[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Interface of `shutdown', 'halt', ... programs



Dmitry Bogatov writes ("Interface of `shutdown', 'halt', ... programs"):
> Recently I asked for inclusion of 'runit-init' into init metapackage
> Pre-Depends: (#838480). Response, in particular mentioned, that
> since 'runit-init' does not provide  'halt', 'reboot' and other
> scripts, it would break things.
> 
> Unfortunately, I did not received reply about following questions:
> 
>  - what exactly would break and what should I test? It is not clear for me,
>    since I am fine without these scripts, running runit-init_2.1.2-8.
> 
>  - what exactly interface is expected from these scripts? For example,
>    shutdown program is rather complex, involving timespecs, access control
>    and tons of command line options, and, if I need to reimplement it (sigh),
>    I would prefer to do as little, as possible.

Quite.

> It was quite misleadingly mentioned on #838480 about sysvinit-like
> system.  Runit is not, and while it support /etc/init.d/ scripts as
> good, as sysvinit, they are still fallback method of managing
> services.
> 
> Any suggestions?

Is it possible to use a pointyclicky desktoppy widgety thing to reboot
a system with runit ?  I guess from your mails you use the command
line.

I think the reference to "sysvinit-like" from Michael Biebl is mainly
there because GNOME expects either that or systemd.  In practice you
might find that the right answer is to implement enough of `shutdown'
to satisfy GNOME.

I doubt there are many non-human callers of shutdown that use all the
exciting options, and those that do exist will probably not change
very much.

I think if I were you I would try installing a system with runit and
GNOME, make enough of an emulation that the GNOME shutdown and reboot
functions work, and call that "done".

Ian.

-- 
Ian Jackson <ijackson@chiark.greenend.org.uk>   These opinions are my own.

If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.


Reply to: