Re: this is an unimplemented feature of update-inetd

[Peter Makholm <peter@makholm.net>]

Josip Rodin <jrodin@public.srce.hr> writes:

> Peter Makholm (brother) already did some work on it, but I haven't
> seen anything else... and we have a big fat warning in current
> update-inetd implemented.

What I did was a drop-in replacement of the present perlscrip behind
update-inetd. It can serve as an temporary solution to the problem but
I don't think it's a nice solution and freeze was just about that time
(first try).

I had some thought about a better solution but they would require some
more rewriting of the present scripts and I didn't really have the
time before the freeze (which never happened).


The design of a future update-inetd would be something like the
following:

  Any package providing inetd-like features implements a module to
  make a configuration file based on some common data.

  Any package providing serveres to be run through inetd register with
  update-inetd (just like now).

  update-inetd constructs with help of the modules conf-files for all
  inetd-like packages.

(Nothing exceptionel here, and I could probally look at the menu
things to se how that works)


The question is, should servers have the posibility to use non-vanilla
inetd features. That is, should update-inetd has the posibillity to
set any other flags than those used by the normal inetd?


-- 
Spinach, carrot, and a melon -- a meal fit for a nurse!