fix(aci): Retry process_delayed_workflows timeouts

src/sentry/sentry_apps/tasks/sentry_apps.py

@@ -330,7 +330,7 @@ def _process_resource_change(
             except model.DoesNotExist as e:
                 # Explicitly requeue the task, so we don't report this to Sentry until
                 # we hit the max number of retries.
-                return retry_task(e)
+                retry_task(e)
 
         org = None
 

src/sentry/taskworker/retry.py

@@ -2,6 +2,7 @@
 
 from enum import Enum
 from multiprocessing.context import TimeoutError
+from typing import NoReturn
 
 from celery import current_task
 from sentry_protos.taskbroker.v1.taskbroker_pb2 import (
@@ -41,7 +42,7 @@ def to_proto(self) -> OnAttemptsExceeded.ValueType:
         raise ValueError(f"Unknown LastAction: {self}")
 
 
-def retry_task(exc: Exception | None = None) -> None:
+def retry_task(exc: Exception | None = None) -> NoReturn:
     """
     Helper for triggering retry errors.
     If all retries have been consumed, this will raise a
@@ -55,6 +56,7 @@ def retry_task(exc: Exception | None = None) -> None:
     celery_retry = getattr(current_task, "retry", None)
     if celery_retry:
         current_task.retry(exc=exc)
+        assert False, "unreachable"
     else:
         current = taskworker_current_task()
         if current and not current.retries_remaining:

src/sentry/workflow_engine/processors/delayed_workflow.py

@@ -65,6 +65,7 @@
 )
 from sentry.workflow_engine.processors.workflow_fire_history import create_workflow_fire_histories
 from sentry.workflow_engine.tasks.actions import build_trigger_action_task_params, trigger_action
+from sentry.workflow_engine.tasks.utils import retry_timeouts
 from sentry.workflow_engine.types import DataConditionHandler, WorkflowEventData
 from sentry.workflow_engine.utils import log_context
 
@@ -760,6 +761,7 @@ def repr_keys[T, V](d: dict[T, V]) -> dict[str, V]:
     ),
 )
 @retry
+@retry_timeouts
 @log_context.root()
 def process_delayed_workflows(
     project_id: int, batch_key: str | None = None, *args: Any, **kwargs: Any

src/sentry/workflow_engine/tasks/utils.py

@@ -1,3 +1,6 @@
+from functools import wraps
+
+import sentry_sdk
 from google.api_core.exceptions import DeadlineExceeded, RetryError, ServiceUnavailable
 
 from sentry import nodestore
@@ -6,6 +9,8 @@
 from sentry.issues.issue_occurrence import IssueOccurrence
 from sentry.models.environment import Environment
 from sentry.models.group import Group
+from sentry.taskworker.retry import retry_task
+from sentry.taskworker.workerchild import ProcessingDeadlineExceeded
 from sentry.types.activity import ActivityType
 from sentry.utils.retries import ConditionalRetryPolicy, exponential_delay
 from sentry.workflow_engine.types import WorkflowEventData
@@ -82,3 +87,20 @@ def build_workflow_event_data_from_event(
         has_escalated=has_escalated,
         workflow_env=workflow_env,
     )
+
+
+def retry_timeouts(func):
+    """
+    Schedule a task retry if the function raises ProcessingDeadlineExceeded.
+    This exists because the standard retry decorator doesn't allow BaseExceptions.
+    """
+
+    @wraps(func)
+    def wrapper(*args, **kwargs):
+        try:
+            return func(*args, **kwargs)
+        except ProcessingDeadlineExceeded:
+            sentry_sdk.capture_exception(level="info")
+            retry_task()
+
+    return wrapper

tests/sentry/tasks/test_base.py

@@ -160,12 +160,17 @@ def test_validate_parameters_call():
     assert "region_task was called with a parameter that cannot be JSON encoded" in str(err)
 
 
+@override_settings(SILO_MODE=SiloMode.CONTROL)
 @patch("sentry.taskworker.retry.current_task")
 @patch("sentry_sdk.capture_exception")
 def test_retry_on(capture_exception, current_task):
+    class ExpectedException(Exception):
+        pass
+
+    current_task.retry.side_effect = ExpectedException("some exception")
 
-    # In reality current_task.retry will cause the given exception to be re-raised but we patch it here so no need to .raises :bufo-shrug:
-    retry_on_task("bruh")
+    with pytest.raises(ExpectedException):
+        retry_on_task("bruh")
 
     assert capture_exception.call_count == 1
     assert current_task.retry.call_count == 1