Agent stuck with Error processing netlog event


we are running our private WPT agent on dedicated windows server.
The agent works great for 30-40 hours with 30 test/h but sometime stuck with open browser (chrome) and the error “Error processing netlog event” on screen.
To restart the agent i need to close the browser and the py script, after that it restart without any problem.
We have 5 agent and all have the same problem after differente time amount.


I checked on the forum but all search gave negative results.
Could someone tell me some ideas for analysis or any fixes?

Agent: current last version
Machine: dedicated windows server 2019
Chrome: 89.0

Would you mind filing an issue here?

Is the agent code sync’d with the latest in github or is it using the latest packaged release? I fixed a timeout condition in trace processing a few months ago which is why I am wondering (could be another one).

The actual error with the netlog event can be ignored. It was handled, ignored and moved on. It looks like it hung waiting for the end of the trace events. There was a bug where the wait timeout wasn’t actually getting updated but it should have been fixed.