Feature Request: Decrease systemd stop job timeout like Fedora 38 did from 2m 45s to 45s

[BS86]

@kagetora13

What suprises me is why such a specific time change … 2min and 45 seconds … to 45 seconds? Why not an even 1 minute.

Maybe slower CPU need the extra time to cleanly shutdown a service?

The commit that Fedora Dev’s initially wanted to upstream to systemd was 15 seconds
Systemd devs asked them to not make such a drastic change but first try a step in between but the upstream pull request somewhat stalled then.
They seem to test the 45s in Fedora and check if users report issues - if not they will certainly lower it further or try to upstream the 45s - let’s see how they decide and what the systemd devs decide.

@dalto

Many people run containers on their desktops to isolate applications or provide services…

There are many services people run on the desktop that need more than 10 seconds to shut down.

well:

@BS86

those services that legitimately take longer than 45 seconds can set their own timeout which can be higher than 45 seconds and that way supersede the default Timeout.

But apparently, there is even another approach for those services, taken from the Fedora-Wiki:

Our users shouldn’t have to randomly sit waiting for their machine to shutdown. It will also encourage the correct use of shutdown inhibit APIs.

If the services would use those “correct […] shutdown inhibit APIs” the issue would also be a non-issue.

[dalto]

@BS86

those services that legitimately take longer than 45 seconds can set their own timeout which can be higher than 45 seconds and that way supersede the default Timeout.

Sure, that is possible but the reality is that most don’t services don’t set that so EndeavourOS changing the default will just cause stuff to break. However, if the change is made upstream, that is different as it could motivate more services to properly take different actions.

Our path should be to follow the upstream defaults and if those default change, that is fine.