Steve Litt <slitt@???> wrote:

> My opinion: If the init specification of a daemon exceeds 25 lines,
> that's a problem. Many sysvinit and OpenRC daemon init specifications
> are over 100 lines, especially if you take into account all the stuff
> imported from the "functions" file. I never want one of those long
> files darkening my door again.

Playing devils advocate ...

What's the difference between importing a few functions and using them, vs embedding that functionality in some C code ?

That seems to me to be one of the big differences between SysVinit and SystemD. In a SysV init script you have a few lines which setup the requirements and call a function to start the daemon. In the other one, you have a few lines that setup the requirements, and then hand over to some black box.

The problem with SysV init scripts isn't what people are making it out. As has been pointed out many times, a lot of that bloat isn't there because SysV needs it - it's there because ... well "it's there" ! They really can be quite short and compact - once you weed out the stuff like LSB metacode to control which runlevels it runs in, prettifying output, and stuff like that.

But what is needed, after you've taken the stuff out that really isn't inherent in SysV, is IMO better than passing a few arguments to some black box binary that can't be debugged. As someone else pointed out, with a shell script you've got somewhere to stick your voltmeter.