)]}'
{"oslo_messaging/rpc/dispatcher.py":[{"author":{"_account_id":6928,"name":"Ben Nemec","email":"openstack@nemebean.com","username":"bnemec"},"change_message_id":"3a7b50d815d774ce5dfea8fe535a9fbd9fc5f990","unresolved":false,"context_lines":[{"line_number":225,"context_line":"                # client has died. Nothing to do here but exit the watchdog"},{"line_number":226,"context_line":"                # thread. If the client is still alive (dead broker) then its"},{"line_number":227,"context_line":"                # RPC will timeout as expected."},{"line_number":228,"context_line":"                LOG.debug(\"Call-monitor heartbeat failed: %(exc)s\""},{"line_number":229,"context_line":"                          % ({\u0027exc\u0027: exc}))"},{"line_number":230,"context_line":"                break"},{"line_number":231,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"3fce034c_a2fbd876","line":228,"updated":"2019-04-18 16:03:59.000000000","message":"I guess this isn\u0027t something we generally need to make operators aware of because the RPC timeout will let them know something went wrong, and that\u0027s really what they\u0027ll care about.","commit_id":"f948e24294dfb7dfa1342d64da749c3adeb5e2e5"}],"oslo_messaging/tests/rpc/test_dispatcher.py":[{"author":{"_account_id":28522,"name":"Hervé Beraud","email":"herveberaud.pro@gmail.com","username":"hberaud"},"change_message_id":"a7190ce58e8400d97308906599661e46e72b9546","unresolved":false,"context_lines":[{"line_number":302,"context_line":""},{"line_number":303,"context_line":"        # only one call to heartbeat should be made since the watchdog thread"},{"line_number":304,"context_line":"        # should exit on the first exception thrown"},{"line_number":305,"context_line":"        self.assertEqual(1, incoming.heartbeat.call_count)"}],"source_content_type":"text/x-python","patch_set":1,"id":"5fc1f717_d1f48d5d","line":305,"updated":"2019-04-09 13:39:22.000000000","message":"Maybe you can check `assertRaises MessageDeliveryFailure` instead of count the calls","commit_id":"f948e24294dfb7dfa1342d64da749c3adeb5e2e5"},{"author":{"_account_id":6928,"name":"Ben Nemec","email":"openstack@nemebean.com","username":"bnemec"},"change_message_id":"3a7b50d815d774ce5dfea8fe535a9fbd9fc5f990","unresolved":false,"context_lines":[{"line_number":302,"context_line":""},{"line_number":303,"context_line":"        # only one call to heartbeat should be made since the watchdog thread"},{"line_number":304,"context_line":"        # should exit on the first exception thrown"},{"line_number":305,"context_line":"        self.assertEqual(1, incoming.heartbeat.call_count)"}],"source_content_type":"text/x-python","patch_set":1,"id":"3fce034c_82433caf","line":305,"in_reply_to":"5fc1f717_d1f48d5d","updated":"2019-04-18 16:03:59.000000000","message":"But that exception isn\u0027t raised anywhere that it could be caught in the test. The point here is that the exception is handled sanely by the calling code. At least IIUC.","commit_id":"f948e24294dfb7dfa1342d64da749c3adeb5e2e5"}]}