enhancementhelp wanted
Repository metrics
- Stars
- (3,844 stars)
- PR merge metrics
- (No merged PRs in 30d)
Description
Right now our existing model_log.log logs one line per line in a stacktrace error which means it's difficult to search for using a logging platform like splunk and creates too many events with the exact same date.
2021-10-06 17:25:11,286 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - --- Logging error ---
2021-10-06 17:25:11,286 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - Traceback (most recent call last):
2021-10-06 17:25:11,286 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - File "/usr/lib/python3.6/logging/__init__.py", line 996, in emit
2021-10-06 17:25:11,286 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - stream.write(msg)
2021-10-06 17:25:11,286 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - UnicodeEncodeError: 'ascii' codec can't encode character '\u201d' in position 55: ordinal not in range(128)
2021-10-06 17:25:11,286 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - Call stack:
2021-10-06 17:25:11,286 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - File "/usr/local/lib/python3.6/dist-packages/ts/model_service_worker.py", line 183, in <module>
2021-10-06 17:25:11,287 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - worker.run_server()
2021-10-06 17:25:11,287 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - File "/usr/local/lib/python3.6/dist-packages/ts/model_service_worker.py", line 155, in run_server
2021-10-06 17:25:11,287 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - self.handle_connection(cl_socket)
2021-10-06 17:25:11,287 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - File "/usr/local/lib/python3.6/dist-packages/ts/model_service_worker.py", line 114, in handle_connection
2021-10-06 17:25:11,287 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - resp = service.predict(msg)
2021-10-06 17:25:11,287 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - File "/usr/local/lib/python3.6/dist-packages/ts/service.py", line 100, in predict
2021-10-06 17:25:11,287 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - ret = self._entry_point(input_batch, self.context)
2021-10-06 17:25:11,287 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - File "/usr/local/lib/python3.6/dist-packages/search_model_serving_handler/search_base_handler.py", line 354, in handle
2021-10-06 17:25:11,287 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - data_preprocess = self.preprocess(data)
2021-10-06 17:25:11,287 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - File "/usr/local/lib/python3.6/dist-packages/search_model_serving_handler/search_base_handler.py", line 131, in preprocess
2021-10-06 17:25:11,287 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - logger.info("Preprocesing request: '%s'", single_request)
2021-10-06 17:25:11,287 [WARN ] W-9001-7301_1.0-stderr MODEL_LOG - Message: "Preprocesing request: '%s'"
Describe the solution
Solution probably looks like changing this line to not create a new line https://github.com/pytorch/serve/blob/2d5204a4cad9242694692ba9c8926d7ab8e8553b/frontend/server/src/main/java/org/pytorch/serve/wlm/WorkerLifeCycle.java#L244