_inventory_hosts(ansible_facts_modified__gte=timeout)
+        else:
+            hosts = self._get_inventory_hosts()
+
+        last_filepath_written = None
         for host in hosts:
             filepath = os.sep.join(map(str, [destination, host.name]))
             if not os.path.realpath(filepath).startswith(destination):
@@ -869,23 +883,38 @@ class Job(UnifiedJob, JobOptions, SurveyJobMixin, JobNotificationMixin, TaskMana
                 with codecs.open(filepath, 'w', encoding='utf-8') as f:
                     os.chmod(f.name, 0o600)
                     json.dump(host.ansible_facts, f)
+                    log_data['written_ct'] += 1
+                    last_filepath_written = filepath
             except IOError:
                 system_tracking_logger.error('facts for host {} could not be cached'.format(smart_str(host.name)))
                 continue
-            # make note of the time we wrote the file so we can check if it changed later
-            modification_times[filepath] = os.path.getmtime(filepath)
+        # make note of the time we wrote the last file so we can check if any file changed later
+        if last_filepath_written:
+            return os.path.getmtime(last_filepath_written)
+        return None
 
-    def finish_job_fact_cache(self, destination, modification_times):
+    @log_excess_runtime(
+        logger,
+        debug_cutoff=0.01,
+        msg='Job {job_id} host facts: updated {updated_ct}, cleared {cleared_ct}, unchanged {unmodified_ct}, took {delta:.3f} s',
+        add_log_data=True,
+    )
+    def finish_job_fact_cache(self, destination, facts_write_time, log_data):
         self.log_lifecycle("finish_job_fact_cache")
+        log_data['job_id'] = self.id
+        log_data['updated_ct'] = 0
+        log_data['unmodified_ct'] = 0
+        log_data['cleared_ct'] = 0
+        hosts_to_update = []
         for host in self._get_inventory_hosts():
             filepath = os.sep.join(map(str, [destination, host.name]))
             if not os.path.realpath(filepath).startswith(destination):
                 system_tracking_logger.error('facts for host {} could not be cached'.format(smart_str(host.name)))
                 continue
             if os.path.exists(filepath):
-                # If the file changed since we wrote it pre-playbook run...
+                # If the file changed since we wrote the last facts file, pre-playbook run...
                 modified = os.path.getmtime(filepath)
-                if modified > modification_times.get(filepath, 0):
+                if (not facts_write_time) or modified > facts_write_time:
                     with codecs.open(filepath, 'r', encoding='utf-8') as f:
                         try:
                             ansible_facts = json.load(f)
@@ -893,7 +922,7 @@ class Job(UnifiedJob, JobOptions, SurveyJobMixin, JobNotificationMixin, TaskMana
                             continue
                         host.ansible_facts = ansible_facts
                         host.ansible_facts_modified = now()
-                        host.save(update_fields=['ansible_facts', 'ansible_facts_modified'])
+                        hosts_to_update.append(host)
                         system_tracking_logger.info(
                             'New fact for inventory {} host {}'.format(smart_str(host.inventory.name), smart_str(host.name)),
                             extra=dict(
@@ -904,12 +933,21 @@ class Job(UnifiedJob, JobOptions, SurveyJobMixin, JobNotificationMixin, TaskMana
                                 job_id=self.id,
                             ),
                         )
+                        log_data['updated_ct'] += 1
+                else:
+                    log_data['unmodified_ct'] += 1
             else:
                 # if the file goes missing, ansible removed it (likely via clear_facts)
                 host.ansible_facts = {}
                 host.ansible_facts_modified = now()
+                hosts_to_update.append(host)
                 system_tracking_logger.info('Facts cleared for inventory {} host {}'.format(smart_str(host.inventory.name), smart_str(host.name)))
-                host.save()
+                log_data['cleared_ct'] += 1
+            if len(hosts_to_update) > 100:
+                self.inventory.hosts.bulk_update(hosts_to_update, ['ansible_facts', 'ansible_facts_modified'])
+                hosts_to_update = []
+        if hosts_to_update:
+            self.inventory.hosts.bulk_update(hosts_to_update, ['ansible_facts', 'ansible_facts_modified'])
 
 
 class LaunchTimeConfigBase(BaseModel):

awx/main/models/projects.py

@@ -471,6 +471,29 @@ class Project(UnifiedJobTemplate, ProjectOptions, ResourceMixin, CustomVirtualEn
     def get_absolute_url(self, request=None):
         return reverse('api:project_detail', kwargs={'pk': self.pk}, request=request)
 
+    def get_reason_if_failed(self):
+        """
+        If the project is in a failed or errored state, return a human-readable
+        error message explaining why. Otherwise return None.
+
+        This is used during validation in the serializer and also by
+        RunProjectUpdate/RunInventoryUpdate.
+        """
+
+        if self.status not in ('error', 'failed'):
+            return None
+
+        latest_update = self.project_updates.last()
+        if latest_update is not None and latest_update.failed:
+            failed_validation_tasks = latest_update.project_update_events.filter(
+                event='runner_on_failed',
+                play="Perform project signature/checksum verification",
+            )
+            if failed_validation_tasks:
+                return _("Last project update failed due to signature validation failure.")
+
+        return _("Missing a revision to run due to failed project update.")
+
     '''
     RelatedJobsMixin
     '''

awx/main/models/unified_jobs.py

@@ -1351,12 +1351,12 @@ class UnifiedJob(
                 if required in defined_fields and not credential.has_input(required):
                     missing_credential_inputs.append(required)
 
-        if missing_credential_inputs:
-            self.job_explanation = '{} cannot start because Credential {} does not provide one or more required fields ({}).'.format(
-                self._meta.verbose_name.title(), credential.name, ', '.join(sorted(missing_credential_inputs))
-            )
-            self.save(update_fields=['job_explanation'])
-            return (False, None)
+            if missing_credential_inputs:
+                self.job_explanation = '{} cannot start because Credential {} does not provide one or more required fields ({}).'.format(
+                    self._meta.verbose_name.title(), credential.name, ', '.join(sorted(missing_credential_inputs))
+                )
+                self.save(update_fields=['job_explanation'])
+                return (False, None)
 
         needed = self.get_passwords_needed_to_start()
         try:

awx/main/notifications/webhook_backend.py

@@ -5,9 +5,6 @@ import json
 import logging
 import requests
 
-from django.utils.encoding import smart_str
-from django.utils.translation import gettext_lazy as _
-
 from awx.main.notifications.base import AWXBaseEmailBackend
 from awx.main.utils import get_awx_http_client_headers
 from awx.main.notifications.custom_notification_base import CustomNotificationBase
@@ -17,6 +14,8 @@ logger = logging.getLogger('awx.main.notifications.webhook_backend')
 
 class WebhookBackend(AWXBaseEmailBackend, CustomNotificationBase):
 
+    MAX_RETRIES = 5
+
     init_parameters = {
         "url": {"label": "Target URL", "type": "string"},
         "http_method": {"label": "HTTP Method", "type": "string", "default": "POST"},
@@ -64,20 +63,67 @@ class WebhookBackend(AWXBaseEmailBackend, CustomNotificationBase):
         if self.http_method.lower() not in ['put', 'post']:
             raise ValueError("HTTP method must be either 'POST' or 'PUT'.")
         chosen_method = getattr(requests, self.http_method.lower(), None)
+
         for m in messages:
+
             auth = None
             if self.username or self.password:
                 auth = (self.username, self.password)
-            r = chosen_method(
-                "{}".format(m.recipients()[0]),
-                auth=auth,
-                data=json.dumps(m.body, ensure_ascii=False).encode('utf-8'),
-                headers=dict(list(get_awx_http_client_headers().items()) + list((self.headers or {}).items())),
-                verify=(not self.disable_ssl_verification),
-            )
-            if r.status_code >= 400:
-                logger.error(smart_str(_("Error sending notification webhook: {}").format(r.status_code)))
+
+            # the constructor for EmailMessage - https://docs.djangoproject.com/en/4.1/_modules/django/core/mail/message will turn an empty dictionary to an empty string
+            # sometimes an empty dict is intentional and we added this conditional to enforce that
+            if not m.body:
+                m.body = {}
+
+            url = str(m.recipients()[0])
+            data = json.dumps(m.body, ensure_ascii=False).encode('utf-8')
+            headers = {**(get_awx_http_client_headers()), **(self.headers or {})}
+
+            err = None
+
+            for retries in range(self.MAX_RETRIES):
+
+                # Sometimes we hit redirect URLs. We must account for this. We still extract the redirect URL from the response headers and try again. Max retires == 5
+                resp = chosen_method(
+                    url=url,
+                    auth=auth,
+                    data=data,
+                    headers=headers,
+                    verify=(not self.disable_ssl_verification),
+                    allow_redirects=False,  # override default behaviour for redirects
+                )
+
+                # either success or error reached if this conditional fires
+                if resp.status_code not in [301, 307]:
+                    break
+
+                # we've hit a redirect. extract the redirect URL out of the first response header and try again
+                logger.warning(
+                    f"Received a {resp.status_code} from {url}, trying to reach redirect url {resp.headers.get('Location', None)}; attempt #{retries+1}"
+                )
+
+                # take the first redirect URL in the response header and try that
+                url = resp.headers.get("Location", None)
+
+                if url is None:
+                    err = f"Webhook notification received redirect to a blank URL from {url}. Response headers={resp.headers}"
+                    break
+            else:
+                # no break condition in the loop encountered; therefore we have hit the maximum number of retries
+                err = f"Webhook notification max number of retries [{self.MAX_RETRIES}] exceeded. Failed to send webhook notification to {url}"
+
+            if resp.status_code >= 400:
+                err = f"Error sending webhook notification: {resp.status_code}"
+
+            # log error message
+            if err:
+                logger.error(err)
                 if not self.fail_silently:
-                    raise Exception(smart_str(_("Error sending notification webhook: {}").format(r.status_code)))
-            sent_messages += 1
+                    raise Exception(err)
+
+            # no errors were encountered therefore we successfully sent off the notification webhook
+            if resp.status_code in range(200, 299):
+                logger.debug(f"Notification webhook successfully sent to {url}. Received {resp.status_code}")
+                sent_messages += 1
+
         return sent_messages

awx/main/registrar.py

@@ -3,6 +3,8 @@
 
 from django.db.models.signals import pre_save, post_save, pre_delete, m2m_changed
 
+from taggit.managers import TaggableManager
+
 
 class ActivityStreamRegistrar(object):
     def __init__(self):
@@ -19,6 +21,8 @@ class ActivityStreamRegistrar(object):
             pre_delete.connect(activity_stream_delete, sender=model, dispatch_uid=str(self.__class__) + str(model) + "_delete")
 
             for m2mfield in model._meta.many_to_many:
+                if isinstance(m2mfield, TaggableManager):
+                    continue  # Special case for taggit app
                 try:
                     m2m_attr = getattr(model, m2mfield.name)
                     m2m_changed.connect(

awx/main/routing.py

@@ -27,8 +27,8 @@ class AWXProtocolTypeRouter(ProtocolTypeRouter):
 
 
 websocket_urlpatterns = [
-    re_path(r'websocket/$', consumers.EventConsumer),
-    re_path(r'websocket/broadcast/$', consumers.BroadcastConsumer),
+    re_path(r'websocket/$', consumers.EventConsumer.as_asgi()),
+    re_path(r'websocket/broadcast/$', consumers.BroadcastConsumer.as_asgi()),
 ]
 
 application = AWXProtocolTypeRouter(

awx/main/scheduler/task_manager.py

@@ -39,12 +39,11 @@ from awx.main.utils import (
     ScheduleTaskManager,
     ScheduleWorkflowManager,
 )
-from awx.main.utils.common import task_manager_bulk_reschedule
+from awx.main.utils.common import task_manager_bulk_reschedule, is_testing
 from awx.main.signals import disable_activity_stream
 from awx.main.constants import ACTIVE_STATES
 from awx.main.scheduler.dependency_graph import DependencyGraph
-from awx.main.scheduler.task_manager_models import TaskManagerInstances
-from awx.main.scheduler.task_manager_models import TaskManagerInstanceGroups
+from awx.main.scheduler.task_manager_models import TaskManagerModels
 import awx.main.analytics.subsystem_metrics as s_metrics
 from awx.main.utils import decrypt_field
 
@@ -71,7 +70,12 @@ class TaskBase:
         # is called later.
         self.subsystem_metrics = s_metrics.Metrics(auto_pipe_execute=False)
         self.start_time = time.time()
+
+        # We want to avoid calling settings in loops, so cache these settings at init time
         self.start_task_limit = settings.START_TASK_LIMIT
+        self.task_manager_timeout = settings.TASK_MANAGER_TIMEOUT
+        self.control_task_impact = settings.AWX_CONTROL_NODE_TASK_IMPACT
+
         for m in self.subsystem_metrics.METRICS:
             if m.startswith(self.prefix):
                 self.subsystem_metrics.set(m, 0)
@@ -79,7 +83,7 @@ class TaskBase:
     def timed_out(self):
         """Return True/False if we have met or exceeded the timeout for the task manager."""
         elapsed = time.time() - self.start_time
-        if elapsed >= settings.TASK_MANAGER_TIMEOUT:
+        if elapsed >= self.task_manager_timeout:
             logger.warning(f"{self.prefix} manager has run for {elapsed} which is greater than TASK_MANAGER_TIMEOUT of {settings.TASK_MANAGER_TIMEOUT}.")
             return True
         return False
@@ -97,7 +101,7 @@ class TaskBase:
         self.all_tasks = [t for t in qs]
 
     def record_aggregate_metrics(self, *args):
-        if not settings.IS_TESTING():
+        if not is_testing():
             # increment task_manager_schedule_calls regardless if the other
             # metrics are recorded
             s_metrics.Metrics(auto_pipe_execute=True).inc(f"{self.prefix}__schedule_calls", 1)
@@ -471,9 +475,8 @@ class TaskManager(TaskBase):
         Init AFTER we know this instance of the task manager will run because the lock is acquired.
         """
         self.dependency_graph = DependencyGraph()
-        self.instances = TaskManagerInstances(self.all_tasks)
-        self.instance_groups = TaskManagerInstanceGroups(instances_by_hostname=self.instances)
-        self.controlplane_ig = self.instance_groups.controlplane_ig
+        self.tm_models = TaskManagerModels()
+        self.controlplane_ig = self.tm_models.instance_groups.controlplane_ig
 
     def job_blocked_by(self, task):
         # TODO: I'm not happy with this, I think blocking behavior should be decided outside of the dependency graph
@@ -505,7 +508,15 @@ class TaskManager(TaskBase):
 
     @timeit
     def start_task(self, task, instance_group, dependent_tasks=None, instance=None):
+        # Just like for process_running_tasks, add the job to the dependency graph and
+        # ask the TaskManagerInstanceGroups object to update consumed capacity on all
+        # implicated instances and container groups.
         self.dependency_graph.add_job(task)
+        if instance_group is not None:
+            task.instance_group = instance_group
+        # We need the instance group assigned to correctly account for container group max_concurrent_jobs and max_forks
+        self.tm_models.consume_capacity(task)
+
         self.subsystem_metrics.inc(f"{self.prefix}_tasks_started", 1)
         self.start_task_limit -= 1
         if self.start_task_limit == 0:
@@ -513,12 +524,6 @@ class TaskManager(TaskBase):
             ScheduleTaskManager().schedule()
         from awx.main.tasks.system import handle_work_error, handle_work_success
 
-        # update capacity for control node and execution node
-        if task.controller_node:
-            self.instances[task.controller_node].consume_capacity(settings.AWX_CONTROL_NODE_TASK_IMPACT)
-        if task.execution_node:
-            self.instances[task.execution_node].consume_capacity(task.task_impact)
-
         dependent_tasks = dependent_tasks or []
 
         task_actual = {
@@ -546,7 +551,6 @@ class TaskManager(TaskBase):
                 ScheduleWorkflowManager().schedule()
             # at this point we already have control/execution nodes selected for the following cases
             else:
-                task.instance_group = instance_group
                 execution_node_msg = f' and execution node {task.execution_node}' if task.execution_node else ''
                 logger.debug(
                     f'Submitting job {task.log_format} controlled by {task.controller_node} to instance group {instance_group.name}{execution_node_msg}.'
@@ -580,6 +584,7 @@ class TaskManager(TaskBase):
             if type(task) is WorkflowJob:
                 ScheduleWorkflowManager().schedule()
             self.dependency_graph.add_job(task)
+            self.tm_models.consume_capacity(task)
 
     @timeit
     def process_pending_tasks(self, pending_tasks):
@@ -611,11 +616,11 @@ class TaskManager(TaskBase):
 
             # Determine if there is control capacity for the task
             if task.capacity_type == 'control':
-                control_impact = task.task_impact + settings.AWX_CONTROL_NODE_TASK_IMPACT
+                control_impact = task.task_impact + self.control_task_impact
             else:
-                control_impact = settings.AWX_CONTROL_NODE_TASK_IMPACT
-            control_instance = self.instance_groups.fit_task_to_most_remaining_capacity_instance(
-                task, instance_group_name=settings.DEFAULT_CONTROL_PLANE_QUEUE_NAME, impact=control_impact, capacity_type='control'
+                control_impact = self.control_task_impact
+            control_instance = self.tm_models.instance_groups.fit_task_to_most_remaining_capacity_instance(
+                task, instance_group_name=self.controlplane_ig.name, impact=control_impact, capacity_type='control'
             )
             if not control_instance:
                 self.task_needs_capacity(task, tasks_to_update_job_explanation)
@@ -626,15 +631,19 @@ class TaskManager(TaskBase):
 
             # All task.capacity_type == 'control' jobs should run on control plane, no need to loop over instance groups
             if task.capacity_type == 'control':
+                if not self.tm_models.instance_groups[self.controlplane_ig.name].has_remaining_capacity(control_impact=True):
+                    continue
                 task.execution_node = control_instance.hostname
-                execution_instance = self.instances[control_instance.hostname].obj
+                execution_instance = self.tm_models.instances[control_instance.hostname].obj
                 task.log_lifecycle("controller_node_chosen")
                 task.log_lifecycle("execution_node_chosen")
                 self.start_task(task, self.controlplane_ig, task.get_jobs_fail_chain(), execution_instance)
                 found_acceptable_queue = True
                 continue
 
-            for instance_group in self.instance_groups.get_instance_groups_from_task_cache(task):
+            for instance_group in self.tm_models.instance_groups.get_instance_groups_from_task_cache(task):
+                if not self.tm_models.instance_groups[instance_group.name].has_remaining_capacity(task):
+                    continue
                 if instance_group.is_container_group:
                     self.start_task(task, instance_group, task.get_jobs_fail_chain(), None)
                     found_acceptable_queue = True
@@ -642,9 +651,9 @@ class TaskManager(TaskBase):
 
                 # at this point we know the instance group is NOT a container group
                 # because if it was, it would have started the task and broke out of the loop.
-                execution_instance = self.instance_groups.fit_task_to_most_remaining_capacity_instance(
+                execution_instance = self.tm_models.instance_groups.fit_task_to_most_remaining_capacity_instance(
                     task, instance_group_name=instance_group.name, add_hybrid_control_cost=True
-                ) or self.instance_groups.find_largest_idle_instance(instance_group_name=instance_group.name, capacity_type=task.capacity_type)
+                ) or self.tm_models.instance_groups.find_largest_idle_instance(instance_group_name=instance_group.name, capacity_type=task.capacity_type)
 
                 if execution_instance:
                     task.execution_node = execution_instance.hostname
@@ -660,7 +669,7 @@ class TaskManager(TaskBase):
                             task.log_format, instance_group.name, execution_instance.hostname, execution_instance.remaining_capacity
                         )
                     )
-                    execution_instance = self.instances[execution_instance.hostname].obj
+                    execution_instance = self.tm_models.instances[execution_instance.hostname].obj
                     self.start_task(task, instance_group, task.get_jobs_fail_chain(), execution_instance)
                     found_acceptable_queue = True
                     break

awx/main/scheduler/task_manager_models.py

@@ -15,15 +15,18 @@ logger = logging.getLogger('awx.main.scheduler')
 class TaskManagerInstance:
     """A class representing minimal data the task manager needs to represent an Instance."""
 
-    def __init__(self, obj):
+    def __init__(self, obj, **kwargs):
         self.obj = obj
         self.node_type = obj.node_type
         self.consumed_capacity = 0
         self.capacity = obj.capacity
         self.hostname = obj.hostname
+        self.jobs_running = 0
 
-    def consume_capacity(self, impact):
+    def consume_capacity(self, impact, job_impact=False):
         self.consumed_capacity += impact
+        if job_impact:
+            self.jobs_running += 1
 
     @property
     def remaining_capacity(self):
@@ -33,9 +36,106 @@ class TaskManagerInstance:
         return remaining
 
 
+class TaskManagerInstanceGroup:
+    """A class representing minimal data the task manager needs to represent an InstanceGroup."""
+
+    def __init__(self, obj, task_manager_instances=None, **kwargs):
+        self.name = obj.name
+        self.is_container_group = obj.is_container_group
+        self.container_group_jobs = 0
+        self.container_group_consumed_forks = 0
+        _instances = obj.instances.all()
+        # We want the list of TaskManagerInstance objects because these are shared across the TaskManagerInstanceGroup objects.
+        # This way when we consume capacity on an instance that is in multiple groups, we tabulate across all the groups correctly.
+        self.instances = [task_manager_instances[instance.hostname] for instance in _instances if instance.hostname in task_manager_instances]
+        self.instance_hostnames = tuple([instance.hostname for instance in _instances if instance.hostname in task_manager_instances])
+        self.max_concurrent_jobs = obj.max_concurrent_jobs
+        self.max_forks = obj.max_forks
+        self.control_task_impact = kwargs.get('control_task_impact', settings.AWX_CONTROL_NODE_TASK_IMPACT)
+
+    def consume_capacity(self, task):
+        """We only consume capacity on an instance group level if it is a container group. Otherwise we consume capacity on an instance level."""
+        if self.is_container_group:
+            self.container_group_jobs += 1
+            self.container_group_consumed_forks += task.task_impact
+        else:
+            raise RuntimeError("We only track capacity for container groups at the instance group level. Otherwise, consume capacity on instances.")
+
+    def get_remaining_instance_capacity(self):
+        return sum(inst.remaining_capacity for inst in self.instances)
+
+    def get_instance_capacity(self):
+        return sum(inst.capacity for inst in self.instances)
+
+    def get_consumed_instance_capacity(self):
+        return sum(inst.consumed_capacity for inst in self.instances)
+
+    def get_instance_jobs_running(self):
+        return sum(inst.jobs_running for inst in self.instances)
+
+    def get_jobs_running(self):
+        if self.is_container_group:
+            return self.container_group_jobs
+        return sum(inst.jobs_running for inst in self.instances)
+
+    def get_capacity(self):
+        """This reports any type of capacity, including that of container group jobs.
+
+        Container groups don't really have capacity, but if they have max_forks set,
+        we can interperet that as how much capacity the user has defined them to have.
+        """
+        if self.is_container_group:
+            return self.max_forks
+        return self.get_instance_capacity()
+
+    def get_consumed_capacity(self):
+        if self.is_container_group:
+            return self.container_group_consumed_forks
+        return self.get_consumed_instance_capacity()
+
+    def get_remaining_capacity(self):
+        return self.get_capacity() - self.get_consumed_capacity()
+
+    def has_remaining_capacity(self, task=None, control_impact=False):
+        """Pass either a task or control_impact=True to determine if the IG has capacity to run the control task or job task."""
+        task_impact = self.control_task_impact if control_impact else task.task_impact
+        job_impact = 0 if control_impact else 1
+        task_string = f"task {task.log_format} with impact of {task_impact}" if task else f"control task with impact of {task_impact}"
+
+        # We only want to loop over instances if self.max_concurrent_jobs is set
+        if self.max_concurrent_jobs == 0:
+            # Override the calculated remaining capacity, because when max_concurrent_jobs == 0 we don't enforce any max
+            remaining_jobs = 0
+        else:
+            remaining_jobs = self.max_concurrent_jobs - self.get_jobs_running() - job_impact
+
+        # We only want to loop over instances if self.max_forks is set
+        if self.max_forks == 0:
+            # Override the calculated remaining capacity, because when max_forks == 0 we don't enforce any max
+            remaining_forks = 0
+        else:
+            remaining_forks = self.max_forks - self.get_consumed_capacity() - task_impact
+
+        if remaining_jobs < 0 or remaining_forks < 0:
+            # A value less than zero means the task will not fit on the group
+            if remaining_jobs < 0:
+                logger.debug(f"{task_string} cannot fit on instance group {self.name} with {remaining_jobs} remaining jobs")
+            if remaining_forks < 0:
+                logger.debug(f"{task_string} cannot fit on instance group {self.name} with {remaining_forks} remaining forks")
+            return False
+
+        # Returning true means there is enough remaining capacity on the group to run the task (or no instance group level limits are being set)
+        logger.debug(f"{task_string} can fit on instance group {self.name} with {remaining_forks} remaining forks and {remaining_jobs}")
+        return True
+
+
 class TaskManagerInstances:
-    def __init__(self, active_tasks, instances=None, instance_fields=('node_type', 'capacity', 'hostname', 'enabled')):
+    def __init__(self, instances=None, instance_fields=('node_type', 'capacity', 'hostname', 'enabled'), **kwargs):
         self.instances_by_hostname = dict()
+        self.instance_groups_container_group_jobs = dict()
+        self.instance_groups_container_group_consumed_forks = dict()
+        self.control_task_impact = kwargs.get('control_task_impact', settings.AWX_CONTROL_NODE_TASK_IMPACT)
+
         if instances is None:
             instances = (
                 Instance.objects.filter(hostname__isnull=False, node_state=Instance.States.READY, enabled=True)
@@ -43,18 +143,15 @@ class TaskManagerInstances:
                 .only('node_type', 'node_state', 'capacity', 'hostname', 'enabled')
             )
         for instance in instances:
-            self.instances_by_hostname[instance.hostname] = TaskManagerInstance(instance)
+            self.instances_by_hostname[instance.hostname] = TaskManagerInstance(instance, **kwargs)
 
-        # initialize remaining capacity based on currently waiting and running tasks
-        for task in active_tasks:
-            if task.status not in ['waiting', 'running']:
-                continue
-            control_instance = self.instances_by_hostname.get(task.controller_node, '')
-            execution_instance = self.instances_by_hostname.get(task.execution_node, '')
-            if execution_instance and execution_instance.node_type in ('hybrid', 'execution'):
-                self.instances_by_hostname[task.execution_node].consume_capacity(task.task_impact)
-            if control_instance and control_instance.node_type in ('hybrid', 'control'):
-                self.instances_by_hostname[task.controller_node].consume_capacity(settings.AWX_CONTROL_NODE_TASK_IMPACT)
+    def consume_capacity(self, task):
+        control_instance = self.instances_by_hostname.get(task.controller_node, '')
+        execution_instance = self.instances_by_hostname.get(task.execution_node, '')
+        if execution_instance and execution_instance.node_type in ('hybrid', 'execution'):
+            self.instances_by_hostname[task.execution_node].consume_capacity(task.task_impact, job_impact=True)
+        if control_instance and control_instance.node_type in ('hybrid', 'control'):
+            self.instances_by_hostname[task.controller_node].consume_capacity(self.control_task_impact)
 
     def __getitem__(self, hostname):
         return self.instances_by_hostname.get(hostname)
@@ -64,42 +161,57 @@ class TaskManagerInstances:
 
 
 class TaskManagerInstanceGroups:
-    """A class representing minimal data the task manager needs to represent an InstanceGroup."""
+    """A class representing minimal data the task manager needs to represent all the InstanceGroups."""
 
-    def __init__(self, instances_by_hostname=None, instance_groups=None, instance_groups_queryset=None):
+    def __init__(self, task_manager_instances=None, instance_groups=None, instance_groups_queryset=None, **kwargs):
         self.instance_groups = dict()
+        self.task_manager_instances = task_manager_instances if task_manager_instances is not None else TaskManagerInstances()
         self.controlplane_ig = None
         self.pk_ig_map = dict()
+        self.control_task_impact = kwargs.get('control_task_impact', settings.AWX_CONTROL_NODE_TASK_IMPACT)
+        self.controlplane_ig_name = kwargs.get('controlplane_ig_name', settings.DEFAULT_CONTROL_PLANE_QUEUE_NAME)
 
         if instance_groups is not None:  # for testing
-            self.instance_groups = instance_groups
+            self.instance_groups = {ig.name: TaskManagerInstanceGroup(ig, self.task_manager_instances, **kwargs) for ig in instance_groups}
+            self.pk_ig_map = {ig.pk: ig for ig in instance_groups}
         else:
             if instance_groups_queryset is None:
-                instance_groups_queryset = InstanceGroup.objects.prefetch_related('instances').only('name', 'instances')
-            for instance_group in instance_groups_queryset:
-                if instance_group.name == settings.DEFAULT_CONTROL_PLANE_QUEUE_NAME:
-                    self.controlplane_ig = instance_group
-                self.instance_groups[instance_group.name] = dict(
-                    instances=[
-                        instances_by_hostname[instance.hostname] for instance in instance_group.instances.all() if instance.hostname in instances_by_hostname
-                    ],
+                instance_groups_queryset = InstanceGroup.objects.prefetch_related('instances').only(
+                    'name', 'instances', 'max_concurrent_jobs', 'max_forks', 'is_container_group'
                 )
+            for instance_group in instance_groups_queryset:
+                if instance_group.name == self.controlplane_ig_name:
+                    self.controlplane_ig = instance_group
+                self.instance_groups[instance_group.name] = TaskManagerInstanceGroup(instance_group, self.task_manager_instances, **kwargs)
                 self.pk_ig_map[instance_group.pk] = instance_group
 
+    def __getitem__(self, ig_name):
+        return self.instance_groups.get(ig_name)
+
+    def __contains__(self, ig_name):
+        return ig_name in self.instance_groups
+
     def get_remaining_capacity(self, group_name):
-        instances = self.instance_groups[group_name]['instances']
-        return sum(inst.remaining_capacity for inst in instances)
+        return self.instance_groups[group_name].get_remaining_instance_capacity()
 
     def get_consumed_capacity(self, group_name):
-        instances = self.instance_groups[group_name]['instances']
-        return sum(inst.consumed_capacity for inst in instances)
+        return self.instance_groups[group_name].get_consumed_capacity()
+
+    def get_jobs_running(self, group_name):
+        return self.instance_groups[group_name].get_jobs_running()
+
+    def get_capacity(self, group_name):
+        return self.instance_groups[group_name].get_capacity()
+
+    def get_instances(self, group_name):
+        return self.instance_groups[group_name].instances
 
     def fit_task_to_most_remaining_capacity_instance(self, task, instance_group_name, impact=None, capacity_type=None, add_hybrid_control_cost=False):
         impact = impact if impact else task.task_impact
         capacity_type = capacity_type if capacity_type else task.capacity_type
         instance_most_capacity = None
         most_remaining_capacity = -1
-        instances = self.instance_groups[instance_group_name]['instances']
+        instances = self.instance_groups[instance_group_name].instances
 
         for i in instances:
             if i.node_type not in (capacity_type, 'hybrid'):
@@ -107,7 +219,7 @@ class TaskManagerInstanceGroups:
             would_be_remaining = i.remaining_capacity - impact
             # hybrid nodes _always_ control their own tasks
             if add_hybrid_control_cost and i.node_type == 'hybrid':
-                would_be_remaining -= settings.AWX_CONTROL_NODE_TASK_IMPACT
+                would_be_remaining -= self.control_task_impact
             if would_be_remaining >= 0 and (instance_most_capacity is None or would_be_remaining > most_remaining_capacity):
                 instance_most_capacity = i
                 most_remaining_capacity = would_be_remaining
@@ -115,10 +227,13 @@ class TaskManagerInstanceGroups:
 
     def find_largest_idle_instance(self, instance_group_name, capacity_type='execution'):
         largest_instance = None
-        instances = self.instance_groups[instance_group_name]['instances']
+        instances = self.instance_groups[instance_group_name].instances
         for i in instances:
             if i.node_type not in (capacity_type, 'hybrid'):
                 continue
+            if i.capacity <= 0:
+                # We don't want to select an idle instance with 0 capacity
+                continue
             if (hasattr(i, 'jobs_running') and i.jobs_running == 0) or i.remaining_capacity == i.capacity:
                 if largest_instance is None:
                     largest_instance = i
@@ -139,3 +254,56 @@ class TaskManagerInstanceGroups:
             logger.warn(f"No instance groups in cache exist, defaulting to global instance groups for task {task}")
             return task.global_instance_groups
         return igs
+
+
+class TaskManagerModels:
+    def __init__(self, **kwargs):
+        # We want to avoid calls to settings over and over in loops, so cache this information here
+        kwargs['control_task_impact'] = kwargs.get('control_task_impact', settings.AWX_CONTROL_NODE_TASK_IMPACT)
+        kwargs['controlplane_ig_name'] = kwargs.get('controlplane_ig_name', settings.DEFAULT_CONTROL_PLANE_QUEUE_NAME)
+        self.instances = TaskManagerInstances(**kwargs)
+        self.instance_groups = TaskManagerInstanceGroups(task_manager_instances=self.instances, **kwargs)
+
+    @classmethod
+    def init_with_consumed_capacity(cls, **kwargs):
+        tmm = cls(**kwargs)
+        tasks = kwargs.get('tasks', None)
+
+        if tasks is None:
+            instance_group_queryset = kwargs.get('instance_groups_queryset', None)
+            # No tasks were provided, so we will fetch them from the database
+            task_status_filter_list = kwargs.get('task_status_filter_list', ['running', 'waiting'])
+            task_fields = kwargs.get('task_fields', ('task_impact', 'controller_node', 'execution_node', 'instance_group'))
+            from awx.main.models import UnifiedJob
+
+            if instance_group_queryset is not None:
+                logger.debug("******************INSTANCE GROUP QUERYSET PASSED -- FILTERING TASKS ****************************")
+                # Sometimes things like the serializer pass a queryset that looks at not all instance groups. in this case,
+                # we also need to filter the tasks we look at
+                tasks = UnifiedJob.objects.filter(status__in=task_status_filter_list, instance_group__in=[ig.id for ig in instance_group_queryset]).only(
+                    *task_fields
+                )
+            else:
+                # No instance group query set, look at all tasks in whole system
+                tasks = UnifiedJob.objects.filter(status__in=task_status_filter_list).only(*task_fields)
+
+        for task in tasks:
+            tmm.consume_capacity(task)
+
+        return tmm
+
+    def consume_capacity(self, task):
+        # Consume capacity on instances, which bubbles up to instance groups they are a member of
+        self.instances.consume_capacity(task)
+
+        # For container group jobs, additionally we must account for capacity consumed since
+        # The container groups have no instances to look at to track how many jobs/forks are consumed
+        if task.instance_group_id:
+            if not task.instance_group_id in self.instance_groups.pk_ig_map.keys():
+                logger.warn(
+                    f"Task {task.log_format} assigned {task.instance_group_id} but this instance group not present in map of instance groups{self.instance_groups.pk_ig_map.keys()}"
+                )
+            else:
+                ig = self.instance_groups.pk_ig_map[task.instance_group_id]
+                if ig.is_container_group:
+                    self.instance_groups[ig.name].consume_capacity(task)

awx/main/tasks/callback.py

@@ -2,8 +2,6 @@ import json
 import time
 import logging
 from collections import deque
-import os
-import stat
 
 # Django
 from django.conf import settings
@@ -206,21 +204,6 @@ class RunnerCallback:
                 self.instance = self.update_model(self.instance.pk, job_args=json.dumps(runner_config.command), job_cwd=runner_config.cwd, job_env=job_env)
             # We opened a connection just for that save, close it here now
             connections.close_all()
-        elif status_data['status'] == 'failed':
-            # For encrypted ssh_key_data, ansible-runner worker will open and write the
-            # ssh_key_data to a named pipe. Then, once the podman container starts, ssh-agent will
-            # read from this named pipe so that the key can be used in ansible-playbook.
-            # Once the podman container exits, the named pipe is deleted.
-            # However, if the podman container fails to start in the first place, e.g. the image
-            # name is incorrect, then this pipe is not cleaned up. Eventually ansible-runner
-            # processor will attempt to write artifacts to the private data dir via unstream_dir, requiring
-            # that it open this named pipe. This leads to a hang. Thus, before any artifacts
-            # are written by the processor, it's important to remove this ssh_key_data pipe.
-            private_data_dir = self.instance.job_env.get('AWX_PRIVATE_DATA_DIR', None)
-            if private_data_dir:
-                key_data_file = os.path.join(private_data_dir, 'artifacts', str(self.instance.id), 'ssh_key_data')
-                if os.path.exists(key_data_file) and stat.S_ISFIFO(os.stat(key_data_file).st_mode):
-                    os.remove(key_data_file)
         elif status_data['status'] == 'error':
             result_traceback = status_data.get('result_traceback', None)
             if result_traceback:

awx/main/tasks/jobs.py

@@ -426,7 +426,7 @@ class BaseTask(object):
         """
         instance.log_lifecycle("post_run")
 
-    def final_run_hook(self, instance, status, private_data_dir, fact_modification_times):
+    def final_run_hook(self, instance, status, private_data_dir):
         """
         Hook for any steps to run after job/task is marked as complete.
         """
@@ -469,7 +469,6 @@ class BaseTask(object):
         self.instance = self.update_model(pk, status='running', start_args='')  # blank field to remove encrypted passwords
         self.instance.websocket_emit_status("running")
         status, rc = 'error', None
-        fact_modification_times = {}
         self.runner_callback.event_ct = 0
 
         '''
@@ -498,14 +497,6 @@ class BaseTask(object):
             if not os.path.exists(settings.AWX_ISOLATION_BASE_PATH):
                 raise RuntimeError('AWX_ISOLATION_BASE_PATH=%s does not exist' % settings.AWX_ISOLATION_BASE_PATH)
 
-            # Fetch "cached" fact data from prior runs and put on the disk
-            # where ansible expects to find it
-            if getattr(self.instance, 'use_fact_cache', False):
-                self.instance.start_job_fact_cache(
-                    os.path.join(private_data_dir, 'artifacts', str(self.instance.id), 'fact_cache'),
-                    fact_modification_times,
-                )
-
             # May have to serialize the value
             private_data_files, ssh_key_data = self.build_private_data_files(self.instance, private_data_dir)
             passwords = self.build_passwords(self.instance, kwargs)
@@ -646,7 +637,7 @@ class BaseTask(object):
             self.instance.send_notification_templates('succeeded' if status == 'successful' else 'failed')
 
         try:
-            self.final_run_hook(self.instance, status, private_data_dir, fact_modification_times)
+            self.final_run_hook(self.instance, status, private_data_dir)
         except Exception:
             logger.exception('{} Final run hook errored.'.format(self.instance.log_format))
 
@@ -767,6 +758,10 @@ class SourceControlMixin(BaseTask):
 
         try:
             original_branch = None
+            failed_reason = project.get_reason_if_failed()
+            if failed_reason:
+                self.update_model(self.instance.pk, status='failed', job_explanation=failed_reason)
+                raise RuntimeError(failed_reason)
             project_path = project.get_project_path(check_if_exists=False)
             if project.scm_type == 'git' and (scm_branch and scm_branch != project.scm_branch):
                 if os.path.exists(project_path):
@@ -1056,22 +1051,25 @@ class RunJob(SourceControlMixin, BaseTask):
             error = _('Job could not start because no Execution Environment could be found.')
             self.update_model(job.pk, status='error', job_explanation=error)
             raise RuntimeError(error)
-        elif job.project.status in ('error', 'failed'):
-            msg = _('The project revision for this job template is unknown due to a failed update.')
-            job = self.update_model(job.pk, status='failed', job_explanation=msg)
-            raise RuntimeError(msg)
 
         if job.inventory.kind == 'smart':
             # cache smart inventory memberships so that the host_filter query is not
             # ran inside of the event saving code
             update_smart_memberships_for_inventory(job.inventory)
 
+        # Fetch "cached" fact data from prior runs and put on the disk
+        # where ansible expects to find it
+        if job.use_fact_cache:
+            self.facts_write_time = self.instance.start_job_fact_cache(os.path.join(private_data_dir, 'artifacts', str(job.id), 'fact_cache'))
+
     def build_project_dir(self, job, private_data_dir):
         self.sync_and_copy(job.project, private_data_dir, scm_branch=job.scm_branch)
 
-    def final_run_hook(self, job, status, private_data_dir, fact_modification_times):
-        super(RunJob, self).final_run_hook(job, status, private_data_dir, fact_modification_times)
-        if not private_data_dir:
+    def post_run_hook(self, job, status):
+        super(RunJob, self).post_run_hook(job, status)
+        job.refresh_from_db(fields=['job_env'])
+        private_data_dir = job.job_env.get('AWX_PRIVATE_DATA_DIR')
+        if (not private_data_dir) or (not hasattr(self, 'facts_write_time')):
             # If there's no private data dir, that means we didn't get into the
             # actual `run()` call; this _usually_ means something failed in
             # the pre_run_hook method
@@ -1079,9 +1077,11 @@ class RunJob(SourceControlMixin, BaseTask):
         if job.use_fact_cache:
             job.finish_job_fact_cache(
                 os.path.join(private_data_dir, 'artifacts', str(job.id), 'fact_cache'),
-                fact_modification_times,
+                self.facts_write_time,
             )
 
+    def final_run_hook(self, job, status, private_data_dir):
+        super(RunJob, self).final_run_hook(job, status, private_data_dir)
         try:
             inventory = job.inventory
         except Inventory.DoesNotExist:

awx/main/tasks/receptor.py

@@ -61,10 +61,15 @@ def read_receptor_config():
             return yaml.safe_load(f)
 
 
-def get_receptor_sockfile():
-    data = read_receptor_config()
+def work_signing_enabled(config_data):
+    for section in config_data:
+        if 'work-verification' in section:
+            return True
+    return False
 
-    for section in data:
+
+def get_receptor_sockfile(config_data):
+    for section in config_data:
         for entry_name, entry_data in section.items():
             if entry_name == 'control-service':
                 if 'filename' in entry_data:
@@ -75,12 +80,11 @@ def get_receptor_sockfile():
         raise RuntimeError(f'Receptor conf {__RECEPTOR_CONF} does not have control-service entry needed to get sockfile')
 
 
-def get_tls_client(use_stream_tls=None):
+def get_tls_client(config_data, use_stream_tls=None):
     if not use_stream_tls:
         return None
 
-    data = read_receptor_config()
-    for section in data:
+    for section in config_data:
         for entry_name, entry_data in section.items():
             if entry_name == 'tls-client':
                 if 'name' in entry_data:
@@ -88,10 +92,12 @@ def get_tls_client(use_stream_tls=None):
     return None
 
 
-def get_receptor_ctl():
-    receptor_sockfile = get_receptor_sockfile()
+def get_receptor_ctl(config_data=None):
+    if config_data is None:
+        config_data = read_receptor_config()
+    receptor_sockfile = get_receptor_sockfile(config_data)
     try:
-        return ReceptorControl(receptor_sockfile, config=__RECEPTOR_CONF, tlsclient=get_tls_client(True))
+        return ReceptorControl(receptor_sockfile, config=__RECEPTOR_CONF, tlsclient=get_tls_client(config_data, True))
     except RuntimeError:
         return ReceptorControl(receptor_sockfile)
 
@@ -159,15 +165,18 @@ def run_until_complete(node, timing_data=None, **kwargs):
     """
     Runs an ansible-runner work_type on remote node, waits until it completes, then returns stdout.
     """
-    receptor_ctl = get_receptor_ctl()
+    config_data = read_receptor_config()
+    receptor_ctl = get_receptor_ctl(config_data)
 
     use_stream_tls = getattr(get_conn_type(node, receptor_ctl), 'name', None) == "STREAMTLS"
-    kwargs.setdefault('tlsclient', get_tls_client(use_stream_tls))
+    kwargs.setdefault('tlsclient', get_tls_client(config_data, use_stream_tls))
     kwargs.setdefault('ttl', '20s')
     kwargs.setdefault('payload', '')
+    if work_signing_enabled(config_data):
+        kwargs['signwork'] = True
 
     transmit_start = time.time()
-    result = receptor_ctl.submit_work(worktype='ansible-runner', node=node, signwork=True, **kwargs)
+    result = receptor_ctl.submit_work(worktype='ansible-runner', node=node, **kwargs)
 
     unit_id = result['unitid']
     run_start = time.time()
@@ -208,7 +217,10 @@ def run_until_complete(node, timing_data=None, **kwargs):
     if state_name.lower() == 'failed':
         work_detail = status.get('Detail', '')
         if work_detail:
-            raise RemoteJobError(f'Receptor error from {node}, detail:\n{work_detail}')
+            if stdout:
+                raise RemoteJobError(f'Receptor error from {node}, detail:\n{work_detail}\nstdout:\n{stdout}')
+            else:
+                raise RemoteJobError(f'Receptor error from {node}, detail:\n{work_detail}')
         else:
             raise RemoteJobError(f'Unknown ansible-runner error on node {node}, stdout:\n{stdout}')
 
@@ -299,7 +311,8 @@ class AWXReceptorJob:
 
     def run(self):
         # We establish a connection to the Receptor socket
-        receptor_ctl = get_receptor_ctl()
+        self.config_data = read_receptor_config()
+        receptor_ctl = get_receptor_ctl(self.config_data)
 
         res = None
         try:
@@ -324,7 +337,7 @@ class AWXReceptorJob:
         if self.work_type == 'ansible-runner':
             work_submit_kw['node'] = self.task.instance.execution_node
             use_stream_tls = get_conn_type(work_submit_kw['node'], receptor_ctl).name == "STREAMTLS"
-            work_submit_kw['tlsclient'] = get_tls_client(use_stream_tls)
+            work_submit_kw['tlsclient'] = get_tls_client(self.config_data, use_stream_tls)
 
         with concurrent.futures.ThreadPoolExecutor(max_workers=1) as executor:
             transmitter_future = executor.submit(self.transmit, sockin)
@@ -474,7 +487,9 @@ class AWXReceptorJob:
 
     @property
     def sign_work(self):
-        return True if self.work_type in ('ansible-runner', 'local') else False
+        if self.work_type in ('ansible-runner', 'local'):
+            return work_signing_enabled(self.config_data)
+        return False
 
     @property
     def work_type(self):

awx/main/tests/functional/api/test_instance.py

@@ -7,7 +7,7 @@ from awx.main.models.ha import Instance
 from django.test.utils import override_settings
 
 
-INSTANCE_KWARGS = dict(hostname='example-host', cpu=6, memory=36000000000, cpu_capacity=6, mem_capacity=42)
+INSTANCE_KWARGS = dict(hostname='example-host', cpu=6, node_type='execution', memory=36000000000, cpu_capacity=6, mem_capacity=42)
 
 
 @pytest.mark.django_db

awx/main/tests/functional/task_management/test_capacity.py

@@ -4,7 +4,7 @@ from awx.main.models import (
     Instance,
     InstanceGroup,
 )
-from awx.main.scheduler.task_manager_models import TaskManagerInstanceGroups, TaskManagerInstances
+from awx.main.scheduler.task_manager_models import TaskManagerInstanceGroups
 
 
 class TestInstanceGroupInstanceMapping(TransactionTestCase):
@@ -23,11 +23,10 @@ class TestInstanceGroupInstanceMapping(TransactionTestCase):
     def test_mapping(self):
         self.sample_cluster()
         with self.assertNumQueries(3):
-            instances = TaskManagerInstances([])  # empty task list
-            instance_groups = TaskManagerInstanceGroups(instances_by_hostname=instances)
+            instance_groups = TaskManagerInstanceGroups()
 
         ig_instance_map = instance_groups.instance_groups
 
-        assert set(i.hostname for i in ig_instance_map['ig_small']['instances']) == set(['i1'])
-        assert set(i.hostname for i in ig_instance_map['ig_large']['instances']) == set(['i2', 'i3'])
-        assert set(i.hostname for i in ig_instance_map['default']['instances']) == set(['i2'])
+        assert set(i.hostname for i in ig_instance_map['ig_small'].instances) == set(['i1'])
+        assert set(i.hostname for i in ig_instance_map['ig_large'].instances) == set(['i2', 'i3'])
+        assert set(i.hostname for i in ig_instance_map['default'].instances) == set(['i2'])

awx/main/tests/functional/task_management/test_container_groups.py

@@ -10,6 +10,10 @@ from awx.main.utils import (
     create_temporary_fifo,
 )
 
+from awx.main.scheduler import TaskManager
+
+from . import create_job
+
 
 @pytest.fixture
 def containerized_job(default_instance_group, kube_credential, job_template_factory):
@@ -34,6 +38,50 @@ def test_containerized_job(containerized_job):
     assert containerized_job.instance_group.credential.kubernetes
 
 
+@pytest.mark.django_db
+def test_max_concurrent_jobs_blocks_start_of_new_jobs(controlplane_instance_group, containerized_job, mocker):
+    """Construct a scenario where only 1 job will fit within the max_concurrent_jobs of the container group.
+
+    Since max_concurrent_jobs is set to 1, even though 2 jobs are in pending
+    and would be launched into the container group, only one will be started.
+    """
+    containerized_job.unified_job_template.allow_simultaneous = True
+    containerized_job.unified_job_template.save()
+    default_instance_group = containerized_job.instance_group
+    default_instance_group.max_concurrent_jobs = 1
+    default_instance_group.save()
+    task_impact = 1
+    # Create a second job that should not be scheduled at first, blocked by the other
+    create_job(containerized_job.unified_job_template)
+    tm = TaskManager()
+    with mock.patch('awx.main.models.Job.task_impact', new_callable=mock.PropertyMock) as mock_task_impact:
+        mock_task_impact.return_value = task_impact
+        with mock.patch.object(TaskManager, "start_task", wraps=tm.start_task) as mock_job:
+            tm.schedule()
+            mock_job.assert_called_once()
+
+
+@pytest.mark.django_db
+def test_max_forks_blocks_start_of_new_jobs(controlplane_instance_group, containerized_job, mocker):
+    """Construct a scenario where only 1 job will fit within the max_forks of the container group.
+
+    In this case, we set the container_group max_forks to 10, and make the task_impact of a job 6.
+    Therefore, only 1 job will fit within the max of 10.
+    """
+    containerized_job.unified_job_template.allow_simultaneous = True
+    containerized_job.unified_job_template.save()
+    default_instance_group = containerized_job.instance_group
+    default_instance_group.max_forks = 10
+    # Create a second job that should not be scheduled
+    create_job(containerized_job.unified_job_template)
+    tm = TaskManager()
+    with mock.patch('awx.main.models.Job.task_impact', new_callable=mock.PropertyMock) as mock_task_impact:
+        mock_task_impact.return_value = 6
+        with mock.patch("awx.main.scheduler.TaskManager.start_task"):
+            tm.schedule()
+            tm.start_task.assert_called_once()
+
+
 @pytest.mark.django_db
 def test_kubectl_ssl_verification(containerized_job, default_job_execution_environment):
     containerized_job.execution_environment = default_job_execution_environment

awx/main/tests/functional/task_management/test_scheduler.py

@@ -248,6 +248,76 @@ def test_multi_jt_capacity_blocking(hybrid_instance, job_template_factory, mocke
         mock_job.assert_called_once_with(j2, controlplane_instance_group, [], instance)
 
 
+@pytest.mark.django_db
+def test_max_concurrent_jobs_ig_capacity_blocking(hybrid_instance, job_template_factory, mocker):
+    """When max_concurrent_jobs of an instance group is more restrictive than capacity of instances, enforce max_concurrent_jobs."""
+    instance = hybrid_instance
+    controlplane_instance_group = instance.rampart_groups.first()
+    # We will expect only 1 job to be started
+    controlplane_instance_group.max_concurrent_jobs = 1
+    controlplane_instance_group.save()
+    num_jobs = 3
+    jobs = []
+    for i in range(num_jobs):
+        jobs.append(
+            create_job(job_template_factory(f'jt{i}', organization=f'org{i}', project=f'proj{i}', inventory=f'inv{i}', credential=f'cred{i}').job_template)
+        )
+    tm = TaskManager()
+    task_impact = 1
+
+    # Sanity check that multiple jobs would run if not for the max_concurrent_jobs setting.
+    assert task_impact * num_jobs < controlplane_instance_group.capacity
+    tm = TaskManager()
+    with mock.patch('awx.main.models.Job.task_impact', new_callable=mock.PropertyMock) as mock_task_impact:
+        mock_task_impact.return_value = task_impact
+        with mock.patch.object(TaskManager, "start_task", wraps=tm.start_task) as mock_job:
+            tm.schedule()
+            mock_job.assert_called_once()
+            jobs[0].status = 'running'
+            jobs[0].controller_node = instance.hostname
+            jobs[0].execution_node = instance.hostname
+            jobs[0].instance_group = controlplane_instance_group
+            jobs[0].save()
+
+    # while that job is running, we should not start another job
+    with mock.patch('awx.main.models.Job.task_impact', new_callable=mock.PropertyMock) as mock_task_impact:
+        mock_task_impact.return_value = task_impact
+        with mock.patch.object(TaskManager, "start_task", wraps=tm.start_task) as mock_job:
+            tm.schedule()
+            mock_job.assert_not_called()
+    # now job is done, we should start one of the two other jobs
+    jobs[0].status = 'successful'
+    jobs[0].save()
+    with mock.patch('awx.main.models.Job.task_impact', new_callable=mock.PropertyMock) as mock_task_impact:
+        mock_task_impact.return_value = task_impact
+        with mock.patch.object(TaskManager, "start_task", wraps=tm.start_task) as mock_job:
+            tm.schedule()
+            mock_job.assert_called_once()
+
+
+@pytest.mark.django_db
+def test_max_forks_ig_capacity_blocking(hybrid_instance, job_template_factory, mocker):
+    """When max_forks of an instance group is less than the capacity of instances, enforce max_forks."""
+    instance = hybrid_instance
+    controlplane_instance_group = instance.rampart_groups.first()
+    controlplane_instance_group.max_forks = 15
+    controlplane_instance_group.save()
+    task_impact = 10
+    num_jobs = 2
+    # Sanity check that 2 jobs would run if not for the max_forks setting.
+    assert controlplane_instance_group.max_forks < controlplane_instance_group.capacity
+    assert task_impact * num_jobs > controlplane_instance_group.max_forks
+    assert task_impact * num_jobs < controlplane_instance_group.capacity
+    for i in range(num_jobs):
+        create_job(job_template_factory(f'jt{i}', organization=f'org{i}', project=f'proj{i}', inventory=f'inv{i}', credential=f'cred{i}').job_template)
+    tm = TaskManager()
+    with mock.patch('awx.main.models.Job.task_impact', new_callable=mock.PropertyMock) as mock_task_impact:
+        mock_task_impact.return_value = task_impact
+        with mock.patch.object(TaskManager, "start_task", wraps=tm.start_task) as mock_job:
+            tm.schedule()
+            mock_job.assert_called_once()
+
+
 @pytest.mark.django_db
 def test_single_job_dependencies_project_launch(controlplane_instance_group, job_template_factory, mocker):
     objects = job_template_factory('jt', organization='org1', project='proj', inventory='inv', credential='cred')

awx/main/tests/functional/test_instances.py

@@ -1,7 +1,7 @@
 import pytest
 from unittest import mock
 
-from awx.main.models import AdHocCommand, InventoryUpdate, JobTemplate
+from awx.main.models import AdHocCommand, InventoryUpdate, JobTemplate, Job
 from awx.main.models.activity_stream import ActivityStream
 from awx.main.models.ha import Instance, InstanceGroup
 from awx.main.tasks.system import apply_cluster_membership_policies
@@ -15,6 +15,24 @@ def test_default_tower_instance_group(default_instance_group, job_factory):
     assert default_instance_group in job_factory().preferred_instance_groups
 
 
+@pytest.mark.django_db
+@pytest.mark.parametrize('node_type', ('execution', 'control'))
+@pytest.mark.parametrize('active', (True, False))
+def test_get_cleanup_task_kwargs_active_jobs(node_type, active):
+    instance = Instance.objects.create(hostname='foobar', node_type=node_type)
+    job_kwargs = dict()
+    job_kwargs['controller_node' if node_type == 'control' else 'execution_node'] = instance.hostname
+    job_kwargs['status'] = 'running' if active else 'successful'
+
+    job = Job.objects.create(**job_kwargs)
+    kwargs = instance.get_cleanup_task_kwargs()
+
+    if active:
+        assert kwargs['exclude_strings'] == [f'awx_{job.pk}_']
+    else:
+        assert 'exclude_strings' not in kwargs
+
+
 @pytest.mark.django_db
 class TestPolicyTaskScheduling:
     """Tests make assertions about when the policy task gets scheduled"""

awx/main/tests/functional/test_licenses.py

@@ -121,8 +121,8 @@ def test_python_and_js_licenses():
         return errors
 
     base_dir = settings.BASE_DIR
-    api_licenses = index_licenses('%s/../docs/licenses' % base_dir)
-    ui_licenses = index_licenses('%s/../docs/licenses/ui' % base_dir)
+    api_licenses = index_licenses('%s/../licenses' % base_dir)
+    ui_licenses = index_licenses('%s/../licenses/ui' % base_dir)
     api_requirements = read_api_requirements('%s/../requirements' % base_dir)
     ui_requirements = read_ui_requirements('%s/ui' % base_dir)
 

awx/main/tests/functional/test_notifications.py

@@ -75,6 +75,7 @@ def test_encrypted_subfields(get, post, user, organization):
     url = reverse('api:notification_template_detail', kwargs={'pk': response.data['id']})
     response = get(url, u)
     assert response.data['notification_configuration']['account_token'] == "$encrypted$"
+
     with mock.patch.object(notification_template_actual.notification_class, "send_messages", assert_send):
         notification_template_actual.send("Test", {'body': "Test"})
 
@@ -175,3 +176,46 @@ def test_custom_environment_injection(post, user, organization):
 
         fake_send.side_effect = _send_side_effect
         template.send('subject', 'message')
+
+
+def mock_post(*args, **kwargs):
+    class MockGoodResponse:
+        def __init__(self):
+            self.status_code = 200
+
+    class MockRedirectResponse:
+        def __init__(self):
+            self.status_code = 301
+            self.headers = {"Location": "http://goodendpoint"}
+
+    if kwargs['url'] == "http://goodendpoint":
+        return MockGoodResponse()
+    else:
+        return MockRedirectResponse()
+
+
+@pytest.mark.django_db
+@mock.patch('requests.post', side_effect=mock_post)
+def test_webhook_notification_pointed_to_a_redirect_launch_endpoint(post, admin, organization):
+
+    n1 = NotificationTemplate.objects.create(
+        name="test-webhook",
+        description="test webhook",
+        organization=organization,
+        notification_type="webhook",
+        notification_configuration=dict(
+            url="http://some.fake.url",
+            disable_ssl_verification=True,
+            http_method="POST",
+            headers={
+                "Content-Type": "application/json",
+            },
+            username=admin.username,
+            password=admin.password,
+        ),
+        messages={
+            "success": {"message": "", "body": "{}"},
+        },
+    )
+
+    assert n1.send("", n1.messages.get("success").get("body")) == 1

awx/main/tests/unit/models/test_ha.py

@@ -1,10 +1,7 @@
 import pytest
-from unittest import mock
-from unittest.mock import Mock
 from decimal import Decimal
 
-from awx.main.models import InstanceGroup, Instance
-from awx.main.scheduler.task_manager_models import TaskManagerInstanceGroups
+from awx.main.models import Instance
 
 
 @pytest.mark.parametrize('capacity_adjustment', [0.0, 0.25, 0.5, 0.75, 1, 1.5, 3])
@@ -17,83 +14,6 @@ def test_capacity_adjustment_no_save(capacity_adjustment):
     assert inst.capacity == (float(inst.capacity_adjustment) * abs(inst.mem_capacity - inst.cpu_capacity) + min(inst.mem_capacity, inst.cpu_capacity))
 
 
-def T(impact):
-    j = mock.Mock(spec_set=['task_impact', 'capacity_type'])
-    j.task_impact = impact
-    j.capacity_type = 'execution'
-    return j
-
-
-def Is(param):
-    """
-    param:
-        [remaining_capacity1, remaining_capacity2, remaining_capacity3, ...]
-        [(jobs_running1, capacity1), (jobs_running2, capacity2), (jobs_running3, capacity3), ...]
-    """
-
-    instances = []
-    if isinstance(param[0], tuple):
-        for (jobs_running, capacity) in param:
-            inst = Mock()
-            inst.capacity = capacity
-            inst.jobs_running = jobs_running
-            inst.node_type = 'execution'
-            instances.append(inst)
-    else:
-        for i in param:
-            inst = Mock()
-            inst.remaining_capacity = i
-            inst.node_type = 'execution'
-            instances.append(inst)
-    return instances
-
-
-class TestInstanceGroup(object):
-    @pytest.mark.parametrize(
-        'task,instances,instance_fit_index,reason',
-        [
-            (T(100), Is([100]), 0, "Only one, pick it"),
-            (T(100), Is([100, 100]), 0, "Two equally good fits, pick the first"),
-            (T(100), Is([50, 100]), 1, "First instance not as good as second instance"),
-            (T(100), Is([50, 0, 20, 100, 100, 100, 30, 20]), 3, "Pick Instance [3] as it is the first that the task fits in."),
-            (T(100), Is([50, 0, 20, 99, 11, 1, 5, 99]), None, "The task don't a fit, you must a quit!"),
-        ],
-    )
-    def test_fit_task_to_most_remaining_capacity_instance(self, task, instances, instance_fit_index, reason):
-        InstanceGroup(id=10)
-        tm_igs = TaskManagerInstanceGroups(instance_groups={'controlplane': {'instances': instances}})
-
-        instance_picked = tm_igs.fit_task_to_most_remaining_capacity_instance(task, 'controlplane')
-
-        if instance_fit_index is None:
-            assert instance_picked is None, reason
-        else:
-            assert instance_picked == instances[instance_fit_index], reason
-
-    @pytest.mark.parametrize(
-        'instances,instance_fit_index,reason',
-        [
-            (Is([(0, 100)]), 0, "One idle instance, pick it"),
-            (Is([(1, 100)]), None, "One un-idle instance, pick nothing"),
-            (Is([(0, 100), (0, 200), (1, 500), (0, 700)]), 3, "Pick the largest idle instance"),
-            (Is([(0, 100), (0, 200), (1, 10000), (0, 700), (0, 699)]), 3, "Pick the largest idle instance"),
-            (Is([(0, 0)]), None, "One idle but down instance, don't pick it"),
-        ],
-    )
-    def test_find_largest_idle_instance(self, instances, instance_fit_index, reason):
-        def filter_offline_instances(*args):
-            return filter(lambda i: i.capacity > 0, instances)
-
-        InstanceGroup(id=10)
-        instances_online_only = filter_offline_instances(instances)
-        tm_igs = TaskManagerInstanceGroups(instance_groups={'controlplane': {'instances': instances_online_only}})
-
-        if instance_fit_index is None:
-            assert tm_igs.find_largest_idle_instance('controlplane') is None, reason
-        else:
-            assert tm_igs.find_largest_idle_instance('controlplane') == instances[instance_fit_index], reason
-
-
 def test_cleanup_params_defaults():
     inst = Instance(hostname='foobar')
     assert inst.get_cleanup_task_kwargs(exclude_strings=['awx_423_']) == {'exclude_strings': ['awx_423_'], 'file_pattern': '/tmp/awx_*_*', 'grace_period': 60}

awx/main/tests/unit/models/test_jobs.py

@@ -36,15 +36,14 @@ def job(mocker, hosts, inventory):
 
 def test_start_job_fact_cache(hosts, job, inventory, tmpdir):
     fact_cache = os.path.join(tmpdir, 'facts')
-    modified_times = {}
-    job.start_job_fact_cache(fact_cache, modified_times, 0)
+    last_modified = job.start_job_fact_cache(fact_cache, timeout=0)
 
     for host in hosts:
         filepath = os.path.join(fact_cache, host.name)
         assert os.path.exists(filepath)
         with open(filepath, 'r') as f:
             assert f.read() == json.dumps(host.ansible_facts)
-        assert filepath in modified_times
+        assert os.path.getmtime(filepath) <= last_modified
 
 
 def test_fact_cache_with_invalid_path_traversal(job, inventory, tmpdir, mocker):
@@ -58,18 +57,16 @@ def test_fact_cache_with_invalid_path_traversal(job, inventory, tmpdir, mocker):
     )
 
     fact_cache = os.path.join(tmpdir, 'facts')
-    job.start_job_fact_cache(fact_cache, {}, 0)
+    job.start_job_fact_cache(fact_cache, timeout=0)
     # a file called "foo" should _not_ be written outside the facts dir
     assert os.listdir(os.path.join(fact_cache, '..')) == ['facts']
 
 
 def test_finish_job_fact_cache_with_existing_data(job, hosts, inventory, mocker, tmpdir):
     fact_cache = os.path.join(tmpdir, 'facts')
-    modified_times = {}
-    job.start_job_fact_cache(fact_cache, modified_times, 0)
+    last_modified = job.start_job_fact_cache(fact_cache, timeout=0)
 
-    for h in hosts:
-        h.save = mocker.Mock()
+    bulk_update = mocker.patch('django.db.models.query.QuerySet.bulk_update')
 
     ansible_facts_new = {"foo": "bar"}
     filepath = os.path.join(fact_cache, hosts[1].name)
@@ -83,23 +80,20 @@ def test_finish_job_fact_cache_with_existing_data(job, hosts, inventory, mocker,
         new_modification_time = time.time() + 3600
         os.utime(filepath, (new_modification_time, new_modification_time))
 
-    job.finish_job_fact_cache(fact_cache, modified_times)
+    job.finish_job_fact_cache(fact_cache, last_modified)
 
     for host in (hosts[0], hosts[2], hosts[3]):
-        host.save.assert_not_called()
         assert host.ansible_facts == {"a": 1, "b": 2}
         assert host.ansible_facts_modified is None
     assert hosts[1].ansible_facts == ansible_facts_new
-    hosts[1].save.assert_called_once_with(update_fields=['ansible_facts', 'ansible_facts_modified'])
+    bulk_update.assert_called_once_with([hosts[1]], ['ansible_facts', 'ansible_facts_modified'])
 
 
 def test_finish_job_fact_cache_with_bad_data(job, hosts, inventory, mocker, tmpdir):
     fact_cache = os.path.join(tmpdir, 'facts')
-    modified_times = {}
-    job.start_job_fact_cache(fact_cache, modified_times, 0)
+    last_modified = job.start_job_fact_cache(fact_cache, timeout=0)
 
-    for h in hosts:
-        h.save = mocker.Mock()
+    bulk_update = mocker.patch('django.db.models.query.QuerySet.bulk_update')
 
     for h in hosts:
         filepath = os.path.join(fact_cache, h.name)
@@ -109,26 +103,22 @@ def test_finish_job_fact_cache_with_bad_data(job, hosts, inventory, mocker, tmpd
             new_modification_time = time.time() + 3600
             os.utime(filepath, (new_modification_time, new_modification_time))
 
-    job.finish_job_fact_cache(fact_cache, modified_times)
+    job.finish_job_fact_cache(fact_cache, last_modified)
 
-    for h in hosts:
-        h.save.assert_not_called()
+    bulk_update.assert_not_called()
 
 
 def test_finish_job_fact_cache_clear(job, hosts, inventory, mocker, tmpdir):
     fact_cache = os.path.join(tmpdir, 'facts')
-    modified_times = {}
-    job.start_job_fact_cache(fact_cache, modified_times, 0)
+    last_modified = job.start_job_fact_cache(fact_cache, timeout=0)
 
-    for h in hosts:
-        h.save = mocker.Mock()
+    bulk_update = mocker.patch('django.db.models.query.QuerySet.bulk_update')
 
     os.remove(os.path.join(fact_cache, hosts[1].name))
-    job.finish_job_fact_cache(fact_cache, modified_times)
+    job.finish_job_fact_cache(fact_cache, last_modified)
 
     for host in (hosts[0], hosts[2], hosts[3]):
-        host.save.assert_not_called()
         assert host.ansible_facts == {"a": 1, "b": 2}
         assert host.ansible_facts_modified is None
     assert hosts[1].ansible_facts == {}
-    hosts[1].save.assert_called_once_with()
+    bulk_update.assert_called_once_with([hosts[1]], ['ansible_facts', 'ansible_facts_modified'])

awx/main/tests/unit/notifications/test_webhook.py

@@ -27,11 +27,12 @@ def test_send_messages_as_POST():
             ]
         )
         requests_mock.post.assert_called_once_with(
-            'http://example.com',
+            url='http://example.com',
             auth=None,
             data=json.dumps({'text': 'test body'}, ensure_ascii=False).encode('utf-8'),
             headers={'Content-Type': 'application/json', 'User-Agent': 'AWX 0.0.1.dev (open)'},
             verify=True,
+            allow_redirects=False,
         )
         assert sent_messages == 1
 
@@ -57,11 +58,12 @@ def test_send_messages_as_PUT():
             ]
         )
         requests_mock.put.assert_called_once_with(
-            'http://example.com',
+            url='http://example.com',
             auth=None,
             data=json.dumps({'text': 'test body 2'}, ensure_ascii=False).encode('utf-8'),
             headers={'Content-Type': 'application/json', 'User-Agent': 'AWX 0.0.1.dev (open)'},
             verify=True,
+            allow_redirects=False,
         )
         assert sent_messages == 1
 
@@ -87,11 +89,12 @@ def test_send_messages_with_username():
             ]
         )
         requests_mock.post.assert_called_once_with(
-            'http://example.com',
+            url='http://example.com',
             auth=('userstring', None),
             data=json.dumps({'text': 'test body'}, ensure_ascii=False).encode('utf-8'),
             headers={'Content-Type': 'application/json', 'User-Agent': 'AWX 0.0.1.dev (open)'},
             verify=True,
+            allow_redirects=False,
         )
         assert sent_messages == 1
 
@@ -117,11 +120,12 @@ def test_send_messages_with_password():
             ]
         )
         requests_mock.post.assert_called_once_with(
-            'http://example.com',
+            url='http://example.com',
             auth=(None, 'passwordstring'),
             data=json.dumps({'text': 'test body'}, ensure_ascii=False).encode('utf-8'),
             headers={'Content-Type': 'application/json', 'User-Agent': 'AWX 0.0.1.dev (open)'},
             verify=True,
+            allow_redirects=False,
         )
         assert sent_messages == 1
 
@@ -147,11 +151,12 @@ def test_send_messages_with_username_and_password():
             ]
         )
         requests_mock.post.assert_called_once_with(
-            'http://example.com',
+            url='http://example.com',
             auth=('userstring', 'passwordstring'),
             data=json.dumps({'text': 'test body'}, ensure_ascii=False).encode('utf-8'),
             headers={'Content-Type': 'application/json', 'User-Agent': 'AWX 0.0.1.dev (open)'},
             verify=True,
+            allow_redirects=False,
         )
         assert sent_messages == 1
 
@@ -177,11 +182,12 @@ def test_send_messages_with_no_verify_ssl():
             ]
         )
         requests_mock.post.assert_called_once_with(
-            'http://example.com',
+            url='http://example.com',
             auth=None,
             data=json.dumps({'text': 'test body'}, ensure_ascii=False).encode('utf-8'),
             headers={'Content-Type': 'application/json', 'User-Agent': 'AWX 0.0.1.dev (open)'},
             verify=False,
+            allow_redirects=False,
         )
         assert sent_messages == 1
 
@@ -207,7 +213,7 @@ def test_send_messages_with_additional_headers():
             ]
         )
         requests_mock.post.assert_called_once_with(
-            'http://example.com',
+            url='http://example.com',
             auth=None,
             data=json.dumps({'text': 'test body'}, ensure_ascii=False).encode('utf-8'),
             headers={
@@ -217,5 +223,6 @@ def test_send_messages_with_additional_headers():
                 'X-Test-Header2': 'test-content-2',
             },
             verify=True,
+            allow_redirects=False,
         )
         assert sent_messages == 1

awx/main/tests/unit/test_capacity.py

@@ -1,6 +1,6 @@
 import pytest
 
-from awx.main.scheduler.task_manager_models import TaskManagerInstanceGroups, TaskManagerInstances
+from awx.main.scheduler.task_manager_models import TaskManagerModels
 
 
 class FakeMeta(object):
@@ -16,38 +16,64 @@ class FakeObject(object):
 
 
 class Job(FakeObject):
-    task_impact = 43
-    is_container_group_task = False
-    controller_node = ''
-    execution_node = ''
+    def __init__(self, **kwargs):
+        self.task_impact = kwargs.get('task_impact', 43)
+        self.is_container_group_task = kwargs.get('is_container_group_task', False)
+        self.controller_node = kwargs.get('controller_node', '')
+        self.execution_node = kwargs.get('execution_node', '')
+        self.instance_group = kwargs.get('instance_group', None)
+        self.instance_group_id = self.instance_group.id if self.instance_group else None
+        self.capacity_type = kwargs.get('capacity_type', 'execution')
 
     def log_format(self):
         return 'job 382 (fake)'
 
