✨Adding new cli option to debug internal status of director-v2

services/director-v2/src/simcore_service_director_v2/api/routes/dynamic_scheduler.py

@@ -22,6 +22,7 @@
 from tenacity.stop import stop_after_delay
 from tenacity.wait import wait_random_exponential
 
+from ...models.dynamic_services_scheduler import SchedulerData
 from ...modules.dynamic_sidecar.scheduler import DynamicSidecarsScheduler
 from ...utils.routes import NoContentResponse
 from ..dependencies.dynamic_sidecar import get_dynamic_sidecar_scheduler
@@ -121,6 +122,24 @@ async def _progress_callback(
         raise HTTPException(status.HTTP_409_CONFLICT, detail=f"{e}") from e
 
 
+@router.get(
+    "/services/{node_uuid}/state",
+    summary="Returns the internals of the scheduler for the given service",
+    status_code=status.HTTP_200_OK,
+    response_model=SchedulerData,
+)
+async def get_service_state(
+    node_uuid: NodeID,
+    dynamic_sidecars_scheduler: Annotated[
+        DynamicSidecarsScheduler, Depends(get_dynamic_sidecar_scheduler)
+    ],
+):
+    # pylint: disable=protected-access
+    return dynamic_sidecars_scheduler._scheduler.get_scheduler_data(
+        node_uuid
+    )  # noqa: SLF001
+
+
 @router.post(
     "/services/{node_uuid}/state:save",
     summary="Starts the saving of the state for the service",

services/director-v2/src/simcore_service_director_v2/cli/__init__.py

@@ -10,7 +10,7 @@
 from ..core.settings import AppSettings
 from ..meta import PROJECT_NAME
 from ._close_and_save_service import async_close_and_save_service
-from ._core import async_project_save_state, async_project_state
+from ._core import async_project_save_state, async_project_state, async_service_state
 
 DEFAULT_NODE_SAVE_ATTEMPTS: Final[int] = 3
 DEFAULT_STATE_UPDATE_INTERVAL_S: Final[int] = 5
@@ -50,6 +50,14 @@ def project_state(
     asyncio.run(async_project_state(project_id, blocking, update_interval))
 
 
+@main.command()
+def service_state(node_id: NodeID):
+    """
+    Prints the state of a services as tracked by director-v2
+    """
+    asyncio.run(async_service_state(node_id))
+
+
 @main.command()
 def close_and_save_service(
     node_id: NodeID,

services/director-v2/src/simcore_service_director_v2/cli/_client.py

@@ -60,3 +60,8 @@ async def delete_service_docker_resources(self, node_uuid: str) -> Response:
         return await self.client.delete(
             self._get_url(f"/services/{node_uuid}/docker-resources")
         )
+
+    @retry_on_errors
+    @expect_status(status.HTTP_200_OK)
+    async def get_service_state(self, node_uuid: str) -> Response:
+        return await self.client.get(self._get_url(f"/services/{node_uuid}/state"))

services/director-v2/src/simcore_service_director_v2/cli/_core.py

@@ -29,6 +29,7 @@
 from ..modules.dynamic_sidecar import api_client
 from ..modules.projects_networks import requires_dynamic_sidecar
 from ..utils.db import get_repository
+from ._client import ThinDV2LocalhostClient
 
 
 @asynccontextmanager
@@ -273,3 +274,9 @@ async def async_project_state(
         await _display(
             app, project_id, update_interval=update_interval, blocking=blocking
         )
+
+
+async def async_service_state(node_id: NodeID) -> None:
+    thin_dv2_localhost_client = ThinDV2LocalhostClient()
+    result = await thin_dv2_localhost_client.get_service_state(node_id)
+    typer.echo(f"Service state: {result.text}")

services/director-v2/tests/unit/with_dbs/test_cli.py

@@ -134,6 +134,9 @@ async def mock_close_service_routes(
         respx_mock.post(
             re.compile(f"{regex_base}/state:save"), name="save_service_state"
         ).respond(status_code=status.HTTP_202_ACCEPTED, json=task_id)
+        respx_mock.get(
+            re.compile(f"{regex_base}/state"), name="service_internal_state"
+        ).respond(status_code=status.HTTP_200_OK)
         respx_mock.post(
             re.compile(f"{regex_base}/outputs:push"), name="push_service_outputs"
         ).respond(status_code=status.HTTP_202_ACCEPTED, json=task_id)
@@ -235,3 +238,11 @@ def test_close_and_save_service(
     result = cli_runner.invoke(main, ["close-and-save-service", f"{node_id}"])
     assert result.exit_code == os.EX_OK, _format_cli_error(result)
     print(result.stdout)
+
+
+def test_service_state(
+    mock_close_service_routes: None, cli_runner: CliRunner, node_id: NodeID
+):
+    result = cli_runner.invoke(main, ["service-state", f"{node_id}"])
+    assert result.exit_code == os.EX_OK, _format_cli_error(result)
+    print(result.stdout)