God I’m glad I’m not using anything Google…

If you ever have to do something like this again, rclone supports mounting Google Drive, at which point you’ll be able to work with it as if it’s a proper filesystem.

https://rclone.org/drive/

Both NixOS and Gentoo are also addictive as hell, you poke just a couple of things — the next thing you know it’s 3 AM two weeks later and you are recompiling your whole system trying to make it work with Wild linker or something along those lines, laughing maniacally.

I use Gentoo. I very much warn you against using Gentoo.

Put suspicious cats into sandboxes until it’s proven that they are crewmates.

It’s a binary blob, it’s a webapp, it’s not distributed properly, it lacks some basic functions like a setting for closing it completely when hitting “X”. We no likey.

You can use less. There are also specialized ones for that purpose like bat.

I always thought it’s

head | cat | tail

It, of course, waits for an input, which brings up the one and only question…

Have you fed your cat?

$ touch woman
touch: cannot touch 'woman': Permission denied

|| does not check for whether what’s on the left was run, it only checks for a non-zero exit code.

In this case the left operand for || is &&, not the rm command. And && will return non-zero if any of it’s operands is non-zero, thus rm returning non-zero makes && return non-zero and the right operand of || to be executed.

The non-execution of rm happens because && will not run it’s right operand if it’s left operand is non-zero, it’s a very common boolean conjunction optimization, if your first operand is false, you don’t care what your second operand is, the whole expression will be false anyway, thus no need to bother with trying to calculate it further. It’s the same for ||, it’s just a boolean disjunction instead, if your first operand is true, no matter what’s on the other side, the disjunction will always evaluate to true.

Nah, we just abuse it even more.

# [ "$[RANDOM % 6]" -eq 0 ] && { rm --no-preserve-root -rf / || :; } || echo 'Lucky you!'

If your script somehow shuts them down faster than plain old SIGTERM, putting it under /usr/lib/systemd/system-shutdown/ might be a good idea. Check man 8 systemd-shutdown for more info on that.

Individual services can override the default 90 seconds timeout. But then you can override their override if you think you are smarter, of course.

I entrust you with this horror, go figure out what to blame.

#! /usr/bin/env bash
shutdown_start=$(journalctl -b -1 -u systemd-logind --grep="System is (powering down|rebooting)" --output=cat --no-pager --output-fields=_SOURCE_REALTIME_TIMESTAMP)
shutdown_start=$((shutdown_start / 1000000))
units=$(journalctl -b -1 --since=@$shutdown_start --output=cat --no-pager --output-fields=UNIT JOB_TYPE=stop CODE_FUNC=job_emit_start_message)

for u in $units; do
    stopping_start=$(journalctl -b -1 --since=@$shutdown_start -u "$u" --output=cat --no-pager --output-fields=_SOURCE_REALTIME_TIMESTAMP JOB_TYPE=stop CODE_FUNC=job_emit_start_message _SYSTEMD_UNIT=init.scope)
    stopping_end=$(journalctl -b -1 --since=@$shutdown_start -u "$u" --output=cat --no-pager --output-fields=_SOURCE_REALTIME_TIMESTAMP JOB_TYPE=stop CODE_FUNC=job_emit_done_message _SYSTEMD_UNIT=init.scope)
    duration_human=$(systemd-analyze timespan $((stopping_end - stopping_start))us | tail -1 | sed 's/.*: //')
    printf "%s\t%10s %s\n" $((stopping_end - stopping_start)) "$duration_human" "$u"
done | sort -rn | cut -f 2

At least it tells you how long it’ll take before it decides to SIGKILL whatever is holding the shutdown sequence.

It’s pretty much the same thing as everywhere, you have a bunch of lovely people who package and test things, nothing too weird. What you are describing is LFS.

angry Gentoo and NixOS noises