 
+class Instances(FakeObject):
+    def add(self, *args):
+        for instance in args:
+            self.obj.instance_list.append(instance)
+
+    def all(self):
+        return self.obj.instance_list
+
+
+class InstanceGroup(FakeObject):
+    def __init__(self, **kwargs):
+        super(InstanceGroup, self).__init__(**kwargs)
+        self.instance_list = []
+        self.pk = self.id = kwargs.get('id', 1)
+
+    @property
+    def instances(self):
+        mgr = Instances(obj=self)
+        return mgr
+
+    @property
+    def is_container_group(self):
+        return False
+
+    @property
+    def max_concurrent_jobs(self):
+        return 0
+
+    @property
+    def max_forks(self):
+        return 0
+
+
+class Instance(FakeObject):
+    def __init__(self, **kwargs):
+        self.node_type = kwargs.get('node_type', 'hybrid')
+        self.capacity = kwargs.get('capacity', 0)
+        self.hostname = kwargs.get('hostname', 'fakehostname')
+        self.consumed_capacity = 0
+        self.jobs_running = 0
+
+
 @pytest.fixture
 def sample_cluster():
     def stand_up_cluster():
-        class Instances(FakeObject):
-            def add(self, *args):
-                for instance in args:
-                    self.obj.instance_list.append(instance)
-
-            def all(self):
-                return self.obj.instance_list
-
-        class InstanceGroup(FakeObject):
-            def __init__(self, **kwargs):
-                super(InstanceGroup, self).__init__(**kwargs)
-                self.instance_list = []
-
-            @property
-            def instances(self):
-                mgr = Instances(obj=self)
-                return mgr
-
-        class Instance(FakeObject):
-            pass
 
         ig_small = InstanceGroup(name='ig_small')
         ig_large = InstanceGroup(name='ig_large')
@@ -66,14 +92,12 @@ def sample_cluster():
 @pytest.fixture
 def create_ig_manager():
     def _rf(ig_list, tasks):
-        instances = TaskManagerInstances(tasks, instances=set(inst for ig in ig_list for inst in ig.instance_list))
-
-        seed_igs = {}
-        for ig in ig_list:
-            seed_igs[ig.name] = {'instances': [instances[inst.hostname] for inst in ig.instance_list]}
-
-        instance_groups = TaskManagerInstanceGroups(instance_groups=seed_igs)
-        return instance_groups
+        tm_models = TaskManagerModels.init_with_consumed_capacity(
+            tasks=tasks,
+            instances=set(inst for ig in ig_list for inst in ig.instance_list),
+            instance_groups=ig_list,
+        )
+        return tm_models.instance_groups
 
     return _rf
 
@@ -126,3 +150,75 @@ def test_RBAC_reduced_filter(sample_cluster, create_ig_manager):
     # Cross-links between groups not visible to current user,
     # so a naieve accounting of capacities is returned instead
     assert instance_groups_mgr.get_consumed_capacity('default') == 43
+
+
+def Is(param):
+    """
+    param:
+        [remaining_capacity1, remaining_capacity2, remaining_capacity3, ...]
+        [(jobs_running1, capacity1), (jobs_running2, capacity2), (jobs_running3, capacity3), ...]
+    """
+
+    instances = []
+    if isinstance(param[0], tuple):
+        for index, (jobs_running, capacity) in enumerate(param):
+            inst = Instance(capacity=capacity, node_type='execution', hostname=f'fakehost-{index}')
+            inst.jobs_running = jobs_running
+            instances.append(inst)
+    else:
+        for index, capacity in enumerate(param):
+            inst = Instance(capacity=capacity, node_type='execution', hostname=f'fakehost-{index}')
+            inst.node_type = 'execution'
+            instances.append(inst)
+    return instances
+
+
+class TestSelectBestInstanceForTask(object):
+    @pytest.mark.parametrize(
+        'task,instances,instance_fit_index,reason',
+        [
+            (Job(task_impact=100), Is([100]), 0, "Only one, pick it"),
+            (Job(task_impact=100), Is([100, 100]), 0, "Two equally good fits, pick the first"),
+            (Job(task_impact=100), Is([50, 100]), 1, "First instance not as good as second instance"),
+            (Job(task_impact=100), Is([50, 0, 20, 100, 100, 100, 30, 20]), 3, "Pick Instance [3] as it is the first that the task fits in."),
+            (Job(task_impact=100), Is([50, 0, 20, 99, 11, 1, 5, 99]), None, "The task don't a fit, you must a quit!"),
+        ],
+    )
+    def test_fit_task_to_most_remaining_capacity_instance(self, task, instances, instance_fit_index, reason):
+        ig = InstanceGroup(id=10, name='controlplane')
+        tasks = []
+        for instance in instances:
+            ig.instances.add(instance)
+            for _ in range(instance.jobs_running):
+                tasks.append(Job(execution_node=instance.hostname, controller_node=instance.hostname, instance_group=ig))
+        tm_models = TaskManagerModels.init_with_consumed_capacity(tasks=tasks, instances=instances, instance_groups=[ig])
+        instance_picked = tm_models.instance_groups.fit_task_to_most_remaining_capacity_instance(task, 'controlplane')
+
+        if instance_fit_index is None:
+            assert instance_picked is None, reason
+        else:
+            assert instance_picked.hostname == instances[instance_fit_index].hostname, reason
+
+    @pytest.mark.parametrize(
+        'instances,instance_fit_index,reason',
+        [
+            (Is([(0, 100)]), 0, "One idle instance, pick it"),
+            (Is([(1, 100)]), None, "One un-idle instance, pick nothing"),
+            (Is([(0, 100), (0, 200), (1, 500), (0, 700)]), 3, "Pick the largest idle instance"),
+            (Is([(0, 100), (0, 200), (1, 10000), (0, 700), (0, 699)]), 3, "Pick the largest idle instance"),
+            (Is([(0, 0)]), None, "One idle but down instance, don't pick it"),
+        ],
+    )
+    def test_find_largest_idle_instance(self, instances, instance_fit_index, reason):
+        ig = InstanceGroup(id=10, name='controlplane')
+        tasks = []
+        for instance in instances:
+            ig.instances.add(instance)
+            for _ in range(instance.jobs_running):
+                tasks.append(Job(execution_node=instance.hostname, controller_node=instance.hostname, instance_group=ig))
+        tm_models = TaskManagerModels.init_with_consumed_capacity(tasks=tasks, instances=instances, instance_groups=[ig])
+
+        if instance_fit_index is None:
+            assert tm_models.instance_groups.find_largest_idle_instance('controlplane') is None, reason
+        else:
+            assert tm_models.instance_groups.find_largest_idle_instance('controlplane').hostname == instances[instance_fit_index].hostname, reason

awx/main/utils/common.py

@@ -11,11 +11,12 @@ import os
 import subprocess
 import re
 import stat
+import sys
 import urllib.parse
 import threading
 import contextlib
 import tempfile
-from functools import reduce, wraps
+import functools
 
 # Django
 from django.core.exceptions import ObjectDoesNotExist, FieldDoesNotExist
@@ -73,6 +74,7 @@ __all__ = [
     'NullablePromptPseudoField',
     'model_instance_diff',
     'parse_yaml_or_json',
+    'is_testing',
     'RequireDebugTrueOrTest',
     'has_model_field_prefetched',
     'set_environ',
@@ -88,6 +90,7 @@ __all__ = [
     'deepmerge',
     'get_event_partition_epoch',
     'cleanup_new_process',
+    'log_excess_runtime',
 ]
 
 
@@ -144,6 +147,19 @@ def underscore_to_camelcase(s):
     return ''.join(x.capitalize() or '_' for x in s.split('_'))
 
 
+@functools.cache
+def is_testing(argv=None):
+    '''Return True if running django or py.test unit tests.'''
+    if 'PYTEST_CURRENT_TEST' in os.environ.keys():
+        return True
+    argv = sys.argv if argv is None else argv
+    if len(argv) >= 1 and ('py.test' in argv[0] or 'py/test.py' in argv[0]):
+        return True
+    elif len(argv) >= 2 and argv[1] == 'test':
+        return True
+    return False
+
+
 class RequireDebugTrueOrTest(logging.Filter):
     """
     Logging filter to output when in DEBUG mode or running tests.
@@ -152,7 +168,7 @@ class RequireDebugTrueOrTest(logging.Filter):
     def filter(self, record):
         from django.conf import settings
 
-        return settings.DEBUG or settings.IS_TESTING()
+        return settings.DEBUG or is_testing()
 
 
 class IllegalArgumentError(ValueError):
@@ -174,7 +190,7 @@ def memoize(ttl=60, cache_key=None, track_function=False, cache=None):
     cache = cache or get_memoize_cache()
 
     def memoize_decorator(f):
-        @wraps(f)
+        @functools.wraps(f)
         def _memoizer(*args, **kwargs):
             if track_function:
                 cache_dict_key = slugify('%r %r' % (args, kwargs))
@@ -992,7 +1008,7 @@ def getattrd(obj, name, default=NoDefaultProvided):
     """
 
     try:
-        return reduce(getattr, name.split("."), obj)
+        return functools.reduce(getattr, name.split("."), obj)
     except AttributeError:
         if default != NoDefaultProvided:
             return default
@@ -1188,7 +1204,7 @@ def cleanup_new_process(func):
     Cleanup django connection, cache connection, before executing new thread or processes entry point, func.
     """
 
-    @wraps(func)
+    @functools.wraps(func)
     def wrapper_cleanup_new_process(*args, **kwargs):
         from awx.conf.settings import SettingsWrapper  # noqa
 
@@ -1200,15 +1216,30 @@ def cleanup_new_process(func):
     return wrapper_cleanup_new_process
 
 
-def log_excess_runtime(func_logger, cutoff=5.0):
+def log_excess_runtime(func_logger, cutoff=5.0, debug_cutoff=5.0, msg=None, add_log_data=False):
     def log_excess_runtime_decorator(func):
-        @wraps(func)
+        @functools.wraps(func)
         def _new_func(*args, **kwargs):
             start_time = time.time()
-            return_value = func(*args, **kwargs)
-            delta = time.time() - start_time
-            if delta > cutoff:
-                logger.info(f'Running {func.__name__!r} took {delta:.2f}s')
+            log_data = {'name': repr(func.__name__)}
+
+            if add_log_data:
+                return_value = func(*args, log_data=log_data, **kwargs)
+            else:
+                return_value = func(*args, **kwargs)
+
+            log_data['delta'] = time.time() - start_time
+            if isinstance(return_value, dict):
+                log_data.update(return_value)
+
+            if msg is None:
+                record_msg = 'Running {name} took {delta:.2f}s'
+            else:
+                record_msg = msg
+            if log_data['delta'] > cutoff:
+                func_logger.info(record_msg.format(**log_data))
+            elif log_data['delta'] > debug_cutoff:
+                func_logger.debug(record_msg.format(**log_data))
             return return_value
 
         return _new_func

awx/main/utils/handlers.py

@@ -110,7 +110,7 @@ if settings.COLOR_LOGS is True:
                 # logs rendered with cyan text
                 previous_level_map = self.level_map.copy()
                 if record.name == "awx.analytics.job_lifecycle":
-                    self.level_map[logging.DEBUG] = (None, 'cyan', True)
+                    self.level_map[logging.INFO] = (None, 'cyan', True)
                 msg = super(ColorHandler, self).colorize(line, record)
                 self.level_map = previous_level_map
                 return msg

awx/settings/defaults.py

@@ -10,28 +10,6 @@ import socket
 from datetime import timedelta
 
 
-# Build paths inside the project like this: os.path.join(BASE_DIR, ...)
-BASE_DIR = os.path.dirname(os.path.dirname(__file__))
-
-
-def is_testing(argv=None):
-    import sys
-
-    '''Return True if running django or py.test unit tests.'''
-    if 'PYTEST_CURRENT_TEST' in os.environ.keys():
-        return True
-    argv = sys.argv if argv is None else argv
-    if len(argv) >= 1 and ('py.test' in argv[0] or 'py/test.py' in argv[0]):
-        return True
-    elif len(argv) >= 2 and argv[1] == 'test':
-        return True
-    return False
-
-
-def IS_TESTING(argv=None):
-    return is_testing(argv)
-
-
 if "pytest" in sys.modules:
     from unittest import mock
 
@@ -40,9 +18,13 @@ if "pytest" in sys.modules:
 else:
     import ldap
 
+
 DEBUG = True
 SQL_DEBUG = DEBUG
 
+# Build paths inside the project like this: os.path.join(BASE_DIR, ...)
+BASE_DIR = os.path.dirname(os.path.dirname(__file__))
+
 # FIXME: it would be nice to cycle back around and allow this to be
 # BigAutoField going forward, but we'd have to be explicit about our
 # existing models.
@@ -101,7 +83,7 @@ USE_L10N = True
 
 USE_TZ = True
 
-STATICFILES_DIRS = (os.path.join(BASE_DIR, 'ui', 'build', 'static'), os.path.join(BASE_DIR, 'static'))
+STATICFILES_DIRS = [os.path.join(BASE_DIR, 'ui', 'build', 'static'), os.path.join(BASE_DIR, 'static')]
 
 # Absolute filesystem path to the directory where static file are collected via
 # the collectstatic command.
@@ -254,6 +236,14 @@ START_TASK_LIMIT = 100
 TASK_MANAGER_TIMEOUT = 300
 TASK_MANAGER_TIMEOUT_GRACE_PERIOD = 60
 
+# Number of seconds _in addition to_ the task manager timeout a job can stay
+# in waiting without being reaped
+JOB_WAITING_GRACE_PERIOD = 60
+
+# Number of seconds after a container group job finished time to wait
+# before the awx_k8s_reaper task will tear down the pods
+K8S_POD_REAPER_GRACE_PERIOD = 60
+
 # Disallow sending session cookies over insecure connections
 SESSION_COOKIE_SECURE = True
 
@@ -314,11 +304,13 @@ INSTALLED_APPS = [
     'django.contrib.messages',
     'django.contrib.sessions',
     'django.contrib.sites',
+    # daphne has to be installed before django.contrib.staticfiles for the app to startup
+    # According to channels 4.0 docs you install daphne instead of channels now
+    'daphne',
     'django.contrib.staticfiles',
     'oauth2_provider',
     'rest_framework',
     'django_extensions',
-    'channels',
     'polymorphic',
     'taggit',
     'social_django',
@@ -993,6 +985,13 @@ DJANGO_GUID = {'GUID_HEADER_NAME': 'X-API-Request-Id'}
 DEFAULT_EXECUTION_QUEUE_NAME = 'default'
 # pod spec used when the default execution queue is a container group, e.g. when deploying on k8s/ocp with the operator
 DEFAULT_EXECUTION_QUEUE_POD_SPEC_OVERRIDE = ''
+# Max number of concurrently consumed forks for the default execution queue
+# Zero means no limit
+DEFAULT_EXECUTION_QUEUE_MAX_FORKS = 0
+# Max number of concurrently running jobs for the default execution queue
+# Zero means no limit
+DEFAULT_EXECUTION_QUEUE_MAX_CONCURRENT_JOBS = 0
+
 # Name of the default controlplane queue
 DEFAULT_CONTROL_PLANE_QUEUE_NAME = 'controlplane'
 
@@ -1004,16 +1003,5 @@ DEFAULT_CONTAINER_RUN_OPTIONS = ['--network', 'slirp4netns:enable_ipv6=true']
 # Mount exposed paths as hostPath resource in k8s/ocp
 AWX_MOUNT_ISOLATED_PATHS_ON_K8S = False
 
-# Time out task managers if they take longer than this many seconds
-TASK_MANAGER_TIMEOUT = 300
-
-# Number of seconds _in addition to_ the task manager timeout a job can stay
-# in waiting without being reaped
-JOB_WAITING_GRACE_PERIOD = 60
-
-# Number of seconds after a container group job finished time to wait
-# before the awx_k8s_reaper task will tear down the pods
-K8S_POD_REAPER_GRACE_PERIOD = 60
-
 # This is overridden downstream via /etc/tower/conf.d/cluster_host_id.py
 CLUSTER_HOST_ID = socket.gethostname()

