Return to BSD News archive
Xref: sserve comp.os.linux.development:23302 comp.os.386bsd.development:3154
Newsgroups: comp.os.linux.development,comp.os.386bsd.development
Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!bunyip.cc.uq.oz.au!munnari.oz.au!news.hawaii.edu!ames!olivea!koriel!cs.utexas.edu!news.cs.utah.edu!news.provo.novell.com!park.uvsc.edu!news
From: Terry Lambert <terry@cs.weber.edu>
Subject: Re: SAMBA and NETWARE mounting
Organization: Utah Valley State College, Orem, Utah
Date: Tue, 31 Jan 1995 18:00:59 GMT
Message-ID: <D3A7Dp.1DI@park.uvsc.edu>
X-Nntp-Posting-Host: hecate.artisoft.com
References: <D2KG6E.CMp@park.uvsc.edu> <D32vvM.5ro@park.uvsc.edu> <1995Jan28.205413.3955@rhrk.uni-kl.de> <D38EGq.5yB@park.uvsc.edu> <3gjjte$4r9@usenet.srv.cis.pitt.edu>
Sender: news@park.uvsc.edu (System Account)
Lines: 85
ddj+@pitt.edu (Doug Dejulio) wrote:
] In article <D38EGq.5yB@park.uvsc.edu>,
] Terry Lambert <terry@cs.weber.edu> wrote:
] >The "problems" of IP subnetting are frequently manufactured. Use
] >a Macintosh and see what I mean.
]
] Huh?
]
] Let's consider the case of a public university that provides PPP
] connections for its faculty, staff and students. Let's say there are
] about 5,000 people that make use of this facility at least once a
] month.
OK.
] How likely do you think it is that the university would be willing to
] provide a distinct, non-dynamic 3-bit (i.e. six host) subnet for each
] of them? Or even a 2-bit (i.e. two host) subnet? Hell, they often
] won't even consider a single static IP address, because there aren't
] enough to give out to everyone who might want one. So, we're stuck
] with a different IP address every time we connect, which is just an
] utter lose.
It's not bloody likely.
By the same token, the services which make a changing IP address
an "utter lose" ar completely unavailable under IPX.
What's needed in this case is either the next rev of IP, which
has a *vastly* larger address space than anything out there now,
*OR* some form of authenticated proxy routing.
The proximal problem is that you are trying to run a host machine
on a dialup IP line without a dedicated address, and dialup IP
lines without dedicated addresses are for client machines only.
As are the third-party-only dialup IPX mechanisms.
Until nomadic computing becomes a bit more advanced, you're pretty
much stuck with this scenario.
And no, I'm not saying that the new IP will fix it, since no one
seems to be considering connection point independing pathing or
service anonymity at all -- which really requires a change in what
we think a connection is in the first place, and starts getting
radically different after that.
] If the local IP namespace were much, much larger, one would have at
] least a snowball's chance in hell of talking folks into doing things
] differently. Doubling the number of bits would be nice.
Use the new IP. You will be using the new IP anyway, probably
within two years.
Like I said, however, unless you limit yourself to a single access
point, as a consumer of nomadic computing, you're pretty much
screwed anyway with almost all commercially available technology.
] >The problems you have are largely the result of the choices you make
] >about the software you will use.
]
] Great! My university will only grant me a temporary, dynamic one-host
] IP address when I connect via PPP or SLIP. I have need for my own
] static 3-bit subnet. What software do I run to solve this problem?
] (Tell me to run an isolated network behind a firewall with proxy
] servers and I'll bite your ankle.)
OK, instead of telling you that, how about buying a point of
presence instead of trying to use the universities?
What you want requires a static address assigned to you, and since
routing is the way it is, requires the ability to cause the machine
you connect to to proxy-arp your gateway's address when the circuit
is up.
As far as a dialup NCP/IPX connection into a network, it's impossible
to adequately firewall. You're just asking for trouble if you, as
the university administrator, allow it.
Terry Lambert
terry@cs.weber.edu
---
Any opinions in this posting are my own and not those of my present
or previous employers.