sentry-arroyo 2.37.1__py3-none-any.whl → 2.38.0__py3-none-any.whl

arroyo/processing/processor.py

@@ -30,6 +30,7 @@ from arroyo.processing.strategies.abstract import (
 from arroyo.types import BrokerValue, Message, Partition, Topic, TStrategyPayload
 from arroyo.utils.logging import handle_internal_error
 from arroyo.utils.metrics import get_metrics
+from arroyo.utils.stuck_detector import get_all_thread_stacks
 
 logger = logging.getLogger(__name__)
 
@@ -40,25 +41,6 @@ DEFAULT_JOIN_TIMEOUT = 25.0  # In seconds
 F = TypeVar("F", bound=Callable[[Any], Any])
 
 
-def get_all_thread_stacks() -> str:
-    """Get stack traces from all threads without using signals."""
-    import sys
-    import threading
-    import traceback
-
-    stacks = []
-    frames = sys._current_frames()
-    threads_by_id = {t.ident: t for t in threading.enumerate()}
-
-    for thread_id, frame in frames.items():
-        thread = threads_by_id.get(thread_id)
-        thread_name = thread.name if thread else f"Unknown-{thread_id}"
-        stack = "".join(traceback.format_stack(frame))
-        stacks.append(f"Thread {thread_name} ({thread_id}):\n{stack}")
-
-    return "\n\n".join(stacks)
-
-
 def _rdkafka_callback(metrics: MetricsBuffer) -> Callable[[F], F]:
     def decorator(f: F) -> F:
         @functools.wraps(f)
@@ -166,9 +148,9 @@ class StreamProcessor(Generic[TStrategyPayload]):
         self.__processor_factory = processor_factory
         self.__metrics_buffer = MetricsBuffer()
 
-        self.__processing_strategy: Optional[
-            ProcessingStrategy[TStrategyPayload]
-        ] = None
+        self.__processing_strategy: Optional[ProcessingStrategy[TStrategyPayload]] = (
+            None
+        )
 
         self.__message: Optional[BrokerValue[TStrategyPayload]] = None
 

arroyo/utils/stuck_detector.py

@@ -0,0 +1,58 @@
+from __future__ import annotations
+
+import logging
+import sys
+import threading
+import time
+import traceback
+from contextlib import contextmanager
+from typing import Iterator
+
+logger = logging.getLogger(__name__)
+
+
+def get_all_thread_stacks() -> str:
+    """Get stack traces from all threads without using signals."""
+    stacks = []
+    frames = sys._current_frames()
+    threads_by_id = {t.ident: t for t in threading.enumerate()}
+
+    for thread_id, frame in frames.items():
+        thread = threads_by_id.get(thread_id)
+        thread_name = thread.name if thread else f"Unknown-{thread_id}"
+        stack = "".join(traceback.format_stack(frame))
+        stacks.append(f"Thread {thread_name} ({thread_id}):\n{stack}")
+
+    return "\n\n".join(stacks)
+
+
+@contextmanager
+def stuck_detector(
+    name: str = "stuck-detector",
+    timeout_seconds: float = 30.0,
+) -> Iterator[None]:
+    """
+    Context manager that spawns a daemon thread to detect stuck operations.
+    If the wrapped code block doesn't complete within timeout_seconds,
+    logs all thread stack traces for debugging.
+    """
+    done = threading.Event()
+
+    def detector() -> None:
+        start = time.time()
+        while not done.wait(timeout=1):
+            if time.time() - start > timeout_seconds:
+                logger.warning(
+                    "%s: Operation stuck for %s seconds, stacks:\n%s",
+                    name,
+                    timeout_seconds,
+                    get_all_thread_stacks(),
+                )
+                return
+
+    thread = threading.Thread(target=detector, daemon=True, name=name)
+    thread.start()
+    try:
+        yield
+    finally:
+        done.set()

{sentry_arroyo-2.37.1.dist-info → sentry_arroyo-2.38.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sentry-arroyo
-Version: 2.37.1
+Version: 2.38.0
 Summary: Arroyo is a Python library for working with streaming data.
 Home-page: https://github.com/getsentry/arroyo
 Author: Sentry

