Oh ok. Thanks for the info 👍

Oh ok. Pyautogui is the one I got working in x11.

Oh bummer. Anyway I’ve made an earlier post from my previous efforts (https://lemmy.world/post/49303063) where I also tested ydotool. At that time there was a known bug in ydotool that prevented absolute mousemove from working properly. The issue is still open on github it seems.

Have a PC with 16gb ram that lives up to the overwatch game system reqs. When I installed it and launched it, it began building some sort of cache. A one time job. During that chache my os would force close steam to prevent an oom crash. I then added 16gb swap and during the owerwatch cache job the used phys ram would max out at 12gb I think and the swap sent up to 10gb. So a 22gb ram job. Yikes. Thanks swap ram 👍

This was a brain fart as the bug specifically prevents absolutte movements so I cannot move the mouse to any specified location. All I can do is move the mouse so much in one direction relatively that it most definitely is at the edge/corner of the screen.

I skipped Input-remapper as the setup for it turned out to be pretty complex and also I need to be able to check current mouse position which it can’t.

I have currrently given up on the project.

I did think about moving to an abritary base position and then move from there. The edge of the screen wont work as it will move the view, but any other position could work. I just dont like that I have to make a workaround for basic functionality. I’ll try input-remapper and see if it works better. If not I might go back and try this.

I’ve been testing ydotool, but I end up running into this bug

https://github.com/ReimuNotMoe/ydotool/issues/250

Which I can’t seem to get around. The workaround to move the mouse to 0,0 and then making non-absolute moves clashes with the game (warcraft 3) because when the mouse is at the edge of the screen it scrolls the view which is really annoying.

I’m going to test input-remapper next.

Just semi automatic, I want a hotkey press activate a mouse move -> click -> move action.

I’m making a hotkey tool for classic warcraft 3 1.27b, so you can calm down.

Well said, sadly

Thank you for your thorough analysis and report. Very interesting read. Just doing the basics, as you say, is more than a layman like me can do!

I had a similar issue on an HP Elitedesk 800 mini G3. It turned out to be a faulty ram stick. Memtest revealed the issue and after unpligging the defective stick the issue was resolved.

Hi, thanks. I’ve been testing out NextCloud and it might be a bit too bloated and buggy for me (document editor iffy on mobile and encountering files and folders being falsely locked and thus cannot be interacted with) so I might give sftp a go.

Is there a dockerhub project you would reccomend? I’ve been looking at sftpgo (https://hub.docker.com/r/drakkan/sftpgo) as a option?

To prevent brute force login attempts into my jellyfin service.

Oh I can see how my title could be misunderstood - what I meant is that both jellyfin and fail2ban are running in (separate) docker containers. Not both services crammed.into one.

Thank you. I did see this list before. My jellyfin instance is not exposed to the net atm., but I’m thinking of exposing it in intervals and would like to have fail2ban working when/if I do.

I did the same. Not because I knew, but because I was unsure if permissions would fail if I didn’t.

Do I just enter the container shell and check GID and UID of a random file?

I’m CURSED I tells ya! I’ll look at the logs tomorrow. Good to know that you can get it working. Maybe I’m close then.

Do I need to enter the container bash and change something then? I tried adding UID and GID to the docker compose file, but it still fails. I updated the google docs notes if you want to see my steps.