Re: Bug#727708: Fsck SystemD and its developers and its users. GR to override this please.
> On Feb 11, 2014, at 5:02 PM, Thomas Goirand <zigo@debian.org> wrote:
>
>> On 02/11/2014 08:13 PM, John Paul Adrian Glaubitz wrote:
>> Yes, but we are not talking about hypothetical things. I am also not
>> planning my life for the case that I am winning the lottery tomorrow.
>
> Chances to win the lottery are 1 against 14 000 000 (at least in my
> country). Claiming this kind of odds for supporting OpenRC is IMO an
> overstatement, especially considering that we have LSB header scripts
> for *all* of our packages right now.
That was not my point. I was not talking about the functionality. I don't have any doubts that OpenRC is doing what it was designed for already.
However, I do not think that the current feature set of OpenRC is enough to justify preferring it over systemd and, now coming back to my lottery metapher, I do not think that it is very likely that OpenRC is going to be able to catch up any time soon.
Don't get me wrong, I really appreciate what you and the other maintainers are working on. But I think that it's not leading anywhere.
The only advantage I have with OpenRC is its portability but that doesn't justify to waive for important systemd features that users need in a productive environment.
People need resource management and process isolation, process tracking, CPU affinity settings, consistent and easily searchable logs, reliable daemon startup and dependencies without having to resort to "sleep".
systemd actually solves lots of real world problems that everyone who gets a salary for administrating a large Linux server setup has heard of before.
Arguing that portability to Hurd or BSD is more important than reliability and robustness in a productive environment is just crazy, sorry!
Adrian
Reply to: