I have just found this thread and it matches my experience after upgrading (I switched laptops, from an ancient setup with Yosemite to macOS 10.3.6 and BTT 2.530).
Data points:
keyboard/touchpad freeze-unfreeze cycles until reboot, though it recovered itself at times
happened with no USB devices connected at all, while switching between apps (to MacPass which may use secure input or something)
happened during a zoom conference (I saw the other video all the time, video from my camera was frozen together with non-working keyboard/touchpad), can't remember if I was switching apps at the time (possibly)
plugging in an external mouse does not unfreeze (the external mouse doesn't work either)
frequent menu bar updates via bitbar (every 30 sec I think)
Kernel logs snippet (from dmesg) after a stall:
loginwindow is not entitled for IOHIDLibUserClient keyboard accessstalling for detach from IOHIDSystem
IOConsoleUsers: time(0) 0->0, lin 1, llk 0,
IOConsoleUsers: gIOScreenLockState 1, hs 0, bs 0, now 0, sm 0x0
stalling for detach from IOHIDSystem
Starting poll type 4
Restarting poll type 4
Command 0x2 completed. inCount: 2 data:0x5Command 0x1 completed. inCount: 2 data:0x1001Command 0x16 completed. inCount: 2 data:0x40e0Command 0xa completed. inCount: 2 data:0x0Command 0xb completed. inCount: 2 data:0x0Command 0x12 completed. inCount: 2 data:0xffffCommand 0x13 completed. inCount: 2 data:0xffffCommand 0xf completed. inCount: 2 data:0x2145Command 0x10 completed. inCount: 2 data:0x2221SmartBattery: finished polling type 4
stalling for detach from IOHIDSystem
Mhh maybe, I'll check whether I can easily integrate the fix in the old code base. In general I assume this will be fixed with one of the next macOS updates as it's a bug in macOS as far as I can tell.
in case it's still happening more samples taken while it happens would be great andreas@folivora.ai.
You can create a sample of BTT using this terminal command (hit enter then switch the window):
Also v2.532 alpha contains another fix that might be remotely connected, so updating to that version is also recommended.
I will soon put together a bugreport and send it to Apple so it can hopefully be fixed at a system level eventually, but I need to investigate a bit further first.
Mhm your samples look very different than the ones I got before. Is the behavior for you the same as described in other posts here? I.e. freezing when switching between windows / apps?
How is the alpha working for everybody else so far?
//edit I have uploaded 2.533 alpha which also disables the function that could be problematic in your case @amnmachado (it shouldn't be necessary anymore anyways - it was used to workaround a bug in an earlier High Sierra version)
No problems here yet, however the longest uninterrupted period I've had is 24 hours. I ran the first alpha, 531, for about 12 hours before the memory-leak-fix version 532 came out, which I ran for 24 hours before I then had to reboot to apply the latest Mac OS update. As I was rebooting for that anyway, I also applied the latest BTT, 533.
For me the issues used to appear at 3-5 day intervals, with any restart of BTT or the system resetting that clock, so I wouldn't have expected to have any problems yet even if the issue isn't fixed. I will now try to stay running on 533 uninterrupted for a few days so I can hopefully confirm it's fixed the issue.
I have been running the different Alphas as well. Till now everything works perfectly, but I need to wait for a longer period of time (since I as well installed the different versions and usually I get errors after a few hours). If it's still fine tomorrow evening I think you found the issue. I'll keep you updated.
Update:
Since I can't write a fourth post: Another day later, still no issues.
However, since for everybody else it seems to be solved, I will try to explain how I can easily reproduce the issue to see if it's maybe due to a different reason from the problem reported here.
I noticed this problem after installing the 10.13.6. I'm positive that the problem just started after that update. I'm also sure that it is somehow related to BetterTouchTool since I can reproduce it multiple times.
For me, the best way to "make it happen" is quite simple:
Open a PDF (however, the problem occurs also while browsing, Word, Excel, etc)
Start scrolling up and down a bit
After a while, PDF scrolling starts to hang...a "small scroll" will hang and then it recovers, scrolling very fast to "catch up" the delay.
This is the same behaviour I can observe using the browser for example.
Stopping BetterTouchTool fully solves the issue. Starting it again brings the issue back. I can reproduce this multiple times if needed.
Should I make more samples (using the method suggested before) to help you debug this?
Once again, thank you so much for the support. BetterTouchTool easily got a place on my workflow so I will do anything needed to help you fix this! Thanks!!
this sounds like a different issue than the other ones in this thread. Do you maybe have the app "Smooth Mouse" installed? (or any similar mouse related apps?). There were some (unsolvable) conflicts with such tools that could lead to such broken scrolling behavior.
