Subject: Re: Re: Summary of /etc/rc*
To: <>
From: None <>
List: amiga
Date: 09/13/1994 17:43:56
Hi folks;  just my two-pennorth, not intended as a flame or slight on anyone's
opinions in any way <grin>

 > from the clients), I fail to see why the current state of affairs should
 > require _one_mammoth_script_ when individual scripts can be intr'd from
 > rc.local as is.

I just wanted to relay my experiences;  I am not mangler of a vast multi-machine
system running huge numbers of daemons, just a single-machiner...  But I have
been involved in writing an introductory document/FAQ to detail how people at
my university can connect to the college computers via SLIP & a phone.

The document covers NetBSD, Linux and AmiTCP users.  As a NetBSD'er, the Linux
coverage was my first real encounter with fiddling with run levels and the like.
Initially I kinda figured it was a kludge designed to make things complicated
<grin>, but experience tells me it can me very handy, particularly when dealing
with beginners (and lets face it, a lot of _any_ free package's users are going
to be beginners dabbling because theres nothing to lose).

Particularly, "automating" the network installation is a deal less scary in the
Linux system if something goes wrong;  because the networking changes etc. can
all go in the MultiUser runlevel scripts, a cockup is pretty simple to fix
just by booting into single-user mode & fiddling a bit :*)  This means a user
with minimal competence has to make a real effort to make the system completely
unusable - a real boon :*).

 >  Or, alternately, you can leave a single flag and "if" statement at the 
 > beginning of /etc/rc.local to boot whatever daemons you please, without
 > any changes to the current implementation.

Surely tho, that is a local change to the implementation?  If you don't
disapprove of this, surely if the multiple-run-level init patches were 
distributed as an optional extra, would that be acceptable?

The only problem with this, of course, is that install scripts provided with
additional packages would assume a standard setup I guess /-|.  Hmmm.
Still, thats the nature of compromise I 'spose <grin>

 > in my opinion, not nearly enough usage to warrant the changes to
 > init, changes to existing startup scripts, and more importantly drift 
 > from the rest of m68k.

Does it have to result in a drift?  I (personally) see no reason why the
changes couldn't be fed back into the general tree if they were regarded as

 > Charles Ewen MacMillan  | Tezcat.COM - Wicker Park 

 __ //||/  Tim Walls/TJay            EMail       
 /_//=|/  My opinions are my own    PGP2.3 Key, finger
  `And with the last line almost drawn, wish you goodbye 'till further on'