[GH-ISSUE #397] [Bug] After a while of moving mouse from screen to screen, mouse will randomly be "stuck" in MacOS. #205

New issue

Open

opened 2026-05-05 22:13:59 -06:00 by gitea-mirror · 2 comments

gitea-mirror commented

2026-05-05 22:13:59 -06:00

Owner

Originally created by @urwrstkn8mare on GitHub (Mar 12, 2026).
Original GitHub issue: https://github.com/feschber/lan-mouse/issues/397

I'm using Hyperland (server) and MacOS (client) and am on the main branch on both. It will work fine most of the time but then randomly the mouse will get stuck in MacOS. The only solution is to quit lan-mouse on MacOS and restart. I can make the issue happen faster by just moving my mouse in between screens very fast. Logs are completely normal (i.e. no different to when its working normally).

I'm assuming this has something to do with the MacOS client releasing capture and never getting it back (so it doesn't know when to switch back). It should be that while a server is controlling the mouse on the client, it should always be checking if the mouse wants to leave. I'm not sure on the architecture of lan-mouse but I don't see how these states could be decoupled.

A bandage fix could be either of the changes you proposed here: https://github.com/feschber/lan-mouse/issues/164#issuecomment-2466214690

Otherwise pretty cool app

Originally created by @urwrstkn8mare on GitHub (Mar 12, 2026). Original GitHub issue: https://github.com/feschber/lan-mouse/issues/397 I'm using Hyperland (server) and MacOS (client) and am on the `main` branch on both. It will work fine most of the time but then randomly the mouse will get stuck in MacOS. The only solution is to quit `lan-mouse` on MacOS and restart. I can make the issue happen faster by just moving my mouse in between screens very fast. Logs are completely normal (i.e. no different to when its working normally). I'm assuming this has something to do with the MacOS client releasing capture and never getting it back (so it doesn't know when to switch back). It should be that while a server is controlling the mouse on the client, it should *always* be checking if the mouse wants to leave. I'm not sure on the architecture of `lan-mouse` but I don't see how these states could be decoupled. A bandage fix could be either of the changes you proposed here: https://github.com/feschber/lan-mouse/issues/164#issuecomment-2466214690 Otherwise pretty cool app

gitea-mirror commented

2026-05-05 22:14:01 -06:00

Author

Owner

@marstaik commented on GitHub (Mar 19, 2026):

I have the opposite issue, driving from MacOS onto linux, and the mouse gets stuck onto Linux. Same solution, I have to kill it on linux and restart it.

@marstaik commented on GitHub (Mar 19, 2026): I have the opposite issue, driving from MacOS onto linux, and the mouse gets stuck onto Linux. Same solution, I have to kill it on linux and restart it.

gitea-mirror commented

2026-05-05 22:14:01 -06:00

Author

Owner

@feschber commented on GitHub (Apr 29, 2026):

might be fixed by #417

@feschber commented on GitHub (Apr 29, 2026): might be fixed by #417