test: fix unit tests & pull CI changes

.github/workflows/ci.yaml

@@ -65,12 +65,16 @@ 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
@@ -79,10 +83,21 @@ jobs:
     steps:
       - name: Checkout
         uses: actions/checkout@v4
+        with:
+          path: main
+      - name: Fetch slurmctld
+        uses: actions/checkout@v4
+        if: ${{ matrix.local }}
+        with:
+          repository: charmed-hpc/slurmctld-operator
+          path: slurmctld-operator
       - name: Setup operator environment
         uses: charmed-kubernetes/actions-operator@main
         with:
           provider: lxd
           juju-channel: 3.1/stable
       - name: Run tests
-        run: tox run -e integration -- --charm-base=${{ 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 }}"

lib/charms/hpc_libs/v0/slurm_ops.py

@@ -62,11 +62,13 @@ def _on_install(self, _) -> None:
     "ConfigurationManager",
     "ServiceType",
     "SlurmManagerBase",
+    "SlurmOpsError",
 ]
 
 import json
 import logging
 import re
+import socket
 import subprocess
 from collections.abc import Mapping
 from enum import Enum
@@ -82,7 +84,7 @@ def _on_install(self, _) -> None:
 
 # Increment this PATCH version before using `charmcraft publish-lib` or reset
 # to 0 if you are raising the major API version
-LIBPATCH = 3
+LIBPATCH = 5
 
 # Charm library dependencies to fetch during `charmcraft pack`.
 PYDEPS = ["pyyaml>=6.0.1"]
@@ -131,7 +133,8 @@ def install() -> None:
 def version() -> str:
     """Get the current version of Slurm installed on the system."""
     info = yaml.safe_load(_snap("info", "slurm"))
-    ver: str = info["installed"]
+    if (ver := info.get("installed")) is None:
+        raise SlurmOpsError("unable to retrive snap info. Ensure slurm is correctly installed")
     return ver.split(maxsplit=1)[0]
 
 
@@ -177,6 +180,7 @@ class ServiceType(Enum):
     """Type of Slurm service to manage."""
 
     MUNGED = "munged"
+    PROMETHEUS_EXPORTER = "slurm-prometheus-exporter"
     SLURMD = "slurmd"
     SLURMCTLD = "slurmctld"
     SLURMDBD = "slurmdbd"
@@ -208,6 +212,17 @@ def restart(self) -> None:
         """Restart service."""
         _snap("restart", f"slurm.{self._service.value}")
 
+    def active(self) -> bool:
+        """Return True if the service is active."""
+        info = yaml.safe_load(_snap("info", "slurm"))
+        if (services := info.get("services")) is None:
+            raise SlurmOpsError("unable to retrive snap info. Ensure slurm is correctly installed")
+
+        # Assume `services` contains the service, since `ServiceManager` is not exposed as a
+        # public interface for now.
+        # We don't do `"active" in state` because the word "active" is also part of "inactive" :)
+        return "inactive" not in services[f"slurm.{self._service.value}"]
+
 
 class ConfigurationManager:
     """Control configuration of a Slurm component."""
@@ -271,10 +286,23 @@ def generate_key(self) -> None:
         _mungectl("key", "generate")
 
 
+class PrometheusExporterManager(ServiceManager):
+    """Manage `slurm-prometheus-exporter` service operations."""
+
+    def __init__(self) -> None:
+        self._service = ServiceType.PROMETHEUS_EXPORTER
+
+
 class SlurmManagerBase(ServiceManager):
     """Base manager for Slurm services."""
 
     def __init__(self, service: ServiceType) -> None:
         self._service = service
         self.config = ConfigurationManager(service.config_name)
         self.munge = MungeManager()
+        self.exporter = PrometheusExporterManager()
+
+    @property
+    def hostname(self) -> str:
+        """The hostname where this manager is running."""
+        return socket.gethostname().split(".")[0]

tests/integration/conftest.py

@@ -73,23 +73,3 @@ async def slurmctld_charm(request, ops_test: OpsTest) -> Union[str, Path]:
             )
 
