-
Notifications
You must be signed in to change notification settings - Fork 82
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Lichess sends invalid move #151
Comments
I have created a new issue because although the result is similar, the cause is completely different. It looks like the local board does not match with Lichess' board or Lichess sent an invalid move for some other reason. @IbaiBuR Can you please post the link to the game. |
Thanks for moving it, you a right, the result is similar but the cause is completely different. Link to the game: https://lichess.org/eN0MpXmc (my bot is black). Full logs of the game:
I also analyzed the position locally to check if it was related to the engine or something, I got a completely different move tho:
|
To be honest, I would just blame it on the Lichess reboot that was done at the time. ToromBot also got stuck, which normally doesn't happen very often anymore. RaspFish continues to run normally. I would be interested to know what the |
Ah no, we're black, then that's not true. |
Yes my bot is black, I think the Lichess reboot has probably something to do with it too |
So it looks like we got a Edit: On the other hand, we would have had to make a move after the imaginary Very interesting but probably caused by the Lichess reboot, as long as it doesn't happen again I'd say it's not a problem. |
Yes, quite weird so far, I was watching the game live and I was shocked when I saw my engine giving a rook for free somehow, which made absolutely no sense. I know BotLi does not make up the moves, thats why I was surprised. So yeah, I think this is solely because of the lichess update because if not I cant find the reason. |
After the error BotLi ignored the game and started matchmaking again, so I guess that is why there were 2 instances of the engine running. That is why my bot lost on time. See:
|
Well, for |
In the game the NPS were normal (1.2M which is what I normally get with 2 threads), the second instance was launched after the error. |
That's what I thought. BotLi does not terminate the engine correctly when an error occurs, I would probably have to solve that better. |
I see :) At least we were able to spot where the problem was, still very weird though |
Created from @IbaiBuR
This is the error I got during the game:
Along with the error, the engine made an stupid move giving up a rook for free (I dont even know why), so, afterwards, I checked btop (similar to htop) because the server's CPU usage stagnated for some reason, and I saw this:
So 2 instances of the engine were runninng at once, for some unknown reason. Even if I am using my own stockfish-derived engine, I do not think the error is related to that, because it never happened to me before.
Some insights about the environment the bot is running in:
The book part is missing since I prefer to keep it private and also the blacklist/whitelist one.
The text was updated successfully, but these errors were encountered: