[Serve] Add log whenever a replica's health check succeeds after fail…

…ing (ray-project#44162) This change makes the Serve controller emit a log whenever the replica's health check succeeds after failing. --------- Signed-off-by: Shreyas Krishnaswamy <[email protected]> Co-authored-by: Edward Oakes <[email protected]>
Mark2000 · Mar 22, 2024 · b83d4d6 · b83d4d6
1 parent c384c1a
commit b83d4d6
Showing 1 changed file with 5 additions and 0 deletions.
diff --git a/python/ray/serve/_private/deployment_state.py b/python/ray/serve/_private/deployment_state.py
@@ -824,6 +824,11 @@ def check_health(self) -> bool:
         elif response is ReplicaHealthCheckResponse.SUCCEEDED:
             # Health check succeeded. Reset the consecutive failure counter
             # and mark the replica healthy.
+            if self._consecutive_health_check_failures > 0:
+                logger.info(
+                    f"{self._replica_id} passed the health check after "
+                    f"{self._consecutive_health_check_failures} consecutive failures."
+                )
             self._consecutive_health_check_failures = 0
             self._healthy = True
         elif response is ReplicaHealthCheckResponse.APP_FAILURE: