athenad: fix test timeouts and comments (#31540)

* these pass in 0.5s since server sends ping on connect

* comments

* unused

* fix

* fix these too

* check end_event while uploading, throw abort exception if we need to shut down/restart

* Revert "check end_event while uploading, throw abort exception if we need to shut down/restart"

This reverts commit f0b822fca9.

* more tol for lte connection

selfdrive/athena/athenad.py

@@ -746,6 +746,9 @@ def ws_manage(ws: WebSocket, end_event: threading.Event) -> None:
       onroad_prev = onroad
 
       if sock is not None:
+        # While not sending data, onroad, we can expect to time out in 7 + (7 * 2) = 21s
+        #                         offroad, we can expect to time out in 30 + (10 * 3) = 60s
+        # FIXME: TCP_USER_TIMEOUT is effectively 2x for some reason (32s), so it's mostly unused
         sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_USER_TIMEOUT, 16000 if onroad else 0)
         sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPIDLE, 7 if onroad else 30)
         sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPINTVL, 7 if onroad else 10)

selfdrive/athena/tests/test_athenad_ping.py

@@ -12,6 +12,8 @@ from openpilot.selfdrive.athena import athenad
 from openpilot.selfdrive.manager.helpers import write_onroad_params
 from openpilot.system.hardware import TICI
 
+TIMEOUT_TOLERANCE = 20  # seconds
+
 
 def wifi_radio(on: bool) -> None:
   if not TICI:
@@ -63,7 +65,7 @@ class TestAthenadPing(unittest.TestCase):
     mock_create_connection.assert_called_once()
     mock_create_connection.reset_mock()
 
-    # check normal behaviour
+    # check normal behaviour, server pings on connection
     with self.subTest("Wi-Fi: receives ping"), Timeout(70, "no ping received"):
       while not self._received_ping():
         time.sleep(0.1)
@@ -92,12 +94,12 @@ class TestAthenadPing(unittest.TestCase):
   @unittest.skipIf(not TICI, "only run on desk")
   def test_offroad(self) -> None:
     write_onroad_params(False, self.params)
-    self.assertTimeout(100)  # expect approx 90s
+    self.assertTimeout(60 + TIMEOUT_TOLERANCE)  # based using TCP keepalive settings
 
   @unittest.skipIf(not TICI, "only run on desk")
   def test_onroad(self) -> None:
     write_onroad_params(True, self.params)
-    self.assertTimeout(30)  # expect 20-30s
+    self.assertTimeout(21 + TIMEOUT_TOLERANCE)
 
 
 if __name__ == "__main__":