From 3f0d8e7f471475cf328666435bbfbf36cf7de36c Mon Sep 17 00:00:00 2001
From: Robert Brennan <accounts@rbren.io>
Date: Thu, 29 May 2025 15:52:13 -0400
Subject: [PATCH] add more logging to debug runtime restarts (#8799)

---
 openhands/runtime/impl/remote/remote_runtime.py | 17 +++++++++++++----
 openhands/server/listen_socket.py               |  6 ++++++
 2 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/openhands/runtime/impl/remote/remote_runtime.py b/openhands/runtime/impl/remote/remote_runtime.py
index dc269124c2..c3c20d58cc 100644
--- a/openhands/runtime/impl/remote/remote_runtime.py
+++ b/openhands/runtime/impl/remote/remote_runtime.py
@@ -101,7 +101,15 @@ class RemoteRuntime(ActionExecutionClient):
 
     def log(self, level: str, message: str, exc_info: bool | None = None) -> None:
         message = f'[runtime session_id={self.sid} runtime_id={self.runtime_id or "unknown"}] {message}'
-        getattr(logger, level)(message, stacklevel=2, exc_info=exc_info)
+        getattr(logger, level)(
+            message,
+            stacklevel=2,
+            exc_info=exc_info,
+            extra={
+                'session_id': self.sid,
+                'runtime_id': self.runtime_id,
+            },
+        )
 
     @property
     def action_execution_server_url(self) -> str:
@@ -285,9 +293,10 @@ class RemoteRuntime(ActionExecutionClient):
             f'{self.config.sandbox.remote_runtime_api_url}/resume',
             json={'runtime_id': self.runtime_id},
         )
+        self.log('info', 'Runtime resumed, waiting for it to be alive...')
         self._wait_until_alive()
         self.setup_initial_env()
-        self.log('debug', 'Runtime resumed.')
+        self.log('info', 'Runtime resumed and alive.')
 
     def _parse_runtime_response(self, response: httpx.Response) -> None:
         start_response = response.json()
@@ -407,7 +416,7 @@ class RemoteRuntime(ActionExecutionClient):
                             f'{self.config.sandbox.remote_runtime_api_url}/pause',
                             json={'runtime_id': self.runtime_id},
                         )
-                        self.log('debug', 'Runtime paused.')
+                        self.log('info', 'Runtime paused.')
                 except Exception as e:
                     self.log('error', f'Unable to pause runtime: {str(e)}')
                     raise e
@@ -420,7 +429,7 @@ class RemoteRuntime(ActionExecutionClient):
                     f'{self.config.sandbox.remote_runtime_api_url}/stop',
                     json={'runtime_id': self.runtime_id},
                 )
-                self.log('debug', 'Runtime stopped.')
+                self.log('info', 'Runtime stopped.')
         except Exception as e:
             self.log('error', f'Unable to stop runtime: {str(e)}')
             raise e
diff --git a/openhands/server/listen_socket.py b/openhands/server/listen_socket.py
index 20a96c9f81..365c63d1d0 100644
--- a/openhands/server/listen_socket.py
+++ b/openhands/server/listen_socket.py
@@ -90,6 +90,9 @@ async def connect(connection_id: str, environ: dict) -> None:
             )
             latest_event_id = -1
         conversation_id = query_params.get('conversation_id', [None])[0]
+        logger.info(
+            f'Socket request for conversation {conversation_id} with connection_id {connection_id}'
+        )
         raw_list = query_params.get('providers_set', [])
         providers_list = []
         for item in raw_list:
@@ -112,6 +115,9 @@ async def connect(connection_id: str, environ: dict) -> None:
         user_id = await conversation_validator.validate(
             conversation_id, cookies_str, authorization_header
         )
+        logger.info(
+            f'User {user_id} is allowed to connect to conversation {conversation_id}'
+        )
 
         conversation_init_data = await setup_init_convo_settings(user_id, providers_set)
         agent_loop_info = await conversation_manager.join_conversation(