On 2019-09-17 at 04:09, Curt wrote:

> On 2019-09-16, Brian <a...@cityscape.co.uk> wrote:

[that on some earlier date which has been clipped out, someone else -
who happens to be The Wanderer - wrote:]

>>> The dist-upgrade will have resulted in installing the
>>> systemd-sysv package, which (despite its name) has nothing to do
>>> with sysvinit; it is the package which sets systemd as the
>>> primary / active / default init system.
>>> 
>>> Installing sysvinit-core will uninstall that package.
>> 
>> What causes systemd-sysv to be installed?
> 
> Why he would say "despite its name" eludes this correspondent,
> because the package has *everything* to do with sysvinit, providing
> as it does the "links needed for systemd to replace sysvinit.
> Installing systemd-sysv will overwrite /sbin/init with a link to
> systemd."

Yes, but unless I'm greatly misunderstanding matters, /sbin/init is not
specific to sysvinit.

It's the path which the kernel will launch by default (i.e., when no
other is specified on the kernel command line) to start bringing up
userspace. It's hardcoded into the kernel source code.

Making _any_ init system the default involves installing an executable
from that init system to /sbin/init (even if as a symlink). This would
be just as true if systemd were replacing e.g. upstart as when it is
replacing sysvinit.

That that path and name have traditionally been taken up by a file built
as part of sysvinit is a historical artifact, and is irrelevant to
proper nomenclature.

-- 
   The Wanderer

The reasonable man adapts himself to the world; the unreasonable one
persists in trying to adapt the world to himself. Therefore all
progress depends on the unreasonable man.         -- George Bernard Shaw

Attachment: signature.asc
Description: OpenPGP digital signature

Reply via email to