Any other way for non-static IPs?

> You need to have a static IP _before_ the dhcp sever starts.
> otherwise it won't be able to serve any IP.

Why is it not possible to say:
- OK, I have eth0. It needs an IP.
- I have a network definition 192.168.0 (netmask 255.255.255.0).
- I can assign from a region 192.168.0.10 - 192.168.0.100.
- OK, let's set it's IP to 192.168.0.44, for example.
- Now, this eth0 has an IP and it's a broadcast network adapter.
- As a DHCP server, I need some broadcast adapter to work on and eth0
is one of them.

The problem I want to address is: why not define only the network (by
netmask) and the range of addresses that machines can get. A DHCP
server should respond to any ethernet (i.e. broadcast) adapter it
finds (or the adapter you give by a command line option). If it sees
that eth doesn't have an IP address, no matter. Since it belongs to
some network, we can configure automatically it by knowing the netmask
and the range of addresses we have. Is this possible? In theory yes,
but I don't know if in practice and if it was implemeted in current
DHCP implemtations.

If not by DHCP, is it possible to be done by any other way? The
requirement is: I define (or get a definition by an upper authority)
of a network (network IP and netmask) and I start a DHCP or whatever.
That computer has ethernet eth0 which is assigned the given network IP
(but NOT host IP!), it automatically assigns one from the given range
of IPs and starts a server there.

Any hints?

Thanks!

On Mon, 25 Oct 2004 05:30:07 -0700, Iceman wrote:

>> You need to have a static IP _before_ the dhcp sever starts.
>> otherwise it won't be able to serve any IP.
>
> Why is it not possible to say:
> - OK, I have eth0. It needs an IP.
> - I have a network definition 192.168.0 (netmask 255.255.255.0).
> - I can assign from a region 192.168.0.10 - 192.168.0.100.
> - OK, let's set it's IP to 192.168.0.44, for example.
> - Now, this eth0 has an IP and it's a broadcast network adapter.
> - As a DHCP server, I need some broadcast adapter to work on and eth0
> is one of them.

You can... Create the scripts to asign static random IP addresses.
Create scripts to test if the IP address can be used on your network. Give
this IP address to your DHCP-server and start it...

> The problem I want to address is: why not define only the network (by
> netmask) and the range of addresses that machines can get. A DHCP server
> should respond to any ethernet (i.e. broadcast) adapter it finds (or the
> adapter you give by a command line option). If it sees that eth doesn't
> have an IP address, no matter. Since it belongs to some network, we can
> configure automatically it by knowing the netmask and the range of
> addresses we have. Is this possible? In theory yes, but I don't know if
> in practice and if it was implemeted in current DHCP implemtations.

This is a proposed solution to a problem... What are you trying to do...?
What's the problem?

> If not by DHCP, is it possible to be done by any other way? The
> requirement is: I define (or get a definition by an upper authority) of
> a network (network IP and netmask) and I start a DHCP or whatever. That
> computer has ethernet eth0 which is assigned the given network IP (but
> NOT host IP!), it automatically assigns one from the given range of IPs
> and starts a server there.

If running 2 NICs you can assign a static IP for one of them. If your
DHCP-server is running "on" this NIC, the other NIC, and other computers
can use this server to get it's IP address.

You'll have to tweak your init scripts first, but it will work.

AFAIK; no DHCP-server can serve and assign IP addresses without using a
working NIC. A working NIC needs an IP address.

Anyway: What are you trying to accomplish and why do you insist on
going in this direction?

--
mvh/regards
Joachim Mæland

If everything seems under control, you're just not going fast enough.
-Mario Andretti

> Anyway: What are you trying to accomplish and why do you insist on
> going in this direction?

I apologize for making such a big deal about a small thing, but it's
more a style and clarity question.

* Problem setup
You have 100 machines (Linux, Windows, MacOS, whatever...). They all
have ONE ethernet adapter. Let's concetrate on 5 Linux machines, which
have their respective eth0 adapters. I will say "some server" instead
of DHCP server here, because this might be a question where answer
could be Dynamic DNS or any other type of name service. Assume that
all 5 Linux machines have that server (let's call it NS - name server)
up and running, one is a master, four are slaves and that any of them
can become master if current master fails (thunderbolt, whatever...).

When any host (even one of the 5 Linuxes running the server) turns on,
it tries to get it's IP. The only unique is their ethernet MAC
address. So it sends MAC to FF:FF:FF:FF:FF:FF broadcast. There is
currently only one master
NS server, which responts. This is something like the DHCP's
DISCOVER/OFFER/REQUEST/ACK protocol. OK, so now it has it's address
and it can work OK.

