chore(similarity): Add seer failure reason to backfill log (#74735)

jangjodi · web-flow · commit fe2c37c41c1d · 2024-07-23T10:32:29.000-07:00
Add the reason seer post request failed to the similarity seer failure
backfill log
diff --git a/src/sentry/seer/similarity/grouping_records.py b/src/sentry/seer/similarity/grouping_records.py
@@ -36,6 +36,7 @@ class CreateGroupingRecordsRequest(TypedDict):
 class BulkCreateGroupingRecordsResponse(TypedDict):
     success: bool
     groups_with_neighbor: NotRequired[dict[str, RawSeerSimilarIssueData]]
+    reason: NotRequired[str | None]
 
 
 seer_grouping_connection_pool = connection_from_url(
@@ -69,15 +70,15 @@ def post_bulk_grouping_records(
     except ReadTimeoutError:
         extra.update({"reason": "ReadTimeoutError", "timeout": POST_BULK_GROUPING_RECORDS_TIMEOUT})
         logger.info("seer.post_bulk_grouping_records.failure", extra=extra)
-        return {"success": False}
+        return {"success": False, "reason": "ReadTimeoutError"}
 
     if response.status >= 200 and response.status < 300:
         logger.info("seer.post_bulk_grouping_records.success", extra=extra)
         return json.loads(response.data.decode("utf-8"))
     else:
         extra.update({"reason": response.reason})
         logger.info("seer.post_bulk_grouping_records.failure", extra=extra)
-        return {"success": False}
+        return {"success": False, "reason": response.reason}
 
 
 def delete_project_grouping_records(
diff --git a/src/sentry/tasks/embeddings_grouping/backfill_seer_grouping_records_for_project.py b/src/sentry/tasks/embeddings_grouping/backfill_seer_grouping_records_for_project.py
@@ -199,6 +199,7 @@ def backfill_seer_grouping_records_for_project(
             extra={
                 "current_project_id": current_project_id,
                 "last_processed_project_index": last_processed_project_index,
+                "reason": seer_response.get("reason"),
             },
         )
         sentry_sdk.capture_exception(Exception("Seer failed during backfill"))
diff --git a/tests/sentry/seer/similarity/test_grouping_records.py b/tests/sentry/seer/similarity/test_grouping_records.py
@@ -68,7 +68,7 @@ def test_post_bulk_grouping_records_success(mock_seer_request: MagicMock, mock_l
 @mock.patch("sentry.seer.similarity.grouping_records.logger")
 @mock.patch("sentry.seer.similarity.grouping_records.seer_grouping_connection_pool.urlopen")
 def test_post_bulk_grouping_records_timeout(mock_seer_request: MagicMock, mock_logger: MagicMock):
-    expected_return_value = {"success": False}
+    expected_return_value = {"success": False, "reason": "ReadTimeoutError"}
     mock_seer_request.side_effect = ReadTimeoutError(
         DUMMY_POOL, settings.SEER_AUTOFIX_URL, "read timed out"
     )
@@ -91,7 +91,7 @@ def test_post_bulk_grouping_records_timeout(mock_seer_request: MagicMock, mock_l
 @mock.patch("sentry.seer.similarity.grouping_records.logger")
 @mock.patch("sentry.seer.similarity.grouping_records.seer_grouping_connection_pool.urlopen")
 def test_post_bulk_grouping_records_failure(mock_seer_request: MagicMock, mock_logger: MagicMock):
-    expected_return_value = {"success": False}
+    expected_return_value = {"success": False, "reason": "INTERNAL SERVER ERROR"}
     mock_seer_request.return_value = HTTPResponse(
         b"<!doctype html>\n<html lang=en>\n<title>500 Internal Server Error</title>\n<h1>Internal Server Error</h1>\n<p>The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.</p>\n",
         reason="INTERNAL SERVER ERROR",

Original file line number	Diff line number	Diff line change
`@@ -199,6 +199,7 @@ def backfill_seer_grouping_records_for_project(`
`199`	`199`	`extra={`
`200`	`200`	`"current_project_id": current_project_id,`
`201`	`201`	`"last_processed_project_index": last_processed_project_index,`
	`202`	`+ "reason": seer_response.get("reason"),`
`202`	`203`	`},`
`203`	`204`	`)`
`204`	`205`	`sentry_sdk.capture_exception(Exception("Seer failed during backfill"))`