Re: Re[2]: s6-rc-repo and initializing compiled databases

From: Guillermo <gdiazhartusch_at_gmail.com>
Date: Sat, 25 Apr 2026 13:09:07 -0300

Hello!

El sáb, 25 abr 2026 a las 9:52, Laurent Bercot escribió:
>
> The point of "s6 system boot", in my mind, is to provide the simplest
> possible command to get the system operational, be it for root services
> at boot time or user services at session start time.

That's the idea I got from reading the documentation —especially, the
part about calling it from rc.init—.

> Do you honestly think I should add an option to
> "s6 system boot" to pass -p to s6-rc-init? I believe this
> would add useless complexity, but you are the users, not me.

I don't have an opinion for or against it at the moment. My interest
and experience so far has been primarily using s6 + s6-rc +
s6-linux-init as an init system, and all I can say is that, in that
context, I've never used -p or felt I needed it. I used it only once,
in tests, to make sure I understood what it did and how it worked :)

Even though I have read its documentation, and it is installed on my
test system (Gentoo has packaged it! [1]) I still haven't gotten
around to testing s6-frontend with my own setup, so I also haven't
gained experience with it yet.

But kurth4cker's setup, as presented in the first messages, does use
-p in the exact way that you explained. And, since exploring the use
of s6-frontend for that case was mentioned, I felt that it was worth
noting that 's6 system boot' can't pass -p.

I still have to read and process kurth4cker's latest reply to
understand the intended usage, and, in particular, what does having
s6-rc databases live at the same time solve...

G.

[1] https://packages.gentoo.org/packages/sys-apps/s6-frontend
Received on Sat Apr 25 2026 - 18:09:07 CEST

This archive was generated by hypermail 2.4.0 : Sat Apr 25 2026 - 18:09:57 CEST