Fix SentryMetrics dropping MetricsFrame from stop_ttfb/stop_processing

SentryMetrics.stop_ttfb_metrics and stop_processing_metrics called the
base FrameProcessorMetrics implementation but discarded its return
value (implicit `return None`). FrameProcessorMetrics.stop_ttfb_metrics
/ stop_processing_metrics build and return a MetricsFrame, which
FrameProcessor.stop_ttfb_metrics / stop_processing_metrics then pushes
downstream so observers (e.g. UserBotLatencyObserver,
MetricsLogObserver) can see TTFB / processing metrics.

Because SentryMetrics returned None, the FrameProcessor never pushed
the MetricsFrame, so any pipeline using metrics=SentryMetrics() on STT
/ LLM / TTS services silently lost all downstream TTFB and processing
MetricsFrames. The metrics were still calculated and logged
internally, and Sentry transactions still finished correctly, but
observers never saw them.

Forward the MetricsFrame returned by the base class so FrameProcessor
can push it into the pipeline.
This commit is contained in:
Radhika Gupta
2026-04-17 14:48:36 +05:30
parent fc1c3b48dc
commit 80fecab4de
2 changed files with 17 additions and 2 deletions

1
changelog/4325.fixed.md Normal file
View File

@@ -0,0 +1 @@
- Fixed `SentryMetrics` silently dropping `MetricsFrame`s from `stop_ttfb_metrics` and `stop_processing_metrics`. `SentryMetrics` called the base `FrameProcessorMetrics` implementation but discarded its return value, so `FrameProcessor` never pushed the `MetricsFrame` downstream. This prevented observers (e.g. `UserBotLatencyObserver`, `MetricsLogObserver`) from seeing TTFB and processing metrics for any service using `metrics=SentryMetrics()`. The metrics were still calculated and Sentry transactions still completed — only the downstream frame push was affected.

View File

@@ -97,13 +97,20 @@ class SentryMetrics(FrameProcessorMetrics):
Args:
end_time: Optional end timestamp override.
Returns:
MetricsFrame produced by the base class, or None if not measuring.
Returning the frame is required so ``FrameProcessor.stop_ttfb_metrics``
can push it downstream to observers.
"""
await super().stop_ttfb_metrics(end_time=end_time)
frame = await super().stop_ttfb_metrics(end_time=end_time)
if self._sentry_available and self._ttfb_metrics_tx:
await self._sentry_queue.put(self._ttfb_metrics_tx)
self._ttfb_metrics_tx = None
return frame
async def start_processing_metrics(self, *, start_time: float | None = None):
"""Start tracking frame processing metrics.
@@ -126,13 +133,20 @@ class SentryMetrics(FrameProcessorMetrics):
Args:
end_time: Optional end timestamp override.
Returns:
MetricsFrame produced by the base class, or None if not measuring.
Returning the frame is required so ``FrameProcessor.stop_processing_metrics``
can push it downstream to observers.
"""
await super().stop_processing_metrics(end_time=end_time)
frame = await super().stop_processing_metrics(end_time=end_time)
if self._sentry_available and self._processing_metrics_tx:
await self._sentry_queue.put(self._processing_metrics_tx)
self._processing_metrics_tx = None
return frame
async def _sentry_task_handler(self):
"""Background task handler for completing Sentry transactions."""
running = True