{sentry_arroyo-2.37.1.dist-info → sentry_arroyo-2.38.0.dist-info}/RECORD

@@ -16,7 +16,7 @@ arroyo/backends/local/storages/__init__.py,sha256=AGYujdAAcn3osoj9jq84IzTywYbkID
 arroyo/backends/local/storages/abstract.py,sha256=1qVQp6roxHkK6XT2aklZyZk1qq7RzcPN6Db_CA5--kg,2901
 arroyo/backends/local/storages/memory.py,sha256=AoKDsVZzBXkOJyWArKWp3vfGfU9xLlKFXE9gsJiMIzQ,2613
 arroyo/processing/__init__.py,sha256=vZVg0wJvJfoVzlzGvnL59bT6YNIRJNQ5t7oU045Qbk4,87
-arroyo/processing/processor.py,sha256=tZH05N4taJz15-BN1UK0lhrNdkeQl4ZJICGija6fIwQ,22598
+arroyo/processing/processor.py,sha256=-ZN1pZhEMLa9fkRA_FSnTgQ7FxXc1xuEXZtzHBvVIpA,22063
 arroyo/processing/strategies/__init__.py,sha256=EU_JMb54eOxMxaC5mIFpI-sAF-X2ZScbE8czBZ7bQkY,1106
 arroyo/processing/strategies/abstract.py,sha256=nu7juEz_aQmQIH35Z8u--FBuLjkK8_LQ1hIG2xpw9AA,4808
 arroyo/processing/strategies/batching.py,sha256=s89xC6lQpBseEaApu1iNTipXGKeO95OMwinj2VBKn9s,4778
@@ -42,11 +42,12 @@ arroyo/utils/metric_defs.py,sha256=Dmw5gBcH80sz_Z5hNH2tmUpGh8IN4QDPA3OEwR_KENQ,9
 arroyo/utils/metrics.py,sha256=kcyUR5cacoPMoU80RHSUhTMNzEcMBDpTXzcyW7yWZBk,3308
 arroyo/utils/profiler.py,sha256=aiYy2RRPX_IiDIO7AnFM3hARaHCctS3rqUS5nrHXbSg,2452
 arroyo/utils/retries.py,sha256=4MRhHUR7da9x1ytlo7YETo8S9HEebXmPF2-mKP4xYz0,3445
+arroyo/utils/stuck_detector.py,sha256=bHJbOuVKa_AGI0i0l3EbdTI0mwIub2svtY9epeS1a4g,1670
 examples/transform_and_produce/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 examples/transform_and_produce/batched.py,sha256=st2R6qTneAtV0JFbKP30Ti3sJDYj8Jkbmta9JckKdZU,2636
 examples/transform_and_produce/script.py,sha256=8kSMIjQNqGYEVyE0PvrfJh-a_UYCrJSstTp_De7kyyg,2306
 examples/transform_and_produce/simple.py,sha256=H7xqxItjl4tx34wVW5dy6mB9G39QucAtxkJSBzVmjgA,1637
-sentry_arroyo-2.37.1.dist-info/licenses/LICENSE,sha256=0Ng3MFdEcnz0sVD1XvGBBzbavvNp_7OAM5yVObB46jU,10829
+sentry_arroyo-2.38.0.dist-info/licenses/LICENSE,sha256=0Ng3MFdEcnz0sVD1XvGBBzbavvNp_7OAM5yVObB46jU,10829
 tests/backends/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 tests/backends/mixins.py,sha256=sfNyE0VTeiD3GHOnBYl-9urvPuURI2G1BWke0cz7Dvc,20445
 tests/backends/test_commit.py,sha256=iTHfK1qsBxim0XwxgMvNNSMqDUMEHoYkYBDcgxGBFbs,831
@@ -75,7 +76,8 @@ tests/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 tests/utils/test_concurrent.py,sha256=Gwdzym2UZ1HO3rhOSGmzxImWcLFygY8P7MXHT3Q0xTE,455
 tests/utils/test_metrics.py,sha256=bI0EtGgPokMQyEqX58i0-8zvLfxRP2nWaWr2wLMaJ_o,917
 tests/utils/test_retries.py,sha256=AxJLkXWeL9AjHv_p1n0pe8CXXJp24ZQIuYBHfNcmiz4,3075
-sentry_arroyo-2.37.1.dist-info/METADATA,sha256=jgESmXFDta8d4EwJ2su-xyxA0Wg84foCb20fqHpRReo,2208
-sentry_arroyo-2.37.1.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
-sentry_arroyo-2.37.1.dist-info/top_level.txt,sha256=DVdMZKysL_iIxm5aY0sYgZtP5ZXMg9YBaBmGQHVmDXA,22
-sentry_arroyo-2.37.1.dist-info/RECORD,,
+tests/utils/test_stuck_detector.py,sha256=IcNqL51L-UwDXA2-w70NB3nYrxLFwWqCdWhI5O5Q5Rs,1574
+sentry_arroyo-2.38.0.dist-info/METADATA,sha256=2D8NJrqb6DLK70kjcx93sAiRdoxSmxBjFBO9IXC35Rg,2208
+sentry_arroyo-2.38.0.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+sentry_arroyo-2.38.0.dist-info/top_level.txt,sha256=DVdMZKysL_iIxm5aY0sYgZtP5ZXMg9YBaBmGQHVmDXA,22
+sentry_arroyo-2.38.0.dist-info/RECORD,,

tests/utils/test_stuck_detector.py

@@ -0,0 +1,46 @@
+from __future__ import annotations
+
+import threading
+from unittest.mock import patch
+
+from arroyo.utils.stuck_detector import get_all_thread_stacks, stuck_detector
+
+
+def test_get_all_thread_stacks() -> None:
+    """Test that get_all_thread_stacks returns stack traces for all threads."""
+    stacks = get_all_thread_stacks()
+    assert "MainThread" in stacks
+    assert "test_get_all_thread_stacks" in stacks  # Current function should be in stack
+
+
+def test_stuck_detector_does_not_trigger_before_timeout() -> None:
+    """Test that stuck detector doesn't trigger when operation completes quickly."""
+    with patch.object(
+        __import__("arroyo.utils.stuck_detector", fromlist=["logger"]).logger,
+        "warning",
+    ) as mock_warn:
+        with stuck_detector(timeout_seconds=30):
+            pass
+
+        mock_warn.assert_not_called()
+
+
+def test_stuck_detector_logs_after_timeout() -> None:
+    """Test that stuck detector logs warning with stack traces when timeout exceeded."""
+    warning_logged = threading.Event()
+
+    def mock_warning(*args: object, **kwargs: object) -> None:
+        warning_logged.set()
+
+    with patch.object(
+        __import__("arroyo.utils.stuck_detector", fromlist=["logger"]).logger,
+        "warning",
+        side_effect=mock_warning,
+    ) as mock_warn:
+        with stuck_detector(timeout_seconds=0.05):
+            warning_logged.wait(timeout=2)
+
+        mock_warn.assert_called_once()
+        call_args = mock_warn.call_args
+        assert "Operation stuck" in call_args[0][0]
+        assert "MainThread" in call_args[0][3]