awx/ui/package-lock.json

@@ -7,9 +7,9 @@
       "name": "ui",
       "dependencies": {
         "@lingui/react": "3.14.0",
-        "@patternfly/patternfly": "4.210.2",
-        "@patternfly/react-core": "^4.239.0",
-        "@patternfly/react-icons": "4.90.0",
+        "@patternfly/patternfly": "4.217.1",
+        "@patternfly/react-core": "^4.250.1",
+        "@patternfly/react-icons": "4.92.10",
         "@patternfly/react-table": "4.108.0",
         "ace-builds": "^1.10.1",
         "ansi-to-html": "0.7.2",
@@ -3747,26 +3747,26 @@
       "dev": true
     },
     "node_modules/@patternfly/patternfly": {
-      "version": "4.210.2",
-      "resolved": "https://registry.npmjs.org/@patternfly/patternfly/-/patternfly-4.210.2.tgz",
-      "integrity": "sha512-aZiW24Bxi6uVmk5RyNTp+6q6ThtlJZotNRJfWVeGuwu1UlbBuV4DFa1bpjA6jfTZpfEpX2YL5+R+4ZVSCFAVdw=="
+      "version": "4.217.1",
+      "resolved": "https://registry.npmjs.org/@patternfly/patternfly/-/patternfly-4.217.1.tgz",
+      "integrity": "sha512-uN7JgfQsyR16YHkuGRCTIcBcnyKIqKjGkB2SGk9x1XXH3yYGenL83kpAavX9Xtozqp17KppOlybJuzcKvZMrgw=="
     },
     "node_modules/@patternfly/react-core": {
-      "version": "4.239.0",
-      "resolved": "https://registry.npmjs.org/@patternfly/react-core/-/react-core-4.239.0.tgz",
-      "integrity": "sha512-6CmYABCJLUXTlzCk6C3WouMNZpS0BCT+aHU8CvYpFQ/NrpYp3MJaDsYbqgCRWV42rmIO5iXun/4WhXBJzJEoQg==",
+      "version": "4.250.1",
+      "resolved": "https://registry.npmjs.org/@patternfly/react-core/-/react-core-4.250.1.tgz",
+      "integrity": "sha512-vAOZPQdZzYXl/vkHnHMIt1eC3nrPDdsuuErPatkNPwmSvilXuXmWP5wxoJ36FbSNRRURkprFwx52zMmWS3iHJA==",
       "dependencies": {
-        "@patternfly/react-icons": "^4.90.0",
-        "@patternfly/react-styles": "^4.89.0",
-        "@patternfly/react-tokens": "^4.91.0",
+        "@patternfly/react-icons": "^4.92.6",
+        "@patternfly/react-styles": "^4.91.6",
+        "@patternfly/react-tokens": "^4.93.6",
         "focus-trap": "6.9.2",
         "react-dropzone": "9.0.0",
         "tippy.js": "5.1.2",
         "tslib": "^2.0.0"
       },
       "peerDependencies": {
-        "react": "^16.8.0 || ^17.0.0",
-        "react-dom": "^16.8.0 || ^17.0.0"
+        "react": "^16.8 || ^17 || ^18",
+        "react-dom": "^16.8 || ^17 || ^18"
       }
     },
     "node_modules/@patternfly/react-core/node_modules/tslib": {
@@ -3775,18 +3775,18 @@
       "integrity": "sha512-77EbyPPpMz+FRFRuAFlWMtmgUWGe9UOG2Z25NqCwiIjRhOf5iKGuzSe5P2w1laq+FkRy4p+PCuVkJSGkzTEKVw=="
     },
     "node_modules/@patternfly/react-icons": {
-      "version": "4.90.0",
-      "resolved": "https://registry.npmjs.org/@patternfly/react-icons/-/react-icons-4.90.0.tgz",
-      "integrity": "sha512-qEnQKbxbUgyiosiKSkeKEBwmhgJwWEqniIAFyoxj+kpzAdeu7ueWe5iBbqo06mvDOedecFiM5mIE1N0MXwk8Yw==",
+      "version": "4.92.10",
+      "resolved": "https://registry.npmjs.org/@patternfly/react-icons/-/react-icons-4.92.10.tgz",
+      "integrity": "sha512-vwCy7b+OyyuvLDSLqLUG2DkJZgMDogjld8tJTdAaG8HiEhC1sJPZac+5wD7AuS3ym/sQolS4vYtNiVDnMEORxA==",
       "peerDependencies": {
-        "react": "^16.8.0 || ^17.0.0",
-        "react-dom": "^16.8.0 || ^17.0.0"
+        "react": "^16.8 || ^17 || ^18",
+        "react-dom": "^16.8 || ^17 || ^18"
       }
     },
     "node_modules/@patternfly/react-styles": {
-      "version": "4.89.0",
-      "resolved": "https://registry.npmjs.org/@patternfly/react-styles/-/react-styles-4.89.0.tgz",
-      "integrity": "sha512-SkT+qx3Xqu70T5s+i/AUT2hI2sKAPDX4ffeiJIUDu/oyWiFdk+/9DEivnLSyJMruroXXN33zKibvzb5rH7DKTQ=="
+      "version": "4.91.10",
+      "resolved": "https://registry.npmjs.org/@patternfly/react-styles/-/react-styles-4.91.10.tgz",
+      "integrity": "sha512-fAG4Vjp63ohiR92F4e/Gkw5q1DSSckHKqdnEF75KUpSSBORzYP0EKMpupSd6ItpQFJw3iWs3MJi3/KIAAfU1Jw=="
     },
     "node_modules/@patternfly/react-table": {
       "version": "4.108.0",
@@ -3811,9 +3811,9 @@
       "integrity": "sha512-d6xOpEDfsi2CZVlPQzGeux8XMwLT9hssAsaPYExaQMuYskwb+x1x7J371tWlbBdWHroy99KnVB6qIkUbs5X3UQ=="
     },
     "node_modules/@patternfly/react-tokens": {
-      "version": "4.91.0",
-      "resolved": "https://registry.npmjs.org/@patternfly/react-tokens/-/react-tokens-4.91.0.tgz",
-      "integrity": "sha512-QeQCy8o8E/16fAr8mxqXIYRmpTsjCHJXi5p5jmgEDFmYMesN6Pqfv6N5D0FHb+CIaNOZWRps7GkWvlIMIE81sw=="
+      "version": "4.93.10",
+      "resolved": "https://registry.npmjs.org/@patternfly/react-tokens/-/react-tokens-4.93.10.tgz",
+      "integrity": "sha512-F+j1irDc9M6zvY6qNtDryhbpnHz3R8ymHRdGelNHQzPTIK88YSWEnT1c9iUI+uM/iuZol7sJmO5STtg2aPIDRQ=="
     },
     "node_modules/@pmmmwh/react-refresh-webpack-plugin": {
       "version": "0.5.4",
@@ -25089,18 +25089,18 @@
       "dev": true
     },
     "@patternfly/patternfly": {
-      "version": "4.210.2",
-      "resolved": "https://registry.npmjs.org/@patternfly/patternfly/-/patternfly-4.210.2.tgz",
-      "integrity": "sha512-aZiW24Bxi6uVmk5RyNTp+6q6ThtlJZotNRJfWVeGuwu1UlbBuV4DFa1bpjA6jfTZpfEpX2YL5+R+4ZVSCFAVdw=="
+      "version": "4.217.1",
+      "resolved": "https://registry.npmjs.org/@patternfly/patternfly/-/patternfly-4.217.1.tgz",
+      "integrity": "sha512-uN7JgfQsyR16YHkuGRCTIcBcnyKIqKjGkB2SGk9x1XXH3yYGenL83kpAavX9Xtozqp17KppOlybJuzcKvZMrgw=="
     },
     "@patternfly/react-core": {
-      "version": "4.239.0",
-      "resolved": "https://registry.npmjs.org/@patternfly/react-core/-/react-core-4.239.0.tgz",
-      "integrity": "sha512-6CmYABCJLUXTlzCk6C3WouMNZpS0BCT+aHU8CvYpFQ/NrpYp3MJaDsYbqgCRWV42rmIO5iXun/4WhXBJzJEoQg==",
+      "version": "4.250.1",
+      "resolved": "https://registry.npmjs.org/@patternfly/react-core/-/react-core-4.250.1.tgz",
+      "integrity": "sha512-vAOZPQdZzYXl/vkHnHMIt1eC3nrPDdsuuErPatkNPwmSvilXuXmWP5wxoJ36FbSNRRURkprFwx52zMmWS3iHJA==",
       "requires": {
-        "@patternfly/react-icons": "^4.90.0",
-        "@patternfly/react-styles": "^4.89.0",
-        "@patternfly/react-tokens": "^4.91.0",
+        "@patternfly/react-icons": "^4.92.6",
+        "@patternfly/react-styles": "^4.91.6",
+        "@patternfly/react-tokens": "^4.93.6",
         "focus-trap": "6.9.2",
         "react-dropzone": "9.0.0",
         "tippy.js": "5.1.2",
@@ -25115,15 +25115,15 @@
       }
     },
     "@patternfly/react-icons": {
-      "version": "4.90.0",
-      "resolved": "https://registry.npmjs.org/@patternfly/react-icons/-/react-icons-4.90.0.tgz",
-      "integrity": "sha512-qEnQKbxbUgyiosiKSkeKEBwmhgJwWEqniIAFyoxj+kpzAdeu7ueWe5iBbqo06mvDOedecFiM5mIE1N0MXwk8Yw==",
+      "version": "4.92.10",
+      "resolved": "https://registry.npmjs.org/@patternfly/react-icons/-/react-icons-4.92.10.tgz",
+      "integrity": "sha512-vwCy7b+OyyuvLDSLqLUG2DkJZgMDogjld8tJTdAaG8HiEhC1sJPZac+5wD7AuS3ym/sQolS4vYtNiVDnMEORxA==",
       "requires": {}
     },
     "@patternfly/react-styles": {
-      "version": "4.89.0",
-      "resolved": "https://registry.npmjs.org/@patternfly/react-styles/-/react-styles-4.89.0.tgz",
-      "integrity": "sha512-SkT+qx3Xqu70T5s+i/AUT2hI2sKAPDX4ffeiJIUDu/oyWiFdk+/9DEivnLSyJMruroXXN33zKibvzb5rH7DKTQ=="
+      "version": "4.91.10",
+      "resolved": "https://registry.npmjs.org/@patternfly/react-styles/-/react-styles-4.91.10.tgz",
+      "integrity": "sha512-fAG4Vjp63ohiR92F4e/Gkw5q1DSSckHKqdnEF75KUpSSBORzYP0EKMpupSd6ItpQFJw3iWs3MJi3/KIAAfU1Jw=="
     },
     "@patternfly/react-table": {
       "version": "4.108.0",
@@ -25146,9 +25146,9 @@
       }
     },
     "@patternfly/react-tokens": {
-      "version": "4.91.0",
-      "resolved": "https://registry.npmjs.org/@patternfly/react-tokens/-/react-tokens-4.91.0.tgz",
-      "integrity": "sha512-QeQCy8o8E/16fAr8mxqXIYRmpTsjCHJXi5p5jmgEDFmYMesN6Pqfv6N5D0FHb+CIaNOZWRps7GkWvlIMIE81sw=="
+      "version": "4.93.10",
+      "resolved": "https://registry.npmjs.org/@patternfly/react-tokens/-/react-tokens-4.93.10.tgz",
+      "integrity": "sha512-F+j1irDc9M6zvY6qNtDryhbpnHz3R8ymHRdGelNHQzPTIK88YSWEnT1c9iUI+uM/iuZol7sJmO5STtg2aPIDRQ=="
     },
     "@pmmmwh/react-refresh-webpack-plugin": {
       "version": "0.5.4",

awx/ui/package.json

@@ -7,9 +7,9 @@
   },
   "dependencies": {
     "@lingui/react": "3.14.0",
-    "@patternfly/patternfly": "4.210.2",
-    "@patternfly/react-core": "^4.239.0",
-    "@patternfly/react-icons": "4.90.0",
+    "@patternfly/patternfly": "4.217.1",
+    "@patternfly/react-core": "^4.250.1",
+    "@patternfly/react-icons": "4.92.10",
     "@patternfly/react-table": "4.108.0",
     "ace-builds": "^1.10.1",
     "ansi-to-html": "0.7.2",

awx/ui/src/components/InstanceGroupLabels/InstanceGroupLabels.js

@@ -0,0 +1,49 @@
+import React from 'react';
+import { arrayOf, bool, number, shape, string } from 'prop-types';
+
+import { Label, LabelGroup } from '@patternfly/react-core';
+import { Link } from 'react-router-dom';
+
+function InstanceGroupLabels({ labels, isLinkable }) {
+  const buildLinkURL = (isContainerGroup) =>
+    isContainerGroup
+      ? '/instance_groups/container_group/'
+      : '/instance_groups/';
+  return (
+    <LabelGroup numLabels={5}>
+      {labels.map(({ id, name, is_container_group }) =>
+        isLinkable ? (
+          <Label
+            color="blue"
+            key={id}
+            render={({ className, content, componentRef }) => (
+              <Link
+                className={className}
+                innerRef={componentRef}
+                to={`${buildLinkURL(is_container_group)}${id}/details`}
+              >
+                {content}
+              </Link>
+            )}
+          >
+            {name}
+          </Label>
+        ) : (
+          <Label color="blue" key={id}>
+            {name}
+          </Label>
+        )
+      )}
+    </LabelGroup>
+  );
+}
+
+InstanceGroupLabels.propTypes = {
+  labels: arrayOf(shape({ id: number.isRequired, name: string.isRequired }))
+    .isRequired,
+  isLinkable: bool,
+};
+
+InstanceGroupLabels.defaultProps = { isLinkable: false };
+
+export default InstanceGroupLabels;

awx/ui/src/components/InstanceGroupLabels/index.js

@@ -0,0 +1 @@
+export { default } from './InstanceGroupLabels';

awx/ui/src/components/PromptDetail/PromptDetail.js

@@ -6,6 +6,7 @@ import { Link } from 'react-router-dom';
 import styled from 'styled-components';
 import { Chip, Divider, Title } from '@patternfly/react-core';
 import { toTitleCase } from 'util/strings';
+import InstanceGroupLabels from 'components/InstanceGroupLabels';
 import CredentialChip from '../CredentialChip';
 import ChipGroup from '../ChipGroup';
 import { DetailList, Detail, UserDateDetail } from '../DetailList';
@@ -227,21 +228,7 @@ function PromptDetail({
                   label={t`Instance Groups`}
                   rows={4}
                   value={
-                    <ChipGroup
-                      numChips={5}
-                      totalChips={overrides.instance_groups.length}
-                      ouiaId="prompt-instance-groups-chips"
-                    >
-                      {overrides.instance_groups.map((instance_group) => (
-                        <Chip
-                          key={instance_group.id}
-                          ouiaId={`instance-group-${instance_group.id}-chip`}
-                          isReadOnly
-                        >
-                          {instance_group.name}
-                        </Chip>
-                      ))}
-                    </ChipGroup>
+                    <InstanceGroupLabels labels={overrides.instance_groups} />
                   }
                 />
               )}

awx/ui/src/components/Schedule/ScheduleDetail/ScheduleDetail.js

@@ -10,6 +10,7 @@ import useRequest, { useDismissableError } from 'hooks/useRequest';
 import { JobTemplatesAPI, SchedulesAPI, WorkflowJobTemplatesAPI } from 'api';
 import { parseVariableField, jsonToYaml } from 'util/yaml';
 import { useConfig } from 'contexts/Config';
+import InstanceGroupLabels from 'components/InstanceGroupLabels';
 import parseRuleObj from '../shared/parseRuleObj';
 import FrequencyDetails from './FrequencyDetails';
 import AlertModal from '../../AlertModal';
@@ -27,11 +28,6 @@ import { VariablesDetail } from '../../CodeEditor';
 import { VERBOSITY } from '../../VerbositySelectField';
 import getHelpText from '../../../screens/Template/shared/JobTemplate.helptext';
 
-const buildLinkURL = (instance) =>
-  instance.is_container_group
-    ? '/instance_groups/container_group/'
-    : '/instance_groups/';
-
 const PromptDivider = styled(Divider)`
   margin-top: var(--pf-global--spacer--lg);
   margin-bottom: var(--pf-global--spacer--lg);
@@ -498,26 +494,7 @@ function ScheduleDetail({ hasDaysToKeepField, schedule, surveyConfig }) {
                 fullWidth
                 label={t`Instance Groups`}
                 value={
-                  <ChipGroup
-                    numChips={5}
-                    totalChips={instanceGroups.length}
-                    ouiaId="instance-group-chips"
-                  >
-                    {instanceGroups.map((ig) => (
-                      <Link
-                        to={`${buildLinkURL(ig)}${ig.id}/details`}
-                        key={ig.id}
-                      >
-                        <Chip
-                          key={ig.id}
-                          ouiaId={`instance-group-${ig.id}-chip`}
-                          isReadOnly
-                        >
-                          {ig.name}
-                        </Chip>
-                      </Link>
-                    ))}
-                  </ChipGroup>
+                  <InstanceGroupLabels labels={instanceGroups} isLinkable />
                 }
                 isEmpty={instanceGroups.length === 0}
               />

awx/ui/src/components/Schedule/shared/ScheduleForm.js

@@ -416,8 +416,14 @@ function ScheduleForm({
 
       if (options.end === 'onDate') {
         if (
-          DateTime.fromISO(values.startDate) >=
-          DateTime.fromISO(options.endDate)
+          DateTime.fromFormat(
+            `${values.startDate} ${values.startTime}`,
+            'yyyy-LL-dd h:mm a'
+          ).toMillis() >=
+          DateTime.fromFormat(
+            `${options.endDate} ${options.endTime}`,
+            'yyyy-LL-dd h:mm a'
+          ).toMillis()
         ) {
           freqErrors.endDate = t`Please select an end date/time that comes after the start date/time.`;
         }

awx/ui/src/components/Schedule/shared/ScheduleForm.test.js

@@ -900,6 +900,36 @@ describe('<ScheduleForm />', () => {
       );
     });
 
+    test('should create schedule with the same start and end date provided that the end date is at a later time', async () => {
+      const today = DateTime.now().toFormat('yyyy-LL-dd');
+      const laterTime = DateTime.now().plus({ hours: 1 }).toFormat('h:mm a');
+      await act(async () => {
+        wrapper.find('DatePicker[aria-label="End date"]').prop('onChange')(
+          today,
+          new Date(today)
+        );
+      });
+      wrapper.update();
+      expect(
+        wrapper
+          .find('FormGroup[data-cy="schedule-End date/time"]')
+          .prop('helperTextInvalid')
+      ).toBe(
+        'Please select an end date/time that comes after the start date/time.'
+      );
+      await act(async () => {
+        wrapper.find('TimePicker[aria-label="End time"]').prop('onChange')(
+          laterTime
+        );
+      });
+      wrapper.update();
+      expect(
+        wrapper
+          .find('FormGroup[data-cy="schedule-End date/time"]')
+          .prop('helperTextInvalid')
+      ).toBe(undefined);
+    });
+
     test('error shown when on day number is not between 1 and 31', async () => {
       await act(async () => {
         wrapper.find('FrequencySelect#schedule-frequency').invoke('onChange')([

awx/ui/src/components/WorkflowOutputNavigation/WorkflowOutputNavigation.js

@@ -24,12 +24,10 @@ function WorkflowOutputNavigation({ relatedJobs, parentRef }) {
   const { id } = useParams();
 
   const relevantResults = relatedJobs.filter(
-    ({
-      job: jobId,
-      summary_fields: {
-        unified_job_template: { unified_job_type },
-      },
-    }) => jobId && `${jobId}` !== id && unified_job_type !== 'workflow_approval'
+    ({ job: jobId, summary_fields }) =>
+      jobId &&
+      `${jobId}` !== id &&
+      summary_fields.job.type !== 'workflow_approval'
   );
 
   const [isOpen, setIsOpen] = useState(false);
@@ -101,16 +99,14 @@ function WorkflowOutputNavigation({ relatedJobs, parentRef }) {
           {sortedJobs?.map((node) => (
             <SelectOption
               key={node.id}
-              to={`/jobs/${
-                JOB_URL_SEGMENT_MAP[
-                  node.summary_fields.unified_job_template.unified_job_type
-                ]
-              }/${node.summary_fields.job?.id}/output`}
+              to={`/jobs/${JOB_URL_SEGMENT_MAP[node.summary_fields.job.type]}/${
+                node.summary_fields.job?.id
+              }/output`}
               component={Link}
-              value={node.summary_fields.unified_job_template.name}
+              value={node.summary_fields.job.name}
             >
               {stringIsUUID(node.identifier)
-                ? node.summary_fields.unified_job_template.name
+                ? node.summary_fields.job.name
                 : node.identifier}
             </SelectOption>
           ))}

awx/ui/src/components/WorkflowOutputNavigation/WorkflowOutputNavigation.test.js

@@ -0,0 +1,85 @@
+import React from 'react';
+import { within, render, screen, waitFor } from '@testing-library/react';
+import userEvent from '@testing-library/user-event';
+import WorkflowOutputNavigation from './WorkflowOutputNavigation';
+import { createMemoryHistory } from 'history';
+import { I18nProvider } from '@lingui/react';
+import { i18n } from '@lingui/core';
+import { en } from 'make-plural/plurals';
+import english from '../../../src/locales/en/messages';
+import { Router } from 'react-router-dom';
+
+jest.mock('react-router-dom', () => ({
+  ...jest.requireActual('react-router-dom'),
+  useParams: () => ({
+    id: 1,
+  }),
+}));
+const jobs = [
+  {
+    id: 1,
+    summary_fields: {
+      job: {
+        name: 'Ansible',
+        type: 'project_update',
+        id: 1,
+        status: 'successful',
+      },
+    },
+    job: 4,
+  },
+  {
+    id: 2,
+    summary_fields: {
+      job: {
+        name: 'Durham',
+        type: 'job',
+        id: 2,
+        status: 'successful',
+      },
+    },
+    job: 3,
+  },
+  {
+    id: 3,
+    summary_fields: {
+      job: {
+        name: 'Red hat',
+        type: 'job',
+        id: 3,
+        status: 'successful',
+      },
+    },
+    job: 2,
+  },
+];
+
+describe('<WorkflowOuputNavigation/>', () => {
+  test('Should open modal and deprovision node', async () => {
+    i18n.loadLocaleData({ en: { plurals: en } });
+    i18n.load({ en: english });
+    i18n.activate('en');
+    const user = userEvent.setup();
+    const ref = jest
+      .spyOn(React, 'useRef')
+      .mockReturnValueOnce({ current: 'div' });
+    const history = createMemoryHistory({
+      initialEntries: ['jobs/playbook/2/output'],
+    });
+    render(
+      <I18nProvider i18n={i18n}>
+        <Router history={history}>
+          <WorkflowOutputNavigation relatedJobs={jobs} parentRef={ref} />
+        </Router>
+      </I18nProvider>
+    );
+
+    const button = screen.getByRole('button');
+    await user.click(button);
+
+    await waitFor(() => screen.getByText('Workflow Nodes'));
+    await waitFor(() => screen.getByText('Red hat'));
+    await waitFor(() => screen.getByText('Durham'));
+    await waitFor(() => screen.getByText('Ansible'));
+  });
+});

awx/ui/src/screens/Credential/CredentialEdit/CredentialEdit.test.js

@@ -282,7 +282,7 @@ const mockInputSources = {
         summary_fields: {
           source_credential: {
             id: 20,
-            name: 'CyberArk Conjur Secret Lookup',
+            name: 'CyberArk Conjur Secrets Manager Lookup',
             description: '',
             kind: 'conjur',
             cloud: false,
@@ -301,7 +301,7 @@ const mockInputSources = {
         summary_fields: {
           source_credential: {
             id: 20,
-            name: 'CyberArk Conjur Secret Lookup',
+            name: 'CyberArk Conjur Secrets Manager Lookup',
             description: '',
             kind: 'conjur',
             cloud: false,

awx/ui/src/screens/Credential/shared/CredentialPlugins/CredentialPluginPrompt/CredentialPluginPrompt.test.js

