Configure redis coordination backend for orquesta integration tests

.github/workflows/ci.yaml

@@ -20,8 +20,25 @@ on:
     # run every night at midnight
     - cron:  '0 0 * * *'
 
+# TODO: Our workflow is far from ideal. We need to refactor it into multiple
+# ones and only run commands which are needed for some steps for those steps and
+# not for all
 jobs:
+  pre_job:
+    name: Skip Duplicate Jobs Pre Job
+    runs-on: ubuntu-latest
+    outputs:
+      should_skip: ${{ steps.skip_check.outputs.should_skip }}
+    steps:
+      - id: skip_check
+        uses: fkirc/skip-duplicate-actions@4c656bbdb6906310fa6213604828008bc28fe55d # v3.3.0
+        with:
+          cancel_others: 'true'
+          github_token: ${{ github.token }}
+
   ci:
+    needs: pre_job
+    if: ${{ needs.pre_job.outputs.should_skip != 'true' }}
     name: '${{ matrix.name }} - python (${{ matrix.python-version }})'
     runs-on: ubuntu-latest
     strategy:
@@ -40,13 +57,13 @@ jobs:
           - name: 'Unit Tests'
             task: 'ci-unit'
             python-version: '3.6'
+          - name: 'Integration Tests'
+            task: 'ci-integration'
+            python-version: '3.6'
           # This job is slow so we only run in on a daily basis
           # - name: 'Micro Benchmarks'
           #   task: 'micro-benchmarks'
           #   python-version: '3.6'
-          - name: 'Integration Tests'
-            task: 'ci-integration'
-            python-version: '3.6'
     services:
       mongo:
         image: mongo:4.0
@@ -82,6 +99,23 @@ jobs:
           #- 15671:15671/tcp # Management: SSL port
           #- 25672:25672/tcp # inter-node or CLI
           #- 4369:4369/tcp   # epmd
+          #
+
+      # Used for the coordination backend for integration tests
+      # TODO: Only start this for integration tests via job step
+      # https://github.bokerqi.topmunity/t/conditional-services-in-a-job/135301/3
+      redis:
+        # Docker Hub image
+        image: redis
+        # Set health checks to wait until redis has started
+        options: >-
+          --name "redis"
+          --health-cmd "redis-cli ping"
+          --health-interval 10s
+          --health-timeout 5s
+          --health-retries 5
+        ports:
+          - 6379:6379/tcp
 
     env:
       TASK: '${{ matrix.task }}'
@@ -191,7 +225,7 @@ jobs:
       - name: Install requirements
         run: |
           ./scripts/ci/install-requirements.sh
-      - name: Setup integration tests
+      - name: Setup Integration Tests
         run: |
           # prep a ci-specific dev conf file that uses runner instead of stanley
           # this user is the username of the user in GitHub actions, used for SSH, etc during

scripts/github/prepare-integration.sh

@@ -10,6 +10,25 @@ fi
 # shellcheck disable=SC1091
 source ./virtualenv/bin/activate
 
+# Enable coordination backend to avoid race conditions with orquesta tests due
+# to the lack of the coordination backend
+sed -i "s#\#url = redis://localhost#url = redis://127.0.0.1#g" ./conf/st2.dev.conf
+sed -i "s#\#url = redis://localhost#url = redis://127.0.0.1#g" ./conf/st2.ci.conf || true
+
+echo "Used config for the tests"
+echo ""
+echo "st2.dev.conf"
+echo ""
+cat conf/st2.dev.conf
+echo ""
+echo "st2.ci.conf"
+echo ""
+cat conf/st2.ci.conf || true
+echo ""
+
+# Needed by the coordination backend
+pip install "redis==3.5.3"
+
 # install st2 client
 python ./st2client/setup.py develop
 st2 --version

st2common/st2common/service_setup.py

@@ -44,6 +44,7 @@
 from st2common.services import coordination
 from st2common.logging.misc import add_global_filters_for_all_loggers
 from st2common.constants.error_messages import PYTHON2_DEPRECATION
+from st2common.services.coordination import get_driver_name
 
 # Note: This is here for backward compatibility.
 # Function has been moved in a standalone module to avoid expensive in-direct
@@ -147,6 +148,9 @@ def setup(
 
     LOG.info("Using logging config: %s", logging_config_path)
 
+    LOG.info("Using coordination driver: %s", get_driver_name())
+    LOG.info("Using metrics driver: %s", cfg.CONF.metrics.driver)
+
     is_debug_enabled = cfg.CONF.debug or cfg.CONF.system.debug
 
     try:

st2common/st2common/services/coordination.py

@@ -175,6 +175,19 @@ def configured():
     return backend_configured and not mock_backend
 
 
+def get_driver_name() -> str:
+    """
+    Return coordination driver name (aka protocol part from the URI / URL).
+    """
+    url = cfg.CONF.coordination.url
+
+    if not url:
+        return None
+
+    driver_name = url.split("://")[0]
+    return driver_name
+
+
 def coordinator_setup(start_heart=True):
     """
     Sets up the client for the coordination service.

