Skip to content

perf(workflow): batch step persistence after execution #722

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

Open
xiami762 wants to merge 5 commits into
base: dev
Choose a base branch
from
Open

perf(workflow): batch step persistence after execution #722

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
5600c59
perf(workflow): remove synchronous step storage waits
Aug 21, 2026
c01cb25
fix(workflow): finalize atomic step persistence
Aug 21, 2026
76f3099
fix(workflow): preserve steps without callback waits
Aug 21, 2026
0908fd3
fix(workflow): persist queued trigger executions
Aug 21, 2026
afc6649
refactor(workflow): simplify step persistence plumbing
Aug 21, 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
29 changes: 16 additions & 13 deletions flocks/ingest/kafka/manager.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -91,6 +91,7 @@ def _worker_count_for_trigger(trigger: TriggerDefinition) -> int:
def _queue_size_for_trigger(trigger: TriggerDefinition) -> int:
return min(_MAX_QUEUE_SIZE, max(1, int(trigger.concurrency.queueSize)))


_KAFKA_STORAGE_LIST_KEYS = DEFAULT_LARGE_LIST_KEYS | frozenset(
{
"duplicate_alerts",
Expand Down Expand Up @@ -446,10 +447,13 @@ async def restart_workflow(
err = "workflow_not_found"
if startup:
self._status[workflow_id] = {"state": "stopped", "error": err}
log.info("kafka.workflow_not_found_on_start", {
"workflow_id": workflow_id,
"action": "stale_config_skipped",
})
log.info(
"kafka.workflow_not_found_on_start",
{
"workflow_id": workflow_id,
"action": "stale_config_skipped",
},
)
return {"state": "stopped", "error": err}
self._status[workflow_id] = {"state": "failed", "error": err}
log.warning("kafka.workflow_not_found", {"workflow_id": workflow_id})
Expand Down Expand Up @@ -689,9 +693,7 @@ async def _worker_loop(
generation_cancel_event: Optional[threading.Event] = None,
) -> None:
run_cancel_event = (
generation_cancel_event
or self._generation_cancel_events.get(workflow_id)
or threading.Event()
generation_cancel_event or self._generation_cancel_events.get(workflow_id) or threading.Event()
)
while not abort.is_set():
try:
Expand Down Expand Up @@ -767,15 +769,10 @@ async def _executor(mapped_inputs: Dict[str, Any]) -> Dict[str, Any]:
input_params=summarized_inputs,
)
exec_id = exec_data["id"]
loop = asyncio.get_running_loop()
start_time = time.time()
trigger_meta = mapped_inputs.get("_flocks", {}).get("trigger", {})
trigger_input_keys = list((trigger.mapping or {}).keys()) or [input_key]
step_recorder = ExecutionStepRecorder(
exec_id=exec_id,
loop=loop,
logger=log,
log_event="kafka.execution_step.write_failed",
step_compactor=lambda step: _compact_step_for_kafka_storage(
step,
input_key=input_key,
Expand Down Expand Up @@ -845,9 +842,15 @@ async def _executor(mapped_inputs: Dict[str, Any]) -> Dict[str, Any]:
}
)
finally:
steps = step_recorder.take_steps()
await cleanup_workflow_tool_context(tool_context)
try:
await record_execution_result(workflow_id, exec_id, exec_data)
await record_execution_result(
workflow_id,
exec_id,
exec_data,
steps=steps,
)
except Exception as exc:
log.warning("kafka.exec_record_failed", {"exec_id": exec_id, "error": str(exc)})
return exec_data
Expand Down
31 changes: 16 additions & 15 deletions flocks/ingest/syslog/manager.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -335,10 +335,13 @@ async def restart_workflow(
err = "workflow_not_found"
if startup:
self._listener_status[workflow_id] = {"state": "stopped", "error": err}
log.info("syslog.workflow_not_found_on_start", {
"workflow_id": workflow_id,
"action": "stale_config_skipped",
})
log.info(
"syslog.workflow_not_found_on_start",
{
"workflow_id": workflow_id,
"action": "stale_config_skipped",
},
)
return {"state": "stopped", "error": err}
self._listener_status[workflow_id] = {"state": "failed", "error": err}
log.warning("syslog.workflow_not_found", {"workflow_id": workflow_id})
Expand Down Expand Up @@ -553,9 +556,7 @@ async def _worker_loop(
of in-flight workflow runs is exactly ``_MAX_CONCURRENT_EXECUTIONS``.
"""
run_cancel_event = (
generation_cancel_event
or self._generation_cancel_events.get(workflow_id)
or threading.Event()
generation_cancel_event or self._generation_cancel_events.get(workflow_id) or threading.Event()
)
while not abort.is_set():
try:
Expand Down Expand Up @@ -620,13 +621,7 @@ async def _executor(mapped_inputs: Dict[str, Any]) -> Dict[str, Any]:
input_params=summarized_inputs,
)
exec_id = exec_data["id"]
loop = asyncio.get_running_loop()
step_recorder = ExecutionStepRecorder(
exec_id=exec_id,
loop=loop,
logger=log,
log_event="syslog.execution_step.write_failed",
)
step_recorder = ExecutionStepRecorder()
start_time = time.time()
trigger_meta = mapped_inputs.get("_flocks", {}).get("trigger", {})
tool_context = None
Expand Down Expand Up @@ -692,9 +687,15 @@ async def _executor(mapped_inputs: Dict[str, Any]) -> Dict[str, Any]:
}
)
finally:
steps = step_recorder.take_steps()
await cleanup_workflow_tool_context(tool_context)
try:
await record_execution_result(workflow_id, exec_id, exec_data)
await record_execution_result(
workflow_id,
exec_id,
exec_data,
steps=steps,
)
except Exception as exc:
log.warning("syslog.exec_record_failed", {"exec_id": exec_id, "error": str(exc)})
return exec_data
Expand Down
Loading