Good point. I’ll have to stop using immutable and stay with atomic (and declarative).
Interestingly /bin
and /usr/bin
are not in PATH by default, so /bin/chewy
can only be executed by its path directly and won’t affect the systems reliability.
You could write a s script which follows jellyfin logs line for line and greps/seds/awks for a keyword corresponding to starting playback. Then you could extract the file path.
It’s also likely that it’s possible to start your specific scheduled task by API like the following [1].
[1] https://github.com/jellyfin/jellyfin/discussions/12224
PS: I don’t know whether it’s possible to run a scheduled task for a specific file only. But I’ve already written the above anyway.