@@ -36,14 +36,14 @@ const mockCredentialTypeDetail = {
     url: '/api/v2/credential_types/20/',
     related: {
       named_url:
-        '/api/v2/credential_types/CyberArk Conjur Secret Lookup+external/',
+        '/api/v2/credential_types/CyberArk Conjur Secrets Manager Lookup+external/',
       credentials: '/api/v2/credential_types/20/credentials/',
       activity_stream: '/api/v2/credential_types/20/activity_stream/',
     },
     summary_fields: { user_capabilities: { edit: false, delete: false } },
     created: '2020-05-18T21:53:35.398260Z',
     modified: '2020-05-18T21:54:05.451444Z',
-    name: 'CyberArk Conjur Secret Lookup',
+    name: 'CyberArk Conjur Secrets Manager Lookup',
     description: '',
     kind: 'external',
     namespace: 'conjur',

awx/ui/src/screens/Credential/shared/data.credentialTypes.json

@@ -546,7 +546,7 @@
     },
     "created": "2020-05-18T21:53:35.398260Z",
     "modified": "2020-05-18T21:54:05.451444Z",
-    "name": "CyberArk Conjur Secret Lookup",
+    "name": "CyberArk Conjur Secrets Manager Lookup",
     "description": "",
     "kind": "external",
     "namespace": "conjur",

awx/ui/src/screens/Credential/shared/data.cyberArkCredential.json

@@ -3,7 +3,7 @@
   "type": "credential",
   "url": "/api/v2/credentials/1/",
   "related": {
-    "named_url": "/api/v2/credentials/CyberArk Conjur Secret Lookup++CyberArk Conjur Secret Lookup+external++/",
+    "named_url": "/api/v2/credentials/CyberArk Conjur Secrets Manager Lookup+external++/",
     "created_by": "/api/v2/users/1/",
     "modified_by": "/api/v2/users/1/",
     "activity_stream": "/api/v2/credentials/1/activity_stream/",
@@ -19,7 +19,7 @@
   "summary_fields": {
     "credential_type": {
       "id": 20,
-      "name": "CyberArk Conjur Secret Lookup",
+      "name": "CyberArk Conjur Secrets Manager Lookup",
       "description": ""
     },
     "created_by": {
@@ -69,7 +69,7 @@
   },
   "created": "2020-05-19T12:51:36.956029Z",
   "modified": "2020-05-19T12:51:36.956086Z",
-  "name": "CyberArk Conjur Secret Lookup",
+  "name": "CyberArk Conjur Secrets Manager Lookup",
   "description": "",
   "organization": null,
   "credential_type": 20,

awx/ui/src/screens/InstanceGroup/ContainerGroupAdd/ContainerGroupAdd.js

@@ -29,6 +29,10 @@ function ContainerGroupAdd() {
     try {
       const { data: response } = await InstanceGroupsAPI.create({
         name: values.name,
+        max_forks: values.max_forks ? values.max_forks : 0,
+        max_concurrent_jobs: values.max_concurrent_jobs
+          ? values.max_concurrent_jobs
+          : 0,
         credential: values?.credential?.id,
         pod_spec_override: values.override
           ? getPodSpecValue(values.pod_spec_override)

awx/ui/src/screens/InstanceGroup/ContainerGroupAdd/ContainerGroupAdd.test.js

@@ -33,6 +33,8 @@ const initialPodSpec = {
 const instanceGroupCreateData = {
   name: 'Fuz',
   credential: { id: 71, name: 'CG' },
+  max_concurrent_jobs: 0,
+  max_forks: 0,
   pod_spec_override:
     'apiVersion: v1\nkind: Pod\nmetadata:\n  namespace: default\nspec:\n  containers:\n    - image: ansible/ansible-runner\n      tty: true\n      stdin: true\n      imagePullPolicy: Always\n      args:\n        - sleep\n        - infinity\n        - test',
 };

awx/ui/src/screens/InstanceGroup/ContainerGroupDetails/ContainerGroupDetails.js

@@ -9,7 +9,12 @@ import AlertModal from 'components/AlertModal';
 import ErrorDetail from 'components/ErrorDetail';
 import { CardBody, CardActionsRow } from 'components/Card';
 import DeleteButton from 'components/DeleteButton';
-import { Detail, DetailList, UserDateDetail } from 'components/DetailList';
+import {
+  Detail,
+  DetailList,
+  UserDateDetail,
+  DetailBadge,
+} from 'components/DetailList';
 import useRequest, { useDismissableError } from 'hooks/useRequest';
 import { jsonToYaml, isJsonString } from 'util/yaml';
 import { InstanceGroupsAPI } from 'api';
@@ -47,6 +52,20 @@ function ContainerGroupDetails({ instanceGroup }) {
           value={t`Container group`}
           dataCy="container-group-type"
         />
+        <DetailBadge
+          label={t`Max concurrent jobs`}
+          dataCy="instance-group-max-concurrent-jobs"
+          helpText={t`Maximum number of jobs to run concurrently on this group.
+          Zero means no limit will be enforced.`}
+          content={instanceGroup.max_concurrent_jobs}
+        />
+        <DetailBadge
+          label={t`Max forks`}
+          dataCy="instance-group-max-forks"
+          helpText={t`Maximum number of forks to allow across all jobs running concurrently on this group.
+          Zero means no limit will be enforced.`}
+          content={instanceGroup.max_forks}
+        />
         {instanceGroup.summary_fields.credential && (
           <Detail
             label={t`Credential`}

awx/ui/src/screens/InstanceGroup/ContainerGroupDetails/ContainerGroupDetails.test.js

@@ -23,6 +23,8 @@ const instanceGroup = {
   created: '2020-09-03T18:26:47.113934Z',
   modified: '2020-09-03T19:34:23.244694Z',
   capacity: 0,
+  max_concurrent_jobs: 0,
+  max_forks: 0,
   committed_capacity: 0,
   consumed_capacity: 0,
   percent_capacity_remaining: 0.0,

awx/ui/src/screens/InstanceGroup/ContainerGroupEdit/ContainerGroupEdit.js

@@ -39,6 +39,10 @@ function ContainerGroupEdit({ instanceGroup }) {
         name: values.name,
         credential: values.credential ? values.credential.id : null,
         pod_spec_override: values.override ? values.pod_spec_override : null,
+        max_forks: values.max_forks ? values.max_forks : 0,
+        max_concurrent_jobs: values.max_concurrent_jobs
+          ? values.max_concurrent_jobs
+          : 0,
         is_container_group: true,
       });
       history.push(detailsIUrl);

awx/ui/src/screens/InstanceGroup/ContainerGroupEdit/ContainerGroupEdit.test.js

@@ -34,6 +34,8 @@ const instanceGroup = {
   policy_instance_percentage: 0,
   policy_instance_minimum: 0,
   policy_instance_list: [],
+  max_concurrent_jobs: 0,
+  max_forks: 0,
   pod_spec_override: '',
   summary_fields: {
     credential: {
@@ -144,6 +146,8 @@ describe('<ContainerGroupEdit/>', () => {
       ...updatedInstanceGroup,
       credential: 12,
       pod_spec_override: null,
+      max_concurrent_jobs: 0,
+      max_forks: 0,
       is_container_group: true,
     });
     expect(history.location.pathname).toEqual(

awx/ui/src/screens/InstanceGroup/InstanceDetails/InstanceDetails.test.js

@@ -42,6 +42,8 @@ const instanceGroup = {
   credential: null,
   policy_instance_percentage: 100,
   policy_instance_minimum: 0,
+  max_concurrent_jobs: 0,
+  max_forks: 0,
   policy_instance_list: ['receptor-1', 'receptor-2'],
   pod_spec_override: '',
   summary_fields: {

awx/ui/src/screens/InstanceGroup/InstanceGroupDetails/InstanceGroupDetails.js

@@ -73,6 +73,20 @@ function InstanceGroupDetails({ instanceGroup }) {
           dataCy="instance-group-policy-instance-percentage"
           content={`${instanceGroup.policy_instance_percentage} %`}
         />
+        <DetailBadge
+          label={t`Max concurrent jobs`}
+          dataCy="instance-group-max-concurrent-jobs"
+          helpText={t`Maximum number of jobs to run concurrently on this group.
+          Zero means no limit will be enforced.`}
+          content={instanceGroup.max_concurrent_jobs}
+        />
+        <DetailBadge
+          label={t`Max forks`}
+          dataCy="instance-group-max-forks"
+          helpText={t`Maximum number of forks to allow across all jobs running concurrently on this group.
+          Zero means no limit will be enforced.`}
+          content={instanceGroup.max_forks}
+        />
         {instanceGroup.capacity ? (
           <DetailBadge
             label={t`Used capacity`}

awx/ui/src/screens/InstanceGroup/InstanceGroupDetails/InstanceGroupDetails.test.js

@@ -19,6 +19,8 @@ const instanceGroups = [
     policy_instance_minimum: 10,
     policy_instance_percentage: 50,
     percent_capacity_remaining: 60,
+    max_concurrent_jobs: 0,
+    max_forks: 0,
     is_container_group: false,
     created: '2020-07-21T18:41:02.818081Z',
     modified: '2020-07-24T20:32:03.121079Z',
@@ -38,6 +40,8 @@ const instanceGroups = [
     policy_instance_minimum: 0,
     policy_instance_percentage: 0,
     percent_capacity_remaining: 0,
+    max_concurrent_jobs: 0,
+    max_forks: 0,
     is_container_group: true,
     created: '2020-07-21T18:41:02.818081Z',
     modified: '2020-07-24T20:32:03.121079Z',

awx/ui/src/screens/InstanceGroup/shared/ContainerGroupForm.js

@@ -11,7 +11,7 @@ import FormField, {
   CheckboxField,
 } from 'components/FormField';
 import FormActionGroup from 'components/FormActionGroup';
-import { required } from 'util/validators';
+import { required, minMaxValue } from 'util/validators';
 import {
   FormColumnLayout,
   FormFullWidthLayout,
@@ -57,6 +57,26 @@ function ContainerGroupFormFields({ instanceGroup }) {
         tooltip={t`Credential to authenticate with Kubernetes or OpenShift. Must be of type "Kubernetes/OpenShift API Bearer Token". If left blank, the underlying Pod's service account will be used.`}
         autoPopulate={!instanceGroup?.id}
       />
+      <FormField
+        id="instance-group-max-concurrent-jobs"
+        label={t`Max concurrent jobs`}
+        name="max_concurrent_jobs"
+        type="number"
+        min="0"
+        validate={minMaxValue(0, 2147483647)}
+        tooltip={t`Maximum number of jobs to run concurrently on this group.
+          Zero means no limit will be enforced.`}
+      />
+      <FormField
+        id="instance-group-max-forks"
+        label={t`Max forks`}
+        name="max_forks"
+        type="number"
+        min="0"
+        validate={minMaxValue(0, 2147483647)}
+        tooltip={t`Maximum number of forks to allow across all jobs running concurrently on this group.
+          Zero means no limit will be enforced.`}
+      />
 
       <FormGroup fieldId="container-groups-option-checkbox" label={t`Options`}>
         <FormCheckboxLayout>
@@ -97,6 +117,8 @@ function ContainerGroupForm({
 
   const initialValues = {
     name: instanceGroup?.name || '',
+    max_concurrent_jobs: instanceGroup.max_concurrent_jobs || 0,
+    max_forks: instanceGroup.max_forks || 0,
     credential: instanceGroup?.summary_fields?.credential,
     pod_spec_override: isCheckboxChecked
       ? instanceGroup?.pod_spec_override

awx/ui/src/screens/InstanceGroup/shared/InstanceGroupForm.js

@@ -42,6 +42,26 @@ function InstanceGroupFormFields() {
           assigned to this group when new instances come online.`}
         validate={minMaxValue(0, 100)}
       />
+      <FormField
+        id="instance-group-max-concurrent-jobs"
+        label={t`Max concurrent jobs`}
+        name="max_concurrent_jobs"
+        type="number"
+        min="0"
+        validate={minMaxValue(0, 2147483647)}
+        tooltip={t`Maximum number of jobs to run concurrently on this group.
+          Zero means no limit will be enforced.`}
+      />
+      <FormField
+        id="instance-group-max-forks"
+        label={t`Max forks`}
+        name="max_forks"
+        type="number"
+        min="0"
+        validate={minMaxValue(0, 2147483647)}
+        tooltip={t`Maximum number of forks to allow across all jobs running concurrently on this group.
+          Zero means no limit will be enforced.`}
+      />
     </>
   );
 }
@@ -57,6 +77,8 @@ function InstanceGroupForm({
     name: instanceGroup.name || '',
     policy_instance_minimum: instanceGroup.policy_instance_minimum || 0,
     policy_instance_percentage: instanceGroup.policy_instance_percentage || 0,
+    max_concurrent_jobs: instanceGroup.max_concurrent_jobs || 0,
+    max_forks: instanceGroup.max_forks || 0,
   };
   return (
     <Formik

awx/ui/src/screens/Instances/InstanceDetail/InstanceDetail.js

@@ -1,6 +1,6 @@
 import React, { useCallback, useEffect, useState } from 'react';
 
-import { Link, useHistory, useParams } from 'react-router-dom';
+import { useHistory, useParams } from 'react-router-dom';
 import { t, Plural } from '@lingui/macro';
 import {
   Button,
@@ -11,7 +11,6 @@ import {
   CodeBlockCode,
   Tooltip,
   Slider,
-  Label,
 } from '@patternfly/react-core';
 import { DownloadIcon, OutlinedClockIcon } from '@patternfly/react-icons';
 import styled from 'styled-components';
@@ -34,6 +33,7 @@ import useRequest, {
   useDismissableError,
 } from 'hooks/useRequest';
 import HealthCheckAlert from 'components/HealthCheckAlert';
+import InstanceGroupLabels from 'components/InstanceGroupLabels';
 import RemoveInstanceButton from '../Shared/RemoveInstanceButton';
 
 const Unavailable = styled.span`
@@ -156,11 +156,6 @@ function InstanceDetail({ setBreadcrumb, isK8s }) {
     </>
   );
 
-  const buildLinkURL = (inst) =>
-    inst.is_container_group
-      ? '/instance_groups/container_group/'
-      : '/instance_groups/';
-
   const { error, dismissError } = useDismissableError(
     updateInstanceError || healthCheckError
   );
@@ -225,25 +220,9 @@ function InstanceDetail({ setBreadcrumb, isK8s }) {
                   label={t`Instance Groups`}
                   dataCy="instance-groups"
                   helpText={t`The Instance Groups to which this instance belongs.`}
-                  value={instanceGroups.map((ig) => (
-                    <React.Fragment key={ig.id}>
-                      <Label
-                        color="blue"
-                        isTruncated
-                        render={({ className, content, componentRef }) => (
-                          <Link
-                            to={`${buildLinkURL(ig)}${ig.id}/details`}
-                            className={className}
-                            innerRef={componentRef}
-                          >
-                            {content}
-                          </Link>
-                        )}
-                      >
-                        {ig.name}
-                      </Label>{' '}
-                    </React.Fragment>
-                  ))}
+                  value={
+                    <InstanceGroupLabels labels={instanceGroups} isLinkable />
+                  }
                   isEmpty={instanceGroups.length === 0}
                 />
               )}

awx/ui/src/screens/Inventory/InventoryDetail/InventoryDetail.js

@@ -23,6 +23,7 @@ import { InventoriesAPI } from 'api';
 import useRequest, { useDismissableError } from 'hooks/useRequest';
 import { Inventory } from 'types';
 import { relatedResourceDeleteRequests } from 'util/getRelatedResourceDeleteDetails';
+import InstanceGroupLabels from 'components/InstanceGroupLabels';
 import getHelpText from '../shared/Inventory.helptext';
 
 function InventoryDetail({ inventory }) {
@@ -105,23 +106,7 @@ function InventoryDetail({ inventory }) {
           <Detail
             fullWidth
             label={t`Instance Groups`}
-            value={
-              <ChipGroup
-                numChips={5}
-                totalChips={instanceGroups?.length}
-                ouiaId="instance-group-chips"
-              >
-                {instanceGroups?.map((ig) => (
-                  <Chip
-                    key={ig.id}
-                    isReadOnly
-                    ouiaId={`instance-group-${ig.id}-chip`}
-                  >
-                    {ig.name}
-                  </Chip>
-                ))}
-              </ChipGroup>
-            }
+            value={<InstanceGroupLabels labels={instanceGroups} isLinkable />}
             isEmpty={instanceGroups.length === 0}
           />
         )}

awx/ui/src/screens/Inventory/InventoryDetail/InventoryDetail.test.js

@@ -131,9 +131,8 @@ describe('<InventoryDetail />', () => {
     expect(InventoriesAPI.readInstanceGroups).toHaveBeenCalledWith(
       mockInventory.id
     );
-    const chip = wrapper.find('Chip').at(0);
-    expect(chip.prop('isReadOnly')).toEqual(true);
-    expect(chip.prop('children')).toEqual('Foo');
+    const label = wrapper.find('Label').at(0);
+    expect(label.prop('children')).toEqual('Foo');
   });
 
   test('should not load instance groups', async () => {

awx/ui/src/screens/Inventory/SmartInventoryDetail/SmartInventoryDetail.js

@@ -2,7 +2,7 @@ import React, { useCallback, useEffect } from 'react';
 import { Link, useHistory } from 'react-router-dom';
 
 import { t } from '@lingui/macro';
-import { Button, Chip, Label } from '@patternfly/react-core';
+import { Button, Label } from '@patternfly/react-core';
 
 import { Inventory } from 'types';
 import { InventoriesAPI, UnifiedJobsAPI } from 'api';
@@ -10,7 +10,6 @@ import useRequest, { useDismissableError } from 'hooks/useRequest';
 
 import AlertModal from 'components/AlertModal';
 import { CardBody, CardActionsRow } from 'components/Card';
-import ChipGroup from 'components/ChipGroup';
 import { VariablesDetail } from 'components/CodeEditor';
 import ContentError from 'components/ContentError';
 import ContentLoading from 'components/ContentLoading';
@@ -18,6 +17,7 @@ import DeleteButton from 'components/DeleteButton';
 import { DetailList, Detail, UserDateDetail } from 'components/DetailList';
 import ErrorDetail from 'components/ErrorDetail';
 import Sparkline from 'components/Sparkline';
+import InstanceGroupLabels from 'components/InstanceGroupLabels';
 
