When sourced within a container, command_not_found_handle causes infinite loop
soulshake opened this issue · 1 comments
soulshake commented
I have PrtSc set up in my i3 profile to a script that runs a command using the AWS CLI: bindsym Print exec "~/bin/screenshot.sh"
When I run "~/bin/screenshot.sh" directly it works as expected, but when it's executed via the i3 bindsym, it spawns more and more aws processes until I kill the container.
This is what I think is happening:
If a user runs a container-command from $HOME, their homedir is mounted in the container. When a container's entrypoint isn't fully qualified, this can result in the aliased version of the command preceding the "real" command in the container-user's $PATH, resulting in an infinite loop.