Skip to content

🐛Ensure Dask client reference is uniquely defined for reference counting #7937

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Jun 20, 2025
Merged

🐛Ensure Dask client reference is uniquely defined for reference counting #7937

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e4cc8a6
fix
sanderegg Jun 19, 2025
e38545b
simplify
sanderegg Jun 19, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 2 additions & 4 deletions ...ces/director-v2/src/simcore_service_director_v2/modules/comp_scheduler/_scheduler_base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -590,9 +590,7 @@ async def _process_executing_tasks(
"""process executing tasks from the 3rd party backend"""

@abstractmethod
async def _release_resources(
self, user_id: UserID, project_id: ProjectID, comp_run: CompRunsAtDB
) -> None:
async def _release_resources(self, comp_run: CompRunsAtDB) -> None:
"""release resources used by the scheduler for a given user and project"""

async def apply(
Expand Down Expand Up @@ -660,7 +658,7 @@ async def apply(

# 7. Are we done scheduling that pipeline?
if not dag.nodes() or pipeline_result in COMPLETED_STATES:
await self._release_resources(user_id, project_id, comp_run)
await self._release_resources(comp_run)
# there is nothing left, the run is completed, we're done here
_logger.info(
"pipeline %s scheduling completed with result %s",
Expand Down
26 changes: 19 additions & 7 deletions ...ces/director-v2/src/simcore_service_director_v2/modules/comp_scheduler/_scheduler_dask.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,7 +56,7 @@

_logger = logging.getLogger(__name__)

_DASK_CLIENT_RUN_REF: Final[str] = "{user_id}:{run_id}"
_DASK_CLIENT_RUN_REF: Final[str] = "{user_id}:{project_id}:{run_id}"


@asynccontextmanager
Expand All @@ -65,6 +65,7 @@ async def _cluster_dask_client(
scheduler: "DaskScheduler",
*,
use_on_demand_clusters: bool,
project_id: ProjectID,
run_id: PositiveInt,
run_metadata: RunMetadataDict,
) -> AsyncIterator[DaskClient]:
Expand All @@ -76,7 +77,10 @@ async def _cluster_dask_client(
wallet_id=run_metadata.get("wallet_id"),
)
async with scheduler.dask_clients_pool.acquire(
cluster, ref=_DASK_CLIENT_RUN_REF.format(user_id=user_id, run_id=run_id)
cluster,
ref=_DASK_CLIENT_RUN_REF.format(
user_id=user_id, project_id=project_id, run_id=run_id
),
) as client:
yield client

Expand Down Expand Up @@ -106,6 +110,7 @@ async def _start_tasks(
user_id,
self,
use_on_demand_clusters=comp_run.use_on_demand_clusters,
project_id=comp_run.project_uuid,
run_id=comp_run.run_id,
run_metadata=comp_run.metadata,
) as client:
Expand Down Expand Up @@ -157,6 +162,7 @@ async def _get_tasks_status(
user_id,
self,
use_on_demand_clusters=comp_run.use_on_demand_clusters,
project_id=comp_run.project_uuid,
run_id=comp_run.run_id,
run_metadata=comp_run.metadata,
) as client:
Expand All @@ -178,6 +184,7 @@ async def _process_executing_tasks(
user_id,
self,
use_on_demand_clusters=comp_run.use_on_demand_clusters,
project_id=comp_run.project_uuid,
run_id=comp_run.run_id,
run_metadata=comp_run.metadata,
) as client:
Expand Down Expand Up @@ -225,20 +232,22 @@ async def _process_executing_tasks(
)
)

async def _release_resources(
self, user_id: UserID, project_id: ProjectID, comp_run: CompRunsAtDB
) -> None:
async def _release_resources(self, comp_run: CompRunsAtDB) -> None:
"""release resources used by the scheduler for a given user and project"""
with (
log_catch(_logger, reraise=False),
log_context(
_logger,
logging.INFO,
msg=f"releasing resources for {user_id=}, {project_id=}, {comp_run.run_id=}",
msg=f"releasing resources for {comp_run.user_id=}, {comp_run.project_uuid=}, {comp_run.run_id=}",
),
):
await self.dask_clients_pool.release_client_ref(
ref=_DASK_CLIENT_RUN_REF.format(user_id=user_id, run_id=comp_run.run_id)
ref=_DASK_CLIENT_RUN_REF.format(
user_id=comp_run.user_id,
project_id=comp_run.project_uuid,
run_id=comp_run.run_id,
)
)

async def _stop_tasks(
Expand All @@ -250,6 +259,7 @@ async def _stop_tasks(
user_id,
self,
use_on_demand_clusters=comp_run.use_on_demand_clusters,
project_id=comp_run.project_uuid,
run_id=comp_run.run_id,
run_metadata=comp_run.metadata,
) as client:
Expand Down Expand Up @@ -284,6 +294,7 @@ async def _process_completed_tasks(
user_id,
self,
use_on_demand_clusters=comp_run.use_on_demand_clusters,
project_id=comp_run.project_uuid,
run_id=comp_run.run_id,
run_metadata=comp_run.metadata,
) as client:
Expand All @@ -304,6 +315,7 @@ async def _process_completed_tasks(
user_id,
self,
use_on_demand_clusters=comp_run.use_on_demand_clusters,
project_id=comp_run.project_uuid,
run_id=comp_run.run_id,
run_metadata=comp_run.metadata,
) as client:
Expand Down
Loading