     return "slurmctld"
-
-
-@pytest.fixture(scope="module")
-async def slurmdbd_charm(request, ops_test: OpsTest) -> Union[str, Path]:
-    """Pack slurmdbd charm to use for integration tests when --use-local is specified.
-
-    Returns:
-        `str` "slurmdbd" if --use-local not specified or if SLURMDBD_DIR does not exist.
-    """
-    if request.config.option.use_local:
-        logger.info("Using local slurmdbd operator rather than pulling from Charmhub")
-        if SLURMDBD_DIR.exists():
-            return await ops_test.build_charm(SLURMDBD_DIR)
-        else:
-            logger.warning(
-                f"{SLURMDBD_DIR} not found. "
-                f"Defaulting to latest/edge slurmdbd operator from Charmhub"
-            )
-
-    return "slurmdbd"

tests/integration/test_charm.py

@@ -35,14 +35,12 @@
 @pytest.mark.skip_if_deployed
 @pytest.mark.order(1)
 async def test_build_and_deploy(
-    ops_test: OpsTest, charm_base: str, slurmd_charm, slurmctld_charm, slurmdbd_charm
+    ops_test: OpsTest, charm_base: str, slurmd_charm, slurmctld_charm
 ) -> None:
     """Test that the slurmd charm can stabilize against slurmctld, slurmdbd and MySQL."""
-    logger.info(f"Deploying {SLURMD} against {SLURMCTLD}, {SLURMDBD}, and {DATABASE}")
+    logger.info(f"Deploying {SLURMD} against {SLURMCTLD}")
     # Pack charms and download NHC resource for slurmd operator.
-    slurmd, slurmctld, slurmdbd = await asyncio.gather(
-        slurmd_charm, slurmctld_charm, slurmdbd_charm
-    )
+    slurmd, slurmctld = await asyncio.gather(slurmd_charm, slurmctld_charm)
     # Deploy the test Charmed SLURM cloud.
     await asyncio.gather(
         ops_test.model.deploy(
@@ -58,33 +56,9 @@ async def test_build_and_deploy(
             num_units=1,
             base=charm_base,
         ),
-        ops_test.model.deploy(
-            str(slurmdbd),
-            application_name=SLURMDBD,
-            channel="edge" if isinstance(slurmdbd, str) else None,
-            num_units=1,
-            base=charm_base,
-        ),
-        ops_test.model.deploy(
-            ROUTER,
-            application_name=f"{SLURMDBD}-{ROUTER}",
-            channel="dpe/edge",
-            num_units=0,
-            base=charm_base,
-        ),
-        ops_test.model.deploy(
-            DATABASE,
-            application_name=DATABASE,
-            channel="8.0/edge",
-            num_units=1,
-            base="[email protected]",
-        ),
     )
     # Set relations for charmed applications.
     await ops_test.model.integrate(f"{SLURMD}:{SLURMCTLD}", f"{SLURMCTLD}:{SLURMD}")
-    await ops_test.model.integrate(f"{SLURMDBD}:{SLURMCTLD}", f"{SLURMCTLD}:{SLURMDBD}")
-    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=[SLURMD], status="active", timeout=1000)

tests/unit/test_charm.py

@@ -56,6 +56,7 @@ def test_install_fail(self, _, defer) -> None:
         defer.assert_called()
 
     @patch("slurmd_ops.SlurmdManager.install")
+    @patch("slurmd_ops.SlurmdManager.version", return_value="23.11.7")
     @patch("pathlib.Path.read_text", return_value="v1.0.0")
     @patch("ops.model.Unit.set_workload_version")
     @patch("ops.model.Resources.fetch")

tests/unit/utils/test_slurmd.py

@@ -26,18 +26,6 @@
 class TestSlurmd(unittest.TestCase):
     """Unit tests for methods in slurmd utility module."""
 
-    @patch("charms.operator_libs_linux.v1.systemd._systemctl")
-    def test_start(self, _) -> None:
-        slurmd.start()
-
-    @patch("charms.operator_libs_linux.v1.systemd._systemctl")
-    def test_stop(self, _) -> None:
-        slurmd.stop()
-
-    @patch("charms.operator_libs_linux.v1.systemd._systemctl")
-    def test_restart(self, _) -> None:
-        slurmd.restart()
-
     @patch("pathlib.Path.write_text")
     def test_overwrite_default(self, _) -> None:
         slurmd.override_default("127.0.0.1")

tox.ini

@@ -78,5 +78,5 @@ commands =
         -s \
         --tb native \
         --log-cli-level=INFO \
-        {[vars]tst_path}integration \
+        {[vars]tst_path}/integration \
         {posargs}