st2common/tests/integration/test_service_setup_log_level_filtering.py

@@ -74,11 +74,13 @@ def test_audit_log_level_is_filtered_if_log_level_is_not_debug_or_audit(self):
         process.send_signal(signal.SIGKILL)
 
         # Verify first 4 environment related log messages
-        stdout = "\n".join(process.stdout.read().decode("utf-8").split("\n")[:4])
+        stdout = "\n".join(process.stdout.read().decode("utf-8").split("\n")[:6])
         self.assertIn("INFO [-] Using Python:", stdout)
         self.assertIn("INFO [-] Using fs encoding:", stdout)
         self.assertIn("INFO [-] Using config files:", stdout)
         self.assertIn("INFO [-] Using logging config:", stdout)
+        self.assertIn("INFO [-] Using coordination driver:", stdout)
+        self.assertIn("INFO [-] Using metrics driver:", stdout)
 
         # 1. INFO log level - audit messages should not be included
         process = self._start_process(config_path=ST2_CONFIG_INFO_LL_PATH)
@@ -88,8 +90,8 @@ def test_audit_log_level_is_filtered_if_log_level_is_not_debug_or_audit(self):
         eventlet.sleep(3)
         process.send_signal(signal.SIGKILL)
 
-        # First 4 log lines are debug messages about the environment which are always logged
-        stdout = "\n".join(process.stdout.read().decode("utf-8").split("\n")[4:])
+        # First 6 log lines are debug messages about the environment which are always logged
+        stdout = "\n".join(process.stdout.read().decode("utf-8").split("\n")[6:])
 
         self.assertIn("INFO [-]", stdout)
         self.assertNotIn("DEBUG [-]", stdout)
@@ -103,8 +105,8 @@ def test_audit_log_level_is_filtered_if_log_level_is_not_debug_or_audit(self):
         eventlet.sleep(5)
         process.send_signal(signal.SIGKILL)
 
-        # First 4 log lines are debug messages about the environment which are always logged
-        stdout = "\n".join(process.stdout.read().decode("utf-8").split("\n")[4:])
+        # First 6 log lines are debug messages about the environment which are always logged
+        stdout = "\n".join(process.stdout.read().decode("utf-8").split("\n")[6:])
 
         self.assertIn("INFO [-]", stdout)
         self.assertIn("DEBUG [-]", stdout)
@@ -118,8 +120,8 @@ def test_audit_log_level_is_filtered_if_log_level_is_not_debug_or_audit(self):
         eventlet.sleep(5)
         process.send_signal(signal.SIGKILL)
 
-        # First 4 log lines are debug messages about the environment which are always logged
-        stdout = "\n".join(process.stdout.read().decode("utf-8").split("\n")[4:])
+        # First 6 log lines are debug messages about the environment which are always logged
+        stdout = "\n".join(process.stdout.read().decode("utf-8").split("\n")[6:])
 
         self.assertNotIn("INFO [-]", stdout)
         self.assertNotIn("DEBUG [-]", stdout)
@@ -133,8 +135,8 @@ def test_audit_log_level_is_filtered_if_log_level_is_not_debug_or_audit(self):
         eventlet.sleep(5)
         process.send_signal(signal.SIGKILL)
 
-        # First 4 log lines are debug messages about the environment which are always logged
-        stdout = "\n".join(process.stdout.read().decode("utf-8").split("\n")[4:])
+        # First 6 log lines are debug messages about the environment which are always logged
+        stdout = "\n".join(process.stdout.read().decode("utf-8").split("\n")[6:])
 
         self.assertIn("INFO [-]", stdout)
         self.assertIn("DEBUG [-]", stdout)

st2common/tests/unit/services/test_synchronization.py

@@ -39,16 +39,28 @@ def tearDownClass(cls):
         super(SynchronizationTest, cls).tearDownClass()
 
     def test_service_configured(self):
+        cfg.CONF.set_override(name="url", override=None, group="coordination")
+        self.assertEqual(coordination.get_driver_name(), None)
+
         cfg.CONF.set_override(
             name="url", override="kazoo://127.0.0.1:2181", group="coordination"
         )
         self.assertTrue(coordination.configured())
+        self.assertEqual(coordination.get_driver_name(), "kazoo")
 
         cfg.CONF.set_override(name="url", override="file:///tmp", group="coordination")
         self.assertFalse(coordination.configured())
+        self.assertEqual(coordination.get_driver_name(), "file")
 
         cfg.CONF.set_override(name="url", override="zake://", group="coordination")
         self.assertFalse(coordination.configured())
+        self.assertEqual(coordination.get_driver_name(), "zake")
+
+        cfg.CONF.set_override(
+            name="url", override="redis://foo:bar@127.0.0.1", group="coordination"
+        )
+        self.assertTrue(coordination.configured())
+        self.assertEqual(coordination.get_driver_name(), "redis")
 
     def test_lock(self):
         name = uuid.uuid4().hex