Mouse - missing or delayed events

[mhanor]

Hello,

I have a problem with my Debian testing PC and some mouse events that are not triggered, the most obvious being the button release event not being triggered or being triggered after moving the mouse. I don't know how to debug this issue.
I've run xev and sudo libinput list-devices and the issue is visible with these tools.
I've tried with two different mouses, both wireless 2.4GHz.

System details in attachment below, kernel details:
Linux 6.10.6-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.10.6-1 (2024-08-19) x86_64 GNU/Linux
Mate Desktop with picom GLX

What more could I do to debug or fix this issue?


- evdev:
- [ 91, 128158, 4, 4, 30] # EV_MSC / MSC_SCAN 30 (obfuscated)
- [ 91, 128158, 1, 272, 1] # EV_KEY / BTN_LEFT 1
- [ 91, 128158, 0, 0, 0] # ------------ SYN_REPORT (0) ---------- +168ms
# Current time is 21:05:10
# Current time is 21:05:16
- evdev:
- [100, 982182, 4, 4, 30] # EV_MSC / MSC_SCAN 30 (obfuscated)
- [100, 982182, 1, 272, 0] # EV_KEY / BTN_LEFT 0
- [100, 982182, 0, 0, 0] # ------------ SYN_REPORT (0) ---------- +9854ms
- evdev:

[ruwolf]

Do you not have any conductive (metal) object between the mouse and the receiver?
In the last line of X.org log Enabling spurious button debouncing is mentioned, so I guess, it could be cause, but it is only my humble opinion.

[mhanor]

Part of the PC chassis/metal sits between the mouse and its receiver.
I don't know how to interpret the libinput record output to determine if it's a spurious button debouncing issue.
Thank you for your answer.

[CwF]

Enabling spurious button debouncing,

Thanks for the link.
I've battled this in the past and it can be very device dependent. Since I use a local unified logitech usb not only for local but also keyboard+mouse device only x2vnc and x2x connections I've noticed much strange behavior including losing the mouse. Worse is my bouncy button trackball that seems temperature sensitive and sometimes double clicks everything and won't hold the button, etc. I continue to deal. Interestingly, this libinput filtering is skipped when passing the device through the network with x2vnc and the button bounce is often fine under windows, that's discouraging. Through x2x this filtering is done by the remote OS and is often different. One mouse, three different interpretations!

[ruwolf]

Part of the PC chassis/metal sits between the mouse and its receiver.
I don't know how to interpret the libinput record output to determine if it's a spurious button debouncing issue.
Thank you for your answer.

Hello.
I am sorry that I did not greet you in the first reply.
In your case, I would try another input drivers, e.g. xserver-xorg-input-evdev.
But you probably should remove (or purge) xserver-xorg-input-libinput for this testing, because X.org in case of availability of multiple drivers chooses one by priority.