Linux: concerns over systemd adoption and Debian's decision to switch

Daniel Corbe corbe at corbe.net
Wed Oct 22 17:35:26 UTC 2014


Andrew Sullivan <asullivan at dyn.com> writes:

> On Wed, Oct 22, 2014 at 12:43:53PM -0400, C. Jon Larsen wrote:
>
>> Incorrect assumption. systemd is a massive security hole waiting to happen
>> and it does not follow the unix philosophy of done 1 thing and do it
>> well/correct. 
>
> But I have no clue what one can do about it.  For many years, I liked
> to keep some Linux and some BSD systems around, because it seemed to
> me that the different styles tended to encourage diversity and that
> was a good thing.  But management of BSD systems -- particularly the
> nonsense of rebuilding things from source all the time -- started to
> look mighty onerous compared to apt-get update; apt-get upgrade.
> Others apparently agreed, and now there are enough things that work
> well on Linux but not as well (or not at all) on BSD that the
> diversity argument isn't as strong.  (Also, of course, certain kinds
> of things, like some kinds of database replication, don't work well
> across platforms, so there's another reason to converge on a single
> system.)  Debian was always the Linux platform that seemed most
> insistent on having more than one way to do it, but in recent years
> that philosophy has made it more work to use than the alternatives;
> and the alternatives have often gotten good enough that one doesn't
> care (Ubuntu is the obvious example here).
>
> So, now we have an encroaching monoculture, and no real option to do
> anything about it.  Maybe this is just the way the Internet is, now.
>
> A

Not to get even further off topic here but when was the last time you
maintained a BSD system?   FreeBSD (at least) adopted binary package
management as its preferred interface to ports through pkg-ng somewhere
in the 9-RELEASE cycle.  

As long as you don't need exotic compile-time options you should be good
to go.  Which is in contrast to the Linux package management paradigm
where you basically enable everything at compile time.  

If you do need to compile something by source though you still have that
option.  

This systemd debacle is an excellent reason to look into stuff that
isn't Linux.  The Linux camp all too often become victims of "not
invented here" and "because we can" is not a good enough reason to
replace something that has worked just fine for 30 or 40 some-odd years.

  



More information about the NANOG mailing list