chore: fix CI pipeline

.github/workflows/ci.yaml

@@ -64,12 +64,16 @@ jobs:
 
   integration-test:
     strategy:
-      fail-fast: true
+      fail-fast: false
       matrix:
-        bases:         
+        bases:
           - [email protected]
-    name: Integration tests (LXD) | ${{ matrix.bases }}
+        local: [true, false]
+    name: Integration tests (LXD) ${{ matrix.local && '|' || '| Charmhub (edge) |'}} ${{ matrix.bases }}
     runs-on: ubuntu-latest
+    # Testing against Charmhub will probably yield errors when doing breaking changes, so don't
+    # block CI on that.
+    continue-on-error: ${{ !matrix.local }}
     needs:
       - inclusive-naming-check
       - lint
@@ -78,10 +82,33 @@ jobs:
     steps:
       - name: Checkout
         uses: actions/checkout@v4
+        with:
+          path: main
+      - name: Fetch slurmd
+        uses: actions/checkout@v4
+        if: ${{ matrix.local }}
+        with:
+          repository: charmed-hpc/slurmd-operator
+          path: slurmd-operator
+      - name: Fetch slurmdbd
+        uses: actions/checkout@v4
+        if: ${{ matrix.local }}
+        with:
+          repository: charmed-hpc/slurmdbd-operator
+          path: slurmdbd-operator
+      - name: Fetch slurmrestd
+        uses: actions/checkout@v4
+        if: ${{ matrix.local }}
+        with:
+          repository: charmed-hpc/slurmrestd-operator
+          path: slurmrestd-operator
       - name: Setup operator environment
         uses: charmed-kubernetes/actions-operator@main
         with:
           provider: lxd
-          juju-channel: 3.1/stable
+          juju-channel: 3.4/stable
       - name: Run tests
-        run: tox run -e integration -- --charm-base=${{ matrix.bases }}
+        run: |
+          cd main && tox run -e integration -- \
+             --charm-base=${{ matrix.bases }} \
+             ${{ matrix.local && '--use-local' || '' }}

.github/workflows/release.yaml

@@ -32,10 +32,10 @@ jobs:
       - name: Checkout
         uses: actions/checkout@v3
       - name: Select charmhub channel
-        uses: canonical/charming-actions/channel@2.2.0
+        uses: canonical/charming-actions/channel@2.5.0-rc
         id: channel
       - name: Upload charm to charmhub
-        uses: canonical/charming-actions/upload-charm@2.2.0
+        uses: canonical/charming-actions/upload-charm@2.5.0-rc
         with:
           credentials: "${{ secrets.CHARMCRAFT_AUTH }}"
           github-token: "${{ secrets.GITHUB_TOKEN }}"

tests/integration/conftest.py

@@ -26,6 +26,7 @@
 logger = logging.getLogger(__name__)
 SLURMD_DIR = Path(os.getenv("SLURMD_DIR", "../slurmd-operator"))
 SLURMDBD_DIR = Path(os.getenv("SLURMDBD_DIR", "../slurmdbd-operator"))
+SLURMRESTD_DIR = Path(os.getenv("SLURMRESTD_DIR", "../slurmrestd-operator"))
 
 
 def pytest_addoption(parser) -> None:
@@ -93,3 +94,23 @@ async def slurmdbd_charm(request, ops_test: OpsTest) -> Union[str, Path]:
             )
 
     return "slurmdbd"
+
+
+@pytest.fixture(scope="module")
+async def slurmrestd_charm(request, ops_test: OpsTest) -> Union[str, Path]:
+    """Pack slurmrestd charm to use for integration tests when --use-local is specified.
+
+    Returns:
+        `str` "slurmrestd" if --use-local not specified or if SLURMRESTD_DIR does not exist.
+    """
+    if request.config.option.use_local:
+        logger.info("Using local slurmrestd operator rather than pulling from Charmhub")
+        if SLURMRESTD_DIR.exists():
+            return await ops_test.build_charm(SLURMRESTD_DIR)
+        else:
+            logger.warning(
+                f"{SLURMRESTD_DIR} not found. "
+                f"Defaulting to latest/edge slurmrestd operator from Charmhub"
+            )
+
+    return "slurmrestd"

tests/integration/test_charm.py

@@ -27,22 +27,29 @@
 SLURMCTLD = "slurmctld"
 SLURMD = "slurmd"
 SLURMDBD = "slurmdbd"
+SLURMRESTD = "slurmrestd"
 DATABASE = "mysql"
 ROUTER = "mysql-router"
-UNIT_NAME = f"{SLURMCTLD}/0"
 
