Re: systemd and Linux are *fundamentally incompatible* -> and I can prove it

[Kevin Chadwick <ma1l1ists@yahoo.co.uk>]

previously on this list Jan Gloser contributed:

Kev, the systemd design document says it all about the lack of design
with statements showing a clear lack of understanding. I would be
ashamed to call it a design document.

> I would also like to ask something the people who dislike systemd (as there
> seem to be more). I am not very proficient with such internals of debian,
> but you say things like systemd breaks things and systemd has no unified
> design and sytemd is possibly a security risk. But can you give some easily
> reproducible examples or setups, code samples, cucumber scenarios,
> whatever, that could clearly demonstrate how systemd breaks anything?

As systemd tries to do so much it is pointless as you simply get side
tracked when your arguments succeed or ironically get called trolls by
real trolls.

I will just say this

OpenBSDs /sbin/init.c is 1448 lines long

Systemds rediculously placed /usr/lib/systemd/systemd.c doesn't exist
but rather there is a directory with many files and just cgroup.c is
nearly 1000 lines long. So good luck evaluating all of that for your
next linux based system when quite clearly the devs don't know even
pid1 like the back of their hand due to having a segfault rather than
die in it. pid1 being potentially larger than your daemon would be quite
laughable too.

As even well audited code tends to have a bug per 1000 lines, is
having this much code permanently resident supposedly for all of
Linux never mind unix ecosystems that have served each other so well a
good design especially when linux is moving onto smaller and smaller
devices? Are the benefits worth it? Is it self managing and open to
competition. It is difficult to get a good design from a bad design.
Evolving /sbin/init simple functionality in the past resulted in
migration of code into the kernel and not the other way around.

Such a fundamental game changer with so much clouding the core
functionality in terms of pulling in so much really shouldn't be decided
by a 50/50 decision. Companies often require 75% or more for game
changing decisions.

> scenario, why not just migrate to Gentoo or BSD?

Gentoo means you have to build otherwise I would for the few things I
need linux for. Sabayon has moved to systemd and my system would have
broken recently because of systemd if I was using it. I use OpenBSD
wherever possible.

Things are often tested on debian or *buntu and the base is well
tested and hence my usage for offline development.

-- 
_______________________________________________________________________

'Write programs that do one thing and do it well. Write programs to work
together. Write programs to handle text streams, because that is a
universal interface'

(Doug McIlroy)

In Other Words - Don't design like polkit or systemd
_______________________________________________________________________

I have no idea why RTFM is used so aggressively on LINUX mailing lists
because whilst 'apropos' is traditionally the most powerful command on
Unix-like systems it's 'modern' replacement 'apropos' on Linux is a tool
to help psychopaths learn to control their anger.

(Kevin Chadwick)

_______________________________________________________________________