However, if there was no NS server available, why not configure
itself? Since you are yourself a NS, you are able to do it and, in the
same time, become a master NS for that network.

A note: this is the simmilar to TokenRing networks, where there is a
protocol for defining the master computer for the network. Of course,
this is just to define master for the ethernet (ISO OSI 2) layer, not
IP (ISO OSI 3).

* Pros
- No static addresses for NS server configurations. Administrators
wouldn't need to configure anything, but a network IP, netmask and
range of IP addresses. They could say: you are a possible NS (there
may be other NS servers) for the 192.168.0.0 network with
255.255.255.0 netmask, you have a range 192.168.0.10 - 192.168.0.100
to lease.
- If eth0 is given such network IP, netmask and range, NS which is
joining on a network without a master could say: OK, no master, so
noone can give me the IP. Since I am a potential NS, I will become a
master on that NETWORK (so I still don't have an IP ADDRESS for my
eth0) and I will issue a client request ON THAT NETWORK again. Since
now there is a master now (itself, but why not?), it will issue an IP
for the client (itself again) and it will get a valid IP.
- No static NS addresses for clients. Clients would only join a
network and look for a NS server. No static DNS, WINS IPs or whatever
- just broadcast and you will get it from the current master NS. Many
NS possible, at any available IP on that network as they wish.

* Cons
- Some security issues may be at stake here, but that's not what I am
"into" right now. This is a potentially big thing in many network, so
a "contra" point here.
- May be slow.

* The real problem
Back to my apology, I have only 2 machines (OK, it's a big story for
such a "huge" network, sorry), one has WinXP/Linux (A) on it and the
other WinXP (B). Static IP works just fine. I wanted to make it
dynamic. If I start Linux (A), I must make it static IP in order for
it to be a DHCP server, right? So what is so dynamic about this? If I
join the other WinXP (B), it will get a dynamic IP. 50% is not
enogugh.

But a better example: assume I had two Linux machines, both of which
could be DHCP servers. How to do it? Since I need a DHCP server to be
on a static IP, both would be on static addresses. 0% is nothing.

* Questions (hopefully clear now)
Is there a way to name 4 machines Dragon, Torpedo, Mushroom and
Turtle, give their network IP, netmask and range of addresses on that
network they can occupy and the system does the rest? I don't want to
care about their IPs, but I want to control them enough. If an upper
authority gave me a network range, I must obey to that, but nothing
else. I have network from "here" to "there", it's mine and I want a
way to fit my 4 computers (and maybe afterwards add up to 60 laptops,
network printers, etc.) without caring who, when, why, ... assigns
IPs. I want simple server configuration. I want to be able to simply
rip of the above 4 machines (all NS severs) and add 15 NS servers
more. I don't want to configure them again by assigning 15 static IPs
- that is boring, error-prone and certainly not DYNAMIC.

* More questions
- Is my thinking bad or good? I.e. do you agree in common?
- Is there anything working in practice like this? Is Dynamic DNS what
I am looking for? Can DHCP do the above?
- If not, any ideas why not - what are the problems for it not being
like this?
- How do you manage (if you do) large networks? Static DNS is also
very demanding, as I see.

* That's all, folks!
Thanks again for your time and answers!
Iceman

P.S.: If you read all this, then you:
a) have a lot of time to waste,
b) are very interested for the topic,
c) are a very good person,
d) (... I surely missed something here, no? ...)

> Anyway: What are you trying to accomplish and why do you insist on
> going in this direction?

I apologize for making such a big deal about a small thing, but it's
more a style and clarity question.

* Problem setup
You have 100 machines (Linux, Windows, MacOS, whatever...). They all
have ONE ethernet adapter. Let's concetrate on 5 Linux machines, which
have their respective eth0 adapters. I will say "some server" instead
of DHCP server here, because this might be a question where answer
could be Dynamic DNS or any other type of name service. Assume that
all 5 Linux machines have that server (let's call it NS - name server)
up and running, one is a master, four are slaves and that any of them
can become master if current master fails (thunderbolt, whatever...).

When any host (even one of the 5 Linuxes running the server) turns on,
it tries to get it's IP. The only unique is their ethernet MAC
address. So it sends MAC to FF:FF:FF:FF:FF:FF broadcast. There is
currently only one master
NS server, which responts. This is something like the DHCP's
DISCOVER/OFFER/REQUEST/ACK protocol. OK, so now it has it's address
and it can work OK.

However, if there was no NS server available, why not configure
itself? Since you are yourself a NS, you are able to do it and, in the
same time, become a master NS for that network.

A note: this is the simmilar to TokenRing networks, where there is a
protocol for defining the master computer for the network. Of course,
this is just to define master for the ethernet (ISO OSI 2) layer, not
IP (ISO OSI 3).

* Pros
- No static addresses for NS server configurations. Administrators
wouldn't need to configure anything, but a network IP, netmask and
range of IP addresses. They could say: you are a possible NS (there
may be other NS servers) for the 192.168.0.0 network with
255.255.255.0 netmask, you have a range 192.168.0.10 - 192.168.0.100
to lease.
- If eth0 is given such network IP, netmask and range, NS which is
joining on a network without a master could say: OK, no master, so
noone can give me the IP. Since I am a potential NS, I will become a
master on that NETWORK (so I still don't have an IP ADDRESS for my
eth0) and I will issue a client request ON THAT NETWORK again. Since
now there is a master now (itself, but why not?), it will issue an IP
for the client (itself again) and it will get a valid IP.
- No static NS addresses for clients. Clients would only join a
network and look for a NS server. No static DNS, WINS IPs or whatever
- just broadcast and you will get it from the current master NS. Many
NS possible, at any available IP on that network as they wish.

* Cons
- Some security issues may be at stake here, but that's not what I am
"into" right now. This is a potentially big thing in many network, so
a "contra" point here.
- May be slow.

* The real problem
Back to my apology, I have only 2 machines (OK, it's a big story for
such a "huge" network, sorry), one has WinXP/Linux (A) on it and the
other WinXP (B). Static IP works just fine. I wanted to make it
dynamic. If I start Linux (A), I must make it static IP in order for
it to be a DHCP server, right? So what is so dynamic about this? If I
join the other WinXP (B), it will get a dynamic IP. 50% is not
enogugh.

But a better example: assume I had two Linux machines, both of which
could be DHCP servers. How to do it? Since I need a DHCP server to be
on a static IP, both would be on static addresses. 0% is nothing.

* Questions (hopefully clear now)
Is there a way to name 4 machines Dragon, Torpedo, Mushroom and
Turtle, give their network IP, netmask and range of addresses on that
network they can occupy and the system does the rest? I don't want to
care about their IPs, but I want to control them enough. If an upper
authority gave me a network range, I must obey to that, but nothing
else. I have network from "here" to "there", it's mine and I want a
way to fit my 4 computers (and maybe afterwards add up to 60 laptops,
network printers, etc.) without caring who, when, why, ... assigns
IPs. I want simple server configuration. I want to be able to simply
rip of the above 4 machines (all NS severs) and add 15 NS servers
more. I don't want to configure them again by assigning 15 static IPs
- that is boring, error-prone and certainly not DYNAMIC.

* More questions
- Is my thinking bad or good? I.e. do you agree in common?
- Is there anything working in practice like this? Is Dynamic DNS what
I am looking for? Can DHCP do the above?
- If not, any ideas why not - what are the problems for it not being
like this?
- How do you manage (if you do) large networks? Static DNS is also
very demanding, as I see.

* That's all, folks!
Thanks again for your time and answers!
Iceman

P.S.: If you read all this, then you:
a) have a lot of time to waste,
b) are very interested for the topic,
c) are a very good person,
d) (... I surely missed something here, no? ...)

On 2004-10-27, Iceman <(E-Mail Removed)> wrote:
> When any host (even one of the 5 Linuxes running the server) turns on,
> it tries to get it's IP.

Servers never ever have dynamic IPs.
Davide

--
Take note of the toes you step on today as they may be connected
to the ass you have to kick tomorrow.
--Ben

In article <(E-Mail Removed) >, Iceman wrote:

>* Problem setup
>You have 100 machines (Linux, Windows, MacOS, whatever...).

No, I have 1700+.

>one is a master, four are slaves and that any of them can become
>master if current master fails (thunderbolt, whatever...).

Indeed - whatever. A _minor_ fly in your ointment - all servers have
to keep track of what addresses have been handed out, so they don't
try to hand it out to some other host.

>When any host (even one of the 5 Linuxes running the server) turns on,
>it tries to get it's IP. The only unique is their ethernet MAC
>address. So it sends MAC to FF:FF:FF:FF:FF:FF broadcast. There is
>currently only one master NS server, which responts. This is something
>like the DHCP's DISCOVER/OFFER/REQUEST/ACK protocol. OK, so now it has
>it's address and it can work OK.

