[28978] in North American Network Operators' Group
Re: pop server in an ISP environment
daemon@ATHENA.MIT.EDU (Adrian Chadd)
Tue May 30 03:48:45 2000
Date: Tue, 30 May 2000 15:46:11 +0800
From: Adrian Chadd <adrian@creative.net.au>
To: "Roeland Meyer (E-mail)" <rmeyer@mhsc.com>
Cc: nanog@merit.edu
Message-ID: <20000530154610.D79030@ewok.creative.net.au>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <004d01bfca08$da760520$eaaf6cc7@PEREGRIN>; from rmeyer@mhsc.com on Tue, May 30, 2000 at 12:29:55AM -0700
Errors-To: owner-nanog-outgoing@merit.edu
On Tue, May 30, 2000, Roeland Meyer (E-mail) wrote:
>
> Angels and pin-heads ...
>
> Y'all both missed the point. Although Greg's dissertation on
> uuids was interesting, the salient fact is that you run out of
> machine way before you run out of uuid, in most practical cases.
> Who uses NIS these days? It's taboo around here and NFS is only
> allowed for Unix server-to-server file shares, even those get
> reviewed, on a must-need basis. The point of my message was to
> answer the question, not to debate relative kernel
> merits/methods. Who cares how many uuids a given kernel can
> support as long is the number is larger than the silicon can
> support? So what is my original numbers were a little dated, they
> were still irrelevent in determining the number of users a system
> can support. My point was that the planned load is the sole
> determinant. The original poster was asking a system capacity
> question, not a religious question.
>
> <sheesh> You'd think I was in DOMAIN-POLICY.
Erm, this dicussion has been rather entertaining .. what are userids?
They aren't "mail ids". Although the original poster IIRC didn't have
source, all this UID stuff is irrelevant because its quite possible to
build a mail server using only one UID, mapping to 'mail. (the same
goes for http if you're not doing CGI, and although I haven't done it
before I'm sure with some magic CGI could be run under one userid
securely..)
I agree with the above mail. The planned load is the sole determinant.
Everything else can be fudged.
Adrian
--
Adrian Chadd Build a man a fire, and he's warm for the
<adrian@creative.net.au> rest of the evening. Set a man on fire and
he's warm for the rest of his life.
