On Wed, 9 Aug 2017 10:16:46 +0200
Narcis Garcia <informat...@actiu.net> wrote:
> El 09/08/17 a les 09:13, Edward Bartolo ha escrit:
> > Using this 'resource' will do a disservice to Devuan. Anyone serious
> > enough to read it will get the wrong impression that Devuan is some
> > 'amateur' distribution not worthy of wasting professional hours on
> > it. Scientific and technical text must at all costs avoid
> > opinionated writing but this resource does the opposite. As said
> > earlier, there is no objective comparism between the different
> > inits.
> If reaching perfection is a requirement, important things will not be
> Imagine Devuan 1.0 without still being released because of Gnome lack.
I agree with you and Edward, and furthermore refuse to see connection
between PID1 and the cgroups feature as something necessarily
desireable. It's not like before systemd, I felt there was something
wrong with any of my systems that could have been fixed by such a
As Edward and Narcis both say, perfection can't be reached. I take one
step farther, and say that not only is a PID1/cgroups connection not
perfection, but its utility is helpful only in edge cases.
And I agree with Edward that the mere discussion of cgroups bends the
discussion in a systemd kind of way, because it's always discussed as a
necessity or at least a good thing.
Cgroups is for restricting/monitoring resources for groups of
processes. Very necessary for Google and perhaps for containers, but
not for PID1s. It's been at least 10 years since an out of control
resource-hog process locked me out of my computer in such a way I was
forced to reboot, and I've never used systemd. On a dare I once ran a C
no output, no sleep forever loop at nice -19, and I was still able to
step in and take control.
The way things are today, the mere discussion of cgroups as a
desireable thing in an init horribly distorts the situation. Like
Edward and Narcis said.
July 2017 featured book: Quit Joblessness: Start Your Own Business
Dng mailing list