[systemd-devel] I wonder… why systemd provokes this amount of polarity and resistance

[Tom Gundersen]

On Sun, Sep 21, 2014 at 3:31 PM, Martin Steigerwald <Martin at lichtvoll.de> wrote:
> I just have one question. In the light of
>
> http://boycottsystemd.org/

Please note that this is just (to the best of my knowledge), the
misinformed rants of an anonymous individual (despite it appearing a
lot more serious due to the guy buying a domain).

> http://uselessd.darknedgy.net/

Hm, missing content?

> Debian Bug report logs - #746715
> the foreseeable outcome of the TC vote on init systems
> https://bugs.debian.org/bug=746715

"this package no longer exists" ?

> Did you ever ask yourself why your project provokes that amount of resistance
> and polarity? Did you ever ask yourself whether this really is just resistance
> against anything new from people who just do not like "new" or whether it
> contains *valuable* and *important* feedback?

We occasinoally got some very critical feedback that is also very
good. However, when you look at the various "debates" in other forums
(such as the ones you mentioned), my impression is that it is almost
entirely useless noise. I.e., either people are complaining about
things that are based entirely on misconceptions, or they are
complaining about things that had a root in something reasonable once
upon a time, but have long since been either explained or fixed.

> I am taking this upstream with you, cause I think too much of this is
> resignately been discussed elsewhere, discussed elsewhere for as I got the
> feedback on various occasions where I recommended to take feedback upstream
> that people have no hope in having their feedback considered at all.

If people bring useful and at least moderately civil feedback upstream
(i.e., technical feedback in terms of bug reports, questions, RFE's or
similar, not rants amounting to "please stop what you are doing and go
away"), we do take it very seriously, and answer them to the best of
our abilities. That does not mean that every patch is accepted, nor
every request is adhered to, but at least you should get an answer
with an explanation.

The only way to find out is to try though.

> For now I use systemd. I like quite some features. But on the other hand I am
> vary about it myself. I look at a 45 KiB binary for /sbin/init as PID1 and a
> 1,3 MiB binary in systemd 215 and wonder myself.

systemd is a lot more powerful than sysvinit, and does take up more
space. We are not really optimizing the size of the binary, so if you
are interested in looking into making it smaller that's certainly
possible.

> I see systemd --user
> processes running and wonder: Why does the user related stuff need to be in the
> systemd binary.

This is rather the other way around: the problem solved by PID1, is
almost entirely the same as the problem that needs to be solved by the
user session manager, so we allow the same code to be reused. The
amount of code specific to user sessions in PID1 is really very small.

> I had it that it didn´t mount an NFS export and while in the
> end it was a syntax error in fstab that sysvinit happily ignored, I needed a
> bug report and dev help to even find that cause. I wonder about the complexity
> involved in one single large binary.

PID1 does not parse your fstab, it is done by an external binary. That
said, if there is a lack of output when a malformed entry is found, we
should probably improve on that (I don't have the bug report in front
of me, but please open an RFE if you think this is still not ideal).
The reason we reject invalid fstab entries, is that we think it is
safer to fail hard if we cannot know for certain what the admin
intended to do. It is regrettable that this means that the transition
is not as smooth as it otherwise would be, but overall we believe it
is the correct thing to do.

Cheers,

Tom