 
 @pytest.mark.abort_on_fail
 @pytest.mark.skip_if_deployed
 @pytest.mark.order(1)
 async def test_build_and_deploy_against_edge(
-    ops_test: OpsTest, charm_base: str, slurmctld_charm, slurmd_charm, slurmdbd_charm
+    ops_test: OpsTest,
+    charm_base: str,
+    slurmctld_charm,
+    slurmd_charm,
+    slurmdbd_charm,
+    slurmrestd_charm,
 ) -> None:
-    """Test that the slurmctld charm can stabilize against slurmd, slurmdbd, and MySQL."""
-    logger.info(f"Deploying {SLURMCTLD} against {SLURMD}, {SLURMDBD}, and {DATABASE}")
+    """Test that the slurmctld charm can stabilize against slurmd, slurmdbd, slurmrestd, and MySQL."""
+    logger.info(
+        f"Deploying {SLURMCTLD} against {SLURMD}, {SLURMDBD}, {SLURMRESTD}, and {DATABASE}"
+    )
     # Pack charms and download NHC resource for the slurmd operator.
-    slurmctld, slurmd, slurmdbd = await asyncio.gather(
-        slurmctld_charm, slurmd_charm, slurmdbd_charm
+    slurmctld, slurmd, slurmdbd, slurmrestd = await asyncio.gather(
+        slurmctld_charm, slurmd_charm, slurmdbd_charm, slurmrestd_charm
     )
     # Deploy the test Charmed SLURM cloud.
     await asyncio.gather(
@@ -66,6 +73,13 @@ async def test_build_and_deploy_against_edge(
             num_units=1,
             base=charm_base,
         ),
+        ops_test.model.deploy(
+            str(slurmrestd),
+            application_name=SLURMRESTD,
+            channel="edge" if isinstance(slurmrestd, str) else None,
+            num_units=1,
+            base=charm_base,
+        ),
         ops_test.model.deploy(
             ROUTER,
             application_name=f"{SLURMDBD}-{ROUTER}",
@@ -84,12 +98,13 @@ async def test_build_and_deploy_against_edge(
     # Set integrations for charmed applications.
     await ops_test.model.integrate(f"{SLURMCTLD}:{SLURMD}", f"{SLURMD}:{SLURMCTLD}")
     await ops_test.model.integrate(f"{SLURMCTLD}:{SLURMDBD}", f"{SLURMDBD}:{SLURMCTLD}")
+    await ops_test.model.integrate(f"{SLURMCTLD}:{SLURMRESTD}", f"{SLURMRESTD}:{SLURMCTLD}")
     await ops_test.model.integrate(f"{SLURMDBD}-{ROUTER}:backend-database", f"{DATABASE}:database")
     await ops_test.model.integrate(f"{SLURMDBD}:database", f"{SLURMDBD}-{ROUTER}:database")
     # Reduce the update status frequency to accelerate the triggering of deferred events.
     async with ops_test.fast_forward():
         await ops_test.model.wait_for_idle(apps=[SLURMCTLD], status="active", timeout=1000)
-        assert ops_test.model.units.get(UNIT_NAME).workload_status == "active"
+        assert ops_test.model.applications["slurmctld"].units[0].workload_status == "active"
 
 
 @pytest.mark.abort_on_fail
@@ -102,7 +117,7 @@ async def test_build_and_deploy_against_edge(
 async def test_slurmctld_is_active(ops_test: OpsTest) -> None:
     """Test that slurmctld is active inside Juju unit."""
     logger.info("Checking that slurmctld is active inside Juju unit")
-    slurmctld_unit = ops_test.model.units.get(UNIT_NAME)
+    slurmctld_unit = ops_test.model.applications["slurmctld"].units[0]
     res = (await slurmctld_unit.ssh("systemctl is-active slurmctld")).strip("\n")
     assert res == "active"
 
@@ -117,7 +132,7 @@ async def test_slurmctld_is_active(ops_test: OpsTest) -> None:
 async def test_slurmctld_port_listen(ops_test: OpsTest) -> None:
     """Test that slurmctld is listening on port 6817."""
     logger.info("Checking that slurmctld is listening on port 6817")
-    slurmctld_unit = ops_test.model.units.get(UNIT_NAME)
+    slurmctld_unit = ops_test.model.applications["slurmctld"].units[0]
     res = await slurmctld_unit.ssh("sudo lsof -t -n -iTCP:6817 -sTCP:LISTEN")
     assert res != ""
 
@@ -132,6 +147,6 @@ async def test_slurmctld_port_listen(ops_test: OpsTest) -> None:
 async def test_munge_is_active(ops_test: OpsTest) -> None:
     """Test that munge is active inside Juju unit."""
     logger.info("Checking that munge is active inside Juju unit")
-    slurmctld_unit = ops_test.model.units.get(UNIT_NAME)
+    slurmctld_unit = ops_test.model.applications["slurmctld"].units[0]
     res = (await slurmctld_unit.ssh("systemctl is-active munge")).strip("\n")
     assert res == "active"