i hate systemd
The Good, The Bad, The Ugly.
And The Survival Guide.
(Work In Progress)
(Work In Progress)
Why does this website exist? What is it for? What (if anything) does it aim to achieve?
Because I created it. I bought a domain name, stole some CSS from people who are smarter than me at that kind of thing, and put some words together.
Beyond that? I don't really know at this moment.
Quite possibly. systemd is a terrible thing, in many ways worse than what it has replaced, and in very few ways is it any better. That, for sure, is a statement that has been made, not only every generation, but indeed every few years, in almost every industry.
It is true that such sentiments are often expressed by those who fear change, who are insecure in their own abilities
to keep with the times. Then again, it is also pretty much what a Ferrari Formula One mechanic or NASA engineer would say
if "promoted" to work on 20-year old Škodas. So - go figure.
I don't think this website is particularly aggressive. It's just relaxed. Most IT documentation, whether written for
public consumption, such as official documentation, textbooks, even blogs, tend to assume that the reader and the author
both believe that the tool being discussed is at least generally beneficial and a positive, useful thing to be working with.
I'm really not certain that that is the case with systemd. It has some terrible flaws, both technically and in the way in
which the developers interact with the community. It does not strike me as helpful to ignore that elephant in the room,
and to timidly try to write up, as best I can, how to get the best out of systemd, without also acknowledging these flaws.
It does the reader a disservice, and makes a liar of the author. It strikes me that a more lighthearted,
honest, maybe even playful and at times disrespectful approach works better for me, and therefore maybe for others. If the
Emperor truly has no clothes, then somebody has to be prepared to stand up and say so!
True, the "i hate systemd" moniker could be construed as, well, hateful. But genuinely, it is the most common reaction I have observed, whether face-to-face or online, to this highly disruptive software which (for better or worse), it is no longer possible to ignore.
Seriously: I want to get to know systemd better. It's part of my toolkit, so I need to know it. And it's not all bad, not at
all. But it's not easy to get to know and to live with, and I know that I am not the only one who feels this way.
The documentation from the developers is pretty much completely useless for the day-to-day sysadmin, DevOps or developer
wanting to achieve some specific task. Some of my more formal writing has apparently been useful to a
great many people; one online tutorial I wrote and maintain gets thousands of readers every day, day-in, day-out. So it may
be that the information I collate here on solving real-world problems with systems (when to use oneshot,
simple, forking, what the trade-offs are with each type, and maybe even how and when to use the
more obscure types) will be helpful to others.
Maybe, even, a community might come together to share war-stories and best practices. Because the clique around systemd
do not seem to be interested in such grass-roots efforts, and any such effort must, of necessity, be open and honest about
the many flaws, foibles, inconsistencies and just stupid design decisions in some parts of systemd, such a thing will not
spring up around the Fedora or FreeDesktop documentation of systemd.
Why ever not? I don't know how, yet. Wiki, Github, Christmas Newsletter: The available options are almost infinite.
There is a website called No systemd which has some further information about some of the shortcomings of Systemd.
A(n)