Still an order of magnitude more complex than systemd.

Then what are we discussing? Use your old style sysvinit.

I don't know again what you mean. The sysvinit init scripts are still working.

Because it doesn't suppose to do that. The journald is used internally by systemd, and services can opt-in to use it. It's also a situation of damn if you do, damn if you don't do. How many times I red on Slashdot the complain that with systemd you must have binary logs, or that it takes over logging.

In Roundcube for example you can opt-in into syslog logging. http://trac.roundcube.net/tick... The same options should be there to enable journald logging. At least systemd captures sysout and syserr into the journal, so it's not lost. Another huge improvement over sysvinit. How many times I had to start a daemon manually so I can see the errors from syserr.

But again, the MySQL logs are still there in /var/log ? It's not systemd fault if Oracle is not updating MySQL to use journald.

In science, it's irrelevant if you admit it or not. Either your model works or not, and if it's not working somebody will come up with a better model and replace yours. People will use the better model because it works better. Or maybe I misunderstand you. Newton's laws of gravity works "almost right", Einstein's theory is better. But in cosmonautics and mechanics we use Newton's laws, because they are good enough.

If we can't observe anything new. And by "new" I mean something that was not predicted by a scientific theory. Of course, there is always the possibility to observe something new, but it's like an asymptote. New scientists will be less and less excited, because they will just confirm established theories.

Aw, boring. I was hoping that everyone was wrong and we get some new physics. Misconception of scientists number one, scientists (and me) like to be shown wrong so we can go and investigate and discover new knowledge. The day it turns out that we know everything will be a very sad day indeed.

"What I want is quite simple. To be able to configure and manage stuff with the least hassle."

Then you want systemd. As I mentioned before, the sysvinit start scripts are huge, complicated and error prone (the loop in Apache init script). In systemd all you have is a 17 lines file for Nginx: https://projects.archlinux.org...

systemd adds a lot of extras, and does not take anything away. It adds, a) info on whether the service was started, b) automatic restart of crashed services, c) journald, d) limit resource usage per service, e) simple and powerful service configuration files, f) inter service dependencies management (start Thin only if Nginx is started, or start MySQL after Apache) g) a lot of tools that were missing like systemd-ask-password.

And systemd does not take anything away from you. You can still use all your old school tools and your old style logging, and even the old sysvinit scripts.

"but no standard tool will help you when some opaque software is capturing the initialization process."

I don't know what you mean by that.

"Do you know which flags are passed to eg. your nginx setup? A quick look at the rc.d script will enighten you."

Same with systemd service files, see my example above.

"Is there a bug on the script (it happens more often than you seem to think)? Easy to fix."

In a 9000 bytes script file? You gotta be kidding me. The systemd file is just 17 lines long. The systemd developers have automated tests to make sure their software runs, if you don't trust them, why do you trust the kernel devs or the devs of the services you are using?

"but if not, why the f*ck is it taking control over MySQL initialization? "

What do you mean by that? It starts the service and watches the service that it is really started, and it gives you info that it is started or it didn't started or it crashed. For example, on my laptop:

But you have both, MySQL log files and systemd logs? So, I still don't get what your problem is. In sysvinit you have zero information. And ps or netstat is not a reliable solution, because it is always different from service to service and from system to system. In ps you have to know the executable, in netstat the name of the service. On my Ubuntu server, for postfix you have "master", for nginx you have "nginx.conf". And of course the name will differ from system to system. Furthermore, services can opt in into the logging of systemd, it is not mandatory. So, I'm really confused what you want. You want that systemd takes over logging, or you still want your old logging files?

What does it matter? If Gnome and systemd are both their developments, good for them.

On systemd I do "systemctl start foo" and I get info that the service was actually started and runs, or that the service did not started. With the init scripts I don't have any response and I need to use ps or netstat to figure out whether or not the service actually runs. That is because systemd is using cgroups to track the started service, but sysvinit have no service management at all.

If you want info and logs why the service didn't started, use "systemctl status foo", or use the journal with "journalctl -n 20" or filter "journalctl _SYSTEMD_UNIT=foo". That's no different from the usual "dmegs|grep foo".

Your logging issues are too unspecific and I have no idea what you actually talking about. On Fedora 20 I have all the usual logging files in /var/log. Even your missed /var/log/messages is there.

No, it will not. It just shows a yellow triangle. If you want a security policy that does what you describe, then Chrome should not display that web site at all.

No, it does not. The Window Manager have dependency on systemd, not the other way around. If the Gnome developers deciding that they need systemd, it's not systemd fault. You are free to open a bug ticket, but in the end it's the decision of the Gnome developers to use a particular technology.

That's why people want systemd. If I look at sysvinit scripts, there are a ton of magic stuff going on that I don't know, and maybe nobody knows. On Ubuntu 12.04 the sysvinit script for Apache is over 7000 bytes long, and it includes bad stuff like "# wait until really stopped" with a loop of kill and sleep. The ClamAV init script is even bigger, over 9000 byes long. Is there any maintainer who really knows what it does?

All they did compare was one memory access vs. another memory access, and showed that Strings are inefficient compared to a byte array. Because the first code will concatenate Strings in memory, the second code will concatenate byte array data in memory, and then both are written to disk. The disk-access test should have been:

    for (int i=0; i numIter; i++) {
                    writer.write(addString);
                    writer.flush();
                    writer.close();
    }

I have the impression that it's their core business. My point was that Red Hat is publicly traded and thus they are much more transparent than private businesses and the value of their stock reflects the performance.

That is your opinion. I wrote reasons why I think it is proven and obviously a lot of Linux distributors think that, too. Lets assume you are correct and it is unproven and will break in RHEL 7 and a lot, if not all, customers will get very angry with Red Hat and switch to sysvinit or stick to 6 or switch to a different distribution or to Windows. Do you think Red Hat will live with that risk? So, why would Red Hat use systemd in their enterprise distribution if they not think it is ready? Red Hat, Inc. is even a publicly traded company and since May 2014 their stock continues to grow in value, i.e. there is no sign that customers are running away from RHEL 7.

You are free to use RHEL 6 or any other distribution that don't use systemd if you think it is unproven. You can even deinstall systemd on RHEL 7 and continue with sysvinit. PS: nice no true Scotsman fallacy.