Xyne, this may be obvious, but have you disabled the key binding (was it in open box) to confirm that it is indeed a XF86Display key event that is triggering arandr?
Sorry for the late follow-up. Yeah, I rebound the key to send a notification instead, which it does every time a screen is detected.
]]>Xyne, this may be obvious, but have you disabled the key binding (was it in open box) to confirm that it is indeed a XF86Display key event that is triggering arandr?
]]>Upon re-reading it, I may have been to hasty.
It does seem the author of that considered XF86Display as "screen change event" .
- X detects a "screen change event"
Which links are a reference for this? This would definitely be new and (I believe) undocumented behavior of the X server. I grepped the source code for the xorg-server package for both that constant symbol and value, and neither were present. I also grepped for all SendEvent calls and could not find any that seemed to serve this function.
Some WMs / DEs would likely want to respond similarly to a press of that key and a screen change event, but I highly doubt the Xorg server simulates a key press in response to a screen change event.
]]>#define XF86XK_Display 0x1008FF59 /* Output switch key */
Combining info from other links[2] it seems that there are atleast 2 cases that generate an XF86Display event :
- a "display" button on laptops (could be also present on so-called multimedia keyboards)
- X detects a "screen change event"
edit
This is probably a wrong interpretation by me, see post #5 & #6 .
end-of-edit
[1]
https://wiki.archlinux.org/index.php/Xmodmap
http://wiki.linuxquestions.org/wiki/Lis … by_Xmodmap
I'm still curious about the origin of the simulated key press.
]]>You might also be able to look at the process tree of the arandr process - what are the parent process(es)?
]]>There is something on all of my Arch systems (all running Openbox) that is launching arandr whenever a new monitor is detected. I think that I first noticed this a few months ago but so far I haven't been able to track down the source. Each time I end up giving up after about 5-10 minutes and moving on.
Where I've looked so far (I may have missed something):
arandr package files (no udev or acpi rules, nothing that seems related)
/etc/udev
/etc/acpi
journalctl
Openbox configuration and log
Does anyone know what it could be?
]]>