[Core] Add a better error message for health checking network failures (

#36957) (#37366) This PR adds a better error message for health checking network failures Currently, it doesn't log why health check RPCs fail.
ray-project · Jul 13, 2023 · c1db53a · c1db53a
1 parent ab2a4f1
commit c1db53a
Showing 1 changed file with 6 additions and 2 deletions.
diff --git a/src/ray/gcs/gcs_server/gcs_health_check_manager.cc b/src/ray/gcs/gcs_server/gcs_health_check_manager.cc
@@ -105,8 +105,12 @@ void GcsHealthCheckManager::HealthCheckContext::StartHealthCheck() {
                 health_check_remaining_ = manager_->failure_threshold_;
               } else {
                 --health_check_remaining_;
-                RAY_LOG(WARNING) << "Health check failed for node " << node_id_
-                                 << ", remaining checks " << health_check_remaining_;
+                RAY_LOG(WARNING)
+                    << "Health check failed for node " << node_id_
+                    << ", remaining checks " << health_check_remaining_ << ", status "
+                    << status.error_code() << ", response status " << response_.status()
+                    << ", status message " << status.error_message()
+                    << ", status details " << status.error_details();
               }
 
               if (health_check_remaining_ == 0) {