 function SmartInventoryDetail({ inventory }) {
   const history = useHistory();
@@ -120,23 +120,7 @@ function SmartInventoryDetail({ inventory }) {
           <Detail
             fullWidth
             label={t`Instance groups`}
-            value={
-              <ChipGroup
-                numChips={5}
-                totalChips={instanceGroups.length}
-                ouiaId="instance-group-chips"
-              >
-                {instanceGroups.map((ig) => (
-                  <Chip
-                    key={ig.id}
-                    isReadOnly
-                    ouiaId={`instance-group-${ig.id}-chip`}
-                  >
-                    {ig.name}
-                  </Chip>
-                ))}
-              </ChipGroup>
-            }
+            value={<InstanceGroupLabels labels={instanceGroups} />}
             isEmpty={instanceGroups.length === 0}
           />
           <VariablesDetail

awx/ui/src/screens/Job/JobOutput/HostEventModal.js

@@ -70,7 +70,6 @@ const getStdOutValue = (hostEvent) => {
 function HostEventModal({ onClose, hostEvent = {}, isOpen = false }) {
   const [hostStatus, setHostStatus] = useState(null);
   const [activeTabKey, setActiveTabKey] = useState(0);
-
   useEffect(() => {
     setHostStatus(processEventStatus(hostEvent));
   }, [setHostStatus, hostEvent]);
@@ -108,11 +107,11 @@ function HostEventModal({ onClose, hostEvent = {}, isOpen = false }) {
             style={{ alignItems: 'center', marginTop: '20px' }}
             gutter="sm"
           >
-            <Detail label={t`Host`} value={hostEvent.host_name} />
-            {hostEvent.summary_fields.host?.description ? (
+            <Detail label={t`Host`} value={hostEvent.event_data?.host} />
+            {hostEvent.summary_fields?.host?.description ? (
               <Detail
                 label={t`Description`}
-                value={hostEvent.summary_fields.host.description}
+                value={hostEvent.summary_fields?.host?.description}
               />
             ) : null}
             {hostStatus ? (
@@ -125,12 +124,9 @@ function HostEventModal({ onClose, hostEvent = {}, isOpen = false }) {
             <Detail label={t`Task`} value={hostEvent.task} />
             <Detail
               label={t`Module`}
-              value={hostEvent.event_data.task_action || t`No result found`}
-            />
-            <Detail
-              label={t`Command`}
-              value={hostEvent?.event_data?.res?.cmd}
+              value={hostEvent.event_data?.task_action || t`No result found`}
             />
+            <Detail label={t`Command`} value={hostEvent.event_data?.res?.cmd} />
           </DetailList>
         </Tab>
         <Tab

awx/ui/src/screens/Job/JobOutput/HostEventModal.test.js

@@ -52,6 +52,47 @@ const hostEvent = {
   },
 };
 
+const partialHostEvent = {
+  changed: true,
+  event: 'runner_on_ok',
+  event_data: {
+    host: 'foo',
+    play: 'all',
+    playbook: 'run_command.yml',
+    res: {
+      ansible_loop_var: 'item',
+      changed: true,
+      item: '1',
+      msg: 'This is a debug message: 1',
+      stdout:
+        '              total        used        free      shared  buff/cache   available\nMem:           7973        3005         960          30        4007        4582\nSwap:          1023           0        1023',
+      stderr: 'problems',
+      cmd: ['free', '-m'],
+      stderr_lines: [],
+      stdout_lines: [
+        '              total        used        free      shared  buff/cache   available',
+        'Mem:           7973        3005         960          30        4007        4582',
+        'Swap:          1023           0        1023',
+      ],
+    },
+    task: 'command',
+    task_action: 'command',
+  },
+  event_display: 'Host OK',
+  event_level: 3,
+  failed: false,
+  host: 1,
+  id: 123,
+  job: 4,
+  play: 'all',
+  playbook: 'run_command.yml',
+  stdout: `stdout: "changed: [localhost] => {"changed": true, "cmd": ["free", "-m"], "delta": "0:00:01.479609", "end": "2019-09-10 14:21:45.469533", "rc": 0, "start": "2019-09-10 14:21:43.989924", "stderr": "", "stderr_lines": [], "stdout": "              total        used        free      shared  buff/cache   available\nMem:           7973        3005         960          30        4007        4582\nSwap:          1023           0        1023", "stdout_lines": ["              total        used        free      shared  buff/cache   available", "Mem:           7973        3005         960          30        4007        4582", "Swap:          1023           0        1023"]}"
+  `,
+  task: 'command',
+  type: 'job_event',
+  url: '/api/v2/job_events/123/',
+};
+
 /*
 Some libraries return a list of string in stdout
 Example: https://github.com/ansible-collections/cisco.ios/blob/main/plugins/modules/ios_command.py#L124-L128
@@ -134,6 +175,13 @@ describe('HostEventModal', () => {
     expect(wrapper).toHaveLength(1);
   });
 
+  test('renders successfully with partial data', () => {
+    const wrapper = shallow(
+      <HostEventModal hostEvent={partialHostEvent} onClose={() => {}} />
+    );
+    expect(wrapper).toHaveLength(1);
+  });
+
   test('should render all tabs', () => {
     const wrapper = shallow(
       <HostEventModal hostEvent={hostEvent} onClose={() => {}} isOpen />

awx/ui/src/screens/Job/useWsJob.js

@@ -4,6 +4,7 @@ import { getJobModel } from 'util/jobs';
 
 export default function useWsJob(initialJob) {
   const [job, setJob] = useState(initialJob);
+  const [pendingMessages, setPendingMessages] = useState([]);
   const lastMessage = useWebsocket({
     jobs: ['status_changed'],
     control: ['limit_reached_1'],
@@ -13,30 +14,48 @@ export default function useWsJob(initialJob) {
     setJob(initialJob);
   }, [initialJob]);
 
+  const processMessage = (message) => {
+    if (message.unified_job_id !== job.id) {
+      return;
+    }
+
+    if (
+      ['successful', 'failed', 'error', 'cancelled'].includes(message.status)
+    ) {
+      fetchJob();
+    }
+    setJob(updateJob(job, message));
+  };
+
+  async function fetchJob() {
+    const { data } = await getJobModel(job.type).readDetail(job.id);
+    setJob(data);
+  }
+
   useEffect(
     () => {
-      async function fetchJob() {
-        const { data } = await getJobModel(job.type).readDetail(job.id);
-        setJob(data);
-      }
-
-      if (!job || lastMessage?.unified_job_id !== job.id) {
+      if (!lastMessage) {
         return;
       }
-
-      if (
-        ['successful', 'failed', 'error', 'cancelled'].includes(
-          lastMessage.status
-        )
-      ) {
-        fetchJob();
-      } else {
-        setJob(updateJob(job, lastMessage));
+      if (job) {
+        processMessage(lastMessage);
+      } else if (lastMessage.unified_job_id) {
+        setPendingMessages(pendingMessages.concat(lastMessage));
       }
     },
     [lastMessage] // eslint-disable-line react-hooks/exhaustive-deps
   );
 
+  useEffect(() => {
+    if (!job || !pendingMessages.length) {
+      return;
+    }
+    pendingMessages.forEach((message) => {
+      processMessage(message);
+    });
+    setPendingMessages([]);
+  }, [job, pendingMessages]); // eslint-disable-line react-hooks/exhaustive-deps
+
   return job;
 }
 

awx/ui/src/screens/Organization/OrganizationDetail/OrganizationDetail.js

@@ -2,7 +2,7 @@ import React, { useEffect, useState, useCallback } from 'react';
 import { Link, useHistory, useRouteMatch } from 'react-router-dom';
 
 import { t } from '@lingui/macro';
-import { Button, Chip } from '@patternfly/react-core';
+import { Button } from '@patternfly/react-core';
 import { OrganizationsAPI } from 'api';
 import { DetailList, Detail, UserDateDetail } from 'components/DetailList';
 import { CardBody, CardActionsRow } from 'components/Card';
@@ -16,6 +16,7 @@ import ErrorDetail from 'components/ErrorDetail';
 import useRequest, { useDismissableError } from 'hooks/useRequest';
 import { useConfig } from 'contexts/Config';
 import ExecutionEnvironmentDetail from 'components/ExecutionEnvironmentDetail';
+import InstanceGroupLabels from 'components/InstanceGroupLabels';
 import { relatedResourceDeleteRequests } from 'util/getRelatedResourceDeleteDetails';
 
 function OrganizationDetail({ organization }) {
@@ -79,11 +80,6 @@ function OrganizationDetail({ organization }) {
     return <ContentError error={contentError} />;
   }
 
-  const buildLinkURL = (instance) =>
-    instance.is_container_group
-      ? '/instance_groups/container_group/'
-      : '/instance_groups/';
-
   return (
     <CardBody>
       <DetailList>
@@ -126,25 +122,7 @@ function OrganizationDetail({ organization }) {
             fullWidth
             label={t`Instance Groups`}
             helpText={t`The Instance Groups for this Organization to run on.`}
-            value={
-              <ChipGroup
-                numChips={5}
-                totalChips={instanceGroups.length}
-                ouiaId="instance-group-chips"
-              >
-                {instanceGroups.map((ig) => (
-                  <Link to={`${buildLinkURL(ig)}${ig.id}/details`} key={ig.id}>
-                    <Chip
-                      key={ig.id}
-                      isReadOnly
-                      ouiaId={`instance-group-${ig.id}-chip`}
-                    >
-                      {ig.name}
-                    </Chip>
-                  </Link>
-                ))}
-              </ChipGroup>
-            }
+            value={<InstanceGroupLabels labels={instanceGroups} isLinkable />}
             isEmpty={instanceGroups.length === 0}
           />
         )}

awx/ui/src/screens/Organization/OrganizationDetail/OrganizationDetail.test.js

@@ -90,7 +90,7 @@ describe('<OrganizationDetail />', () => {
     await waitForElement(component, 'ContentLoading', (el) => el.length === 0);
     expect(
       component
-        .find('Chip')
+        .find('Label')
         .findWhere((el) => el.text() === 'One')
         .exists()
     ).toBe(true);

awx/ui/src/screens/Template/JobTemplateDetail/JobTemplateDetail.js

@@ -34,6 +34,7 @@ import useRequest, { useDismissableError } from 'hooks/useRequest';
 import useBrandName from 'hooks/useBrandName';
 import ExecutionEnvironmentDetail from 'components/ExecutionEnvironmentDetail';
 import { relatedResourceDeleteRequests } from 'util/getRelatedResourceDeleteDetails';
+import InstanceGroupLabels from 'components/InstanceGroupLabels';
 import getHelpText from '../shared/JobTemplate.helptext';
 
 function JobTemplateDetail({ template }) {
@@ -167,11 +168,6 @@ function JobTemplateDetail({ template }) {
     );
   };
 
-  const buildLinkURL = (instance) =>
-    instance.is_container_group
-      ? '/instance_groups/container_group/'
-      : '/instance_groups/';
-
   if (instanceGroupsError) {
     return <ContentError error={instanceGroupsError} />;
   }
@@ -422,25 +418,7 @@ function JobTemplateDetail({ template }) {
           label={t`Instance Groups`}
           dataCy="jt-detail-instance-groups"
           helpText={helpText.instanceGroups}
-          value={
-            <ChipGroup
-              numChips={5}
-              totalChips={instanceGroups.length}
-              ouiaId="instance-group-chips"
-            >
-              {instanceGroups.map((ig) => (
-                <Link to={`${buildLinkURL(ig)}${ig.id}/details`} key={ig.id}>
-                  <Chip
-                    key={ig.id}
-                    ouiaId={`instance-group-${ig.id}-chip`}
-                    isReadOnly
-                  >
-                    {ig.name}
-                  </Chip>
-                </Link>
-              ))}
-            </ChipGroup>
-          }
+          value={<InstanceGroupLabels labels={instanceGroups} isLinkable />}
           isEmpty={instanceGroups.length === 0}
         />
         {job_tags && (

awx_collection/plugins/modules/credential.py

@@ -52,7 +52,7 @@ options:
         - The credential type being created.
         - Can be a built-in credential type such as "Machine", or a custom credential type such as "My Credential Type"
         - Choices include Amazon Web Services, Ansible Galaxy/Automation Hub API Token, Centrify Vault Credential Provider Lookup,
-          Container Registry, CyberArk AIM Central Credential Provider Lookup, CyberArk Conjur Secret Lookup, Google Compute Engine,
+          Container Registry, CyberArk AIM Central Credential Provider Lookup, CyberArk Conjur Secrets Manager Lookup, Google Compute Engine,
           GitHub Personal Access Token, GitLab Personal Access Token, GPG Public Key, HashiCorp Vault Secret Lookup, HashiCorp Vault Signed SSH,
           Insights, Machine, Microsoft Azure Key Vault, Microsoft Azure Resource Manager, Network, OpenShift or Kubernetes API
           Bearer Token, OpenStack, Red Hat Ansible Automation Platform, Red Hat Satellite 6, Red Hat Virtualization, Source Control,

awx_collection/plugins/modules/export.py

@@ -28,52 +28,64 @@ options:
       default: 'False'
     organizations:
       description:
-        - organization name to export
-      type: str
+        - organization names to export
+      type: list
+      elements: str
     users:
       description:
-        - user name to export
-      type: str
+        - user names to export
+      type: list
+      elements: str
     teams:
       description:
-        - team name to export
-      type: str
+        - team names to export
+      type: list
+      elements: str
     credential_types:
       description:
-        - credential type name to export
-      type: str
+        - credential type names to export
+      type: list
+      elements: str
     credentials:
       description:
-        - credential name to export
-      type: str
+        - credential names to export
+      type: list
+      elements: str
     execution_environments:
       description:
-        - execution environment name to export
-      type: str
+        - execution environment names to export
+      type: list
+      elements: str
     notification_templates:
       description:
-        - notification template name to export
-      type: str
+        - notification template names to export
+      type: list
+      elements: str
     inventory_sources:
       description:
-        - inventory soruce to export
-      type: str
+        - inventory soruces to export
+      type: list
+      elements: str
     inventory:
       description:
-        - inventory name to export
-      type: str
+        - inventory names to export
+      type: list
+      elements: str
     projects:
       description:
-        - project name to export
-      type: str
+        - project names to export
+      type: list
+      elements: str
     job_templates:
       description:
-        - job template name to export
-      type: str
+        - job template names to export
+      type: list
+      elements: str
     workflow_job_templates:
       description:
-        - workflow name to export
-      type: str
+        - workflow names to export
+      type: list
+      elements: str
 requirements:
   - "awxkit >= 9.3.0"
 notes:
@@ -94,6 +106,10 @@ EXAMPLES = '''
   export:
     job_templates: "My Template"
     credential: 'all'
+
+- name: Export a list of inventories
+  export:
+    inventory: ['My Inventory 1', 'My Inventory 2']
 '''
 
 import logging
@@ -111,24 +127,12 @@ except ImportError:
 def main():
     argument_spec = dict(
         all=dict(type='bool', default=False),
-        credential_types=dict(type='str'),
-        credentials=dict(type='str'),
-        execution_environments=dict(type='str'),
-        inventory=dict(type='str'),
-        inventory_sources=dict(type='str'),
-        job_templates=dict(type='str'),
-        notification_templates=dict(type='str'),
-        organizations=dict(type='str'),
-        projects=dict(type='str'),
-        teams=dict(type='str'),
-        users=dict(type='str'),
-        workflow_job_templates=dict(type='str'),
     )
 
     # We are not going to raise an error here because the __init__ method of ControllerAWXKitModule will do that for us
     if HAS_EXPORTABLE_RESOURCES:
         for resource in EXPORTABLE_RESOURCES:
-            argument_spec[resource] = dict(type='str')
+            argument_spec[resource] = dict(type='list', elements='str')
 
     module = ControllerAWXKitModule(argument_spec=argument_spec)
 

awx_collection/plugins/modules/instance_group.py

@@ -54,6 +54,18 @@ options:
       required: False
       type: int
       default: '0'
+    max_concurrent_jobs:
+      description:
+        - Maximum number of concurrent jobs to run on this group. Zero means no limit.
+      required: False
+      type: int
+      default: '0'
+    max_forks:
+      description:
+        - Max forks to execute on this group. Zero means no limit.
+      required: False
+      type: int
+      default: '0'
     policy_instance_list:
       description:
         - List of exact-match Instances that will be assigned to this group
@@ -95,6 +107,8 @@ def main():
         is_container_group=dict(type='bool', default=False),
         policy_instance_percentage=dict(type='int', default='0'),
         policy_instance_minimum=dict(type='int', default='0'),
+        max_concurrent_jobs=dict(type='int', default='0'),
+        max_forks=dict(type='int', default='0'),
         policy_instance_list=dict(type='list', elements='str'),
         pod_spec_override=dict(),
         instances=dict(required=False, type="list", elements='str', default=None),
@@ -111,6 +125,8 @@ def main():
     is_container_group = module.params.get('is_container_group')
     policy_instance_percentage = module.params.get('policy_instance_percentage')
     policy_instance_minimum = module.params.get('policy_instance_minimum')
+    max_concurrent_jobs = module.params.get('max_concurrent_jobs')
+    max_forks = module.params.get('max_forks')
     policy_instance_list = module.params.get('policy_instance_list')
     pod_spec_override = module.params.get('pod_spec_override')
     instances = module.params.get('instances')
@@ -144,6 +160,10 @@ def main():
         new_fields['policy_instance_percentage'] = policy_instance_percentage
     if policy_instance_minimum is not None:
         new_fields['policy_instance_minimum'] = policy_instance_minimum
+    if max_concurrent_jobs is not None:
+        new_fields['max_concurrent_jobs'] = max_concurrent_jobs
+    if max_forks is not None:
+        new_fields['max_forks'] = max_forks
     if policy_instance_list is not None:
         new_fields['policy_instance_list'] = policy_instance_list
     if pod_spec_override is not None:

awx_collection/plugins/modules/job_template.py

@@ -84,7 +84,7 @@ options:
       type: str
     execution_environment:
       description:
-        - Execution Environment to use for the JT.
+        - Execution Environment to use for the job template.
       type: str
     custom_virtualenv:
       description:

awx_collection/plugins/modules/project_update.py

@@ -114,7 +114,12 @@ def main():
     # Update the project
     result = module.post_endpoint(project['related']['update'])
 
-    if result['status_code'] != 202:
+    if result['status_code'] == 405:
+        module.fail_json(
+            msg="Unable to trigger a project update because the project scm_type ({0}) does not support it.".format(project['scm_type']),
+            response=result
+        )
+    elif result['status_code'] != 202:
         module.fail_json(msg="Failed to update project, see response for details", response=result)
 
     module.json_output['changed'] = True