Re: [SAGE] System Naming Conventions

[Brad Knowles <brad@stop.mail-abuse.org>]

At 10:24 PM -0800 2005-03-09, Neil Waybright wrote:

>  I am looking for formal studies of system naming conventions.

	We recently had a big "discussion" of this issue on another 
SA-related mailing list that I'm on.  Most of the best advice seems 
to already be in RFC 1178 (which I see you've already found).  Most 
of what is not in RFC 1178 seems to relate to the scaling of naming 
conventions.

>                                                                 I read
>  with a great deal of interest the thread on the subject on 31 March
>  2004
>  (<http://www.sage.org/mailarchive/sage-members-archive/2004/msg00803.html>
>  was part of the series).

	In this thread, the messages I found most useful are 
<http://www.sage.org/mailarchive/sage-members-archive/2004/msg00815.html>, 
<http://www.sage.org/mailarchive/sage-members-archive/2004/msg00821.html>, 
<http://www.sage.org/mailarchive/sage-members-archive/2004/msg00822.html>, 
<http://www.sage.org/mailarchive/sage-members-archive/2004/msg00826.html>, 
and 
<http://www.sage.org/mailarchive/sage-members-archive/2004/msg00853.html>.

	Yes, the last one is mine, but I still think it's a good point.

>  [3] <http://itmanager.blogs.com/notes/2004/03/system_naming_c.html>

	Alex Scoble does have some good points.

>  [4] <http://www.sage.org/mailarchive/sage-members-archive/1998/msg00369.html>

	In this thread, the three most useful messages I found were 
<http://www.sage.org/mailarchive/sage-members-archive/1998/msg00376.html>, 
<http://www.sage.org/mailarchive/sage-members-archive/1998/msg00395.html>, 
and 
<http://www.sage.org/mailarchive/sage-members-archive/1998/msg00391.html>.

	In particular, I like the advice to pick your battles, and to set 
broad naming scheme guidelines and then let the various sub-groups 
pick their own specific naming schemes within those.  That's one way 
to get around the scaling issues -- IME, most "cute" naming schemes 
start running out of steam once you've gotten beyond a dozen to a few 
dozen machines, but if each department had their own "cute" naming 
scheme, the overall aggregate could still potentially scale to a 
thousand machines or more.

	One advantage of using descriptive or meaningless names is if the 
machines can be re-installed when they are moved.  Indeed, I would 
suggest that this is a requirement if you want to have any kind of 
truly scalable naming convention.  We did this a lot at AOL.


	I guess my most important piece of advice for a normal-size 
business, academic environment, etc... would be to use both types of 
naming conventions -- the SAs are likely to be able to deal with 
descriptive names best, while users are going to be more likely to be 
able to deal better with "cute" names.  So, use both -- via aliases 
in the DNS.

	The only place where you wouldn't need to use both types of names 
would be if you had a situation where you had thousands or tens of 
thousands of machines, all given descriptive or meaningless names, 
probably organized by subdomain.  In other words, an 
AOL/Yahoo/Hotmail/Google-like situation.


	For normal-size groups, I would drive the DNS alias 
recommendation one step further -- you should name the services you 
provide with aliases, and then have those aliases resolve to the 
machine(s) which perform those services.

	Even if you only have one machine today that provides e-mail 
services, having separate service aliases for relay.example.com 
(outbound mail from clients), pop.example.com (mailbox access), 
imap.example.com (other mailbox access), smtp.example.com (inbound 
mail), etc... will make your life much, much easier when it comes 
time to split those services across multiple machines.



	But if you're looking for a more formal treatment of this topic, 
I doubt that you're going to get it.

	Maybe this would be a good subject for a SAGE Short Topics 
booklet?  I think Neil just volunteered to write it....  ;)

-- 
Brad Knowles, <brad@stop.mail-abuse.org>

"Those who would give up essential Liberty, to purchase a little
temporary Safety, deserve neither Liberty nor Safety."

     -- Benjamin Franklin (1706-1790), reply of the Pennsylvania
     Assembly to the Governor, November 11, 1755

   SAGE member since 1995.  See <http://www.sage.org/> for more info.