Skip to content

[benchmarker][scd] Ensure OVN list is properly cleaned up #1833

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
barroco merged 2 commits into from
Oct 21, 2026
Merged

[benchmarker][scd] Ensure OVN list is properly cleaned up #1833

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
e0476ba
[benchmarker][scd] Ensure OVN list is properly cleanup
barroco Oct 20, 2026
3b53daf
lint
barroco Oct 20, 2026
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
7 changes: 7 additions & 0 deletions monitoring/benchmarker/engine/coordination.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -68,3 +68,10 @@ def subscribe(
group.subscribers.add(subscriber)
for msg in group.messages:
subscriber.receive_coordination_message(msg)

def unsubscribe(
self, subscriber: CoordinationSubscriber, group_id: CoordinationGroupID
) -> None:
group = self._groups[group_id]
with group.lock:
group.subscribers.discard(subscriber)
61 changes: 39 additions & 22 deletions monitoring/benchmarker/engine/users/flight_planner/scd.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,7 +65,7 @@ class SCDHandler(CoordinationSubscriber):

op_intent_refs: dict[FlightID, api.OperationalIntentReference]

key: list[api.EntityOVN]
key: set[api.EntityOVN]
"""OVNs currently known by this user."""

key_lock: Lock
Expand All @@ -79,7 +79,7 @@ def __init__(
):
self.user = user

self.key = []
self.key = set()
self.key_lock = Lock()

self.dss_instances = []
Expand Down Expand Up @@ -178,14 +178,14 @@ def receive_coordination_message(self, msg: CoordinationMessage) -> None:
f"Expected an EntityOVN content in a CoordinationMessage with subject {msg.subject} but instead found a {type(msg.content).__name__}"
)
with self.key_lock:
self.key.append(msg.content)
self.key.add(msg.content)
elif msg.subject == COORDINATION_SUBJECT_REMOVE_OVN:
if not isinstance(msg.content, api.EntityOVN):
raise RuntimeError(
f"Expected an EntityOVN content in a CoordinationMessage with subject {msg.subject} but instead found a {type(msg.content).__name__}"
)
with self.key_lock:
self.key.remove(msg.content)
self.key.discard(msg.content)

def select_dss_instance(self) -> DSSInstance:
if self.dss_selection_strategy == ASTMDSSSelectionStrategy.First:
Expand Down Expand Up @@ -425,7 +425,7 @@ async def upsert_op_intent_ref(
CoordinationMessage(
group_id=None,
subject=COORDINATION_SUBJECT_REMOVE_OVN,
content=op_intent_ref.ovn,
content=old_ovn,
)
)
if old_ovn is None or op_intent_ref.ovn != old_ovn:
Expand Down Expand Up @@ -520,6 +520,29 @@ def get_delete_actions(
run_on_shutdown=True,
)

def _remove_ovn(self, ovn: api.EntityOVN | None) -> None:
"""Remove the OVN of an op intent that no longer exists from the key of all coordinating users."""
if ovn is None:
return

if (
"ovn_coordination_group" in self.op_intent_ref_creation_strategy
and self.op_intent_ref_creation_strategy.ovn_coordination_group
):
self.user.coordinator.publish(
self.op_intent_ref_creation_strategy.ovn_coordination_group,
COORDINATION_SUBJECT_REMOVE_OVN,
ovn,
)
else:
self.receive_coordination_message(
CoordinationMessage(
group_id=None,
subject=COORDINATION_SUBJECT_REMOVE_OVN,
content=ovn,
)
)

async def delete_op_intent_ref(
self, flight: Flight, op_intent_id: api.EntityID
) -> list[FlightAction]:
Expand Down Expand Up @@ -548,23 +571,7 @@ async def delete_op_intent_ref(
)

if success:
if (
"ovn_coordination_group" in self.op_intent_ref_creation_strategy
and self.op_intent_ref_creation_strategy.ovn_coordination_group
):
self.user.coordinator.publish(
self.op_intent_ref_creation_strategy.ovn_coordination_group,
COORDINATION_SUBJECT_REMOVE_OVN,
op_intent_ref.ovn,
)
else:
self.receive_coordination_message(
CoordinationMessage(
group_id=None,
subject=COORDINATION_SUBJECT_REMOVE_OVN,
content=op_intent_ref.ovn,
)
)
self._remove_ovn(op_intent_ref.ovn)
else:
with self.key_lock:
self.op_intent_refs[flight.id] = op_intent_ref
Expand All @@ -589,12 +596,14 @@ async def cleanup(self) -> None:
)
self.user.record_query(query, True)
n_deleted += 1
self._remove_ovn(op_intent_ref.ovn)
except QueryError as e:
success = e.queries[0].status_code == 404
for query in e.queries:
self.user.record_query(query, success)
if success:
n_already_gone += 1
self._remove_ovn(op_intent_ref.ovn)
else:
logger.warning(
f"{self.user.user_id}'s SCDHandler was unable to clean up op intent {op_intent_ref.id} for flight {flight_id} from {dss_instance.participant_id}'s DSS; HTTP code {e.queries[0].status_code}"
Expand All @@ -613,6 +622,14 @@ async def cleanup(self) -> None:
msg += f"; {len(undeleted_ids)} could not be deleted"
logger.debug(msg)

if (
"ovn_coordination_group" in self.op_intent_ref_creation_strategy
and self.op_intent_ref_creation_strategy.ovn_coordination_group
):
self.user.coordinator.unsubscribe(
self, self.op_intent_ref_creation_strategy.ovn_coordination_group
)

@staticmethod
def enumerate_coordination_groups(
behavior: BehaviorSpecification,
Expand Down
Loading