Yes, but to avoid a huge security hole (this host _claims_ to be $FOO,
and is allowed to do $BAR - how do I know it really is who it claims
to be - even though authentication by IP address is but a baby step
beyond no security), you really want to be handing out specific
addresses to specific hosts. Here's a test for you - what is the
MAC address that COMPUTER A would see when it asks itself for an IP?

>However, if there was no NS server available, why not configure
>itself? Since you are yourself a NS, you are able to do it and, in the
>same time, become a master NS for that network.

http://www.ietf.org/internet-drafts/...nklocal-17.txt

That was "current" in early July. Microsoft has been trying to get it
past the IETF for over two years. When a system can't find it's DHCP
server, I reaches up it's ass and grabs an IP in the 169.254.0.0/16
block - so that sales weasels meeting in airport waiting areas can trade
p0rn and viruses without needing to know anything more than how to plug
in a crossover cable. Luckily I don't have to deal with windoze, and when
a computer is brought back into our facility, the first thing security
does is hand it to IT for a wipe and install. We also rarely allow
systems out the door.

>- No static addresses for NS server configurations. Administrators
>wouldn't need to configure anything, but a network IP, netmask and
>range of IP addresses.

Oh, so your DHCP server configures itself? I think you are totally
missing a concept here. It's simple. When a computer comes into the
facility, it gets a clean install, and is configured with a fixed
address. Serial numbers, MAC address, location, IP address, and user
information all goes into the security and network databases. The box
gets put on a cart, delivered to the luser, plugged in and tested, and
we're done.

>They could say: you are a possible NS (there may be other NS servers)
>for the 192.168.0.0 network with 255.255.255.0 netmask, you have a
>range 192.168.0.10 - 192.168.0.100 to lease.

Well, that's only 91 hosts, but what is the advantage?

>- If eth0 is given such network IP, netmask and range, NS which is
>joining on a network without a master could say: OK, no master, so
>noone can give me the IP. Since I am a potential NS, I will become a
>master on that NETWORK (so I still don't have an IP ADDRESS for my
>eth0) and I will issue a client request ON THAT NETWORK again.

Why would your servers all be down? Are they solar powered or something?

>Since now there is a master now (itself, but why not?), it will issue an
>IP for the client (itself again) and it will get a valid IP.

Except DHCP servers can't issue an address to themselves. Because they
can't talk on the net, What happens if two OR MORE of your servers boot
at the same time (there was a building power outage) They all look for
the DHCP server - and fine none. So, they all become "masters" and hand
out the same address to themselves. Honest and true - this BOOTP/DHCP crap
has been around for more than a week (try 1985), and there have been
people with lots of network skills looking at it. Things really are done
certain ways for a reason.

>- No static NS addresses for clients. Clients would only join a
>network and look for a NS server. No static DNS, WINS IPs or whatever

and DNS doesn't work easily with dynamic addresses.

>- just broadcast and you will get it from the current master NS. Many
>NS possible, at any available IP on that network as they wish.
>
>* Cons
>- Some security issues may be at stake here,

Wow, what an understatement. You have a system with nearly zero security.
What happens when your baddass buddy drops by with his box, which takes
over - even though you never noticed it.

>but that's not what I am "into" right now.

I'm sorry - but there are a few of us who are "into" that stuff. It's called
the real world.

>This is a potentially big thing in many network, so a "contra" point here.
>- May be slow.

It would cost a fraction of a second. How often are you rebooting your
systems?

>* The real problem
>Back to my apology, I have only 2 machines (OK, it's a big story for
>such a "huge" network, sorry), one has WinXP/Linux (A) on it and the
>other WinXP (B). Static IP works just fine. I wanted to make it
>dynamic.

Why? You still haven't shown any benefit for having dynamic addresses,
and you don't seem to want to think about the disadvantages. Say you
have five systems, and want to share the hard disks of all. How do you
know which is which? The only sane way around that is to use reservations
where MAC address 08:00:20:cd:44:1f will ALWAYS get address 192.168.1.11
whenever it boots - that's doable, at only a moderate increase in the
setup of your DHCP server - but why bother, if it's going to get the
same address from a DHCP server, just make it static and don't even HAVE
the DHCP server - it's a waste of a CPU.

>Is there a way to name 4 machines Dragon, Torpedo, Mushroom and
>Turtle, give their network IP, netmask and range of addresses on that
>network they can occupy and the system does the rest?

Reserved addresses. This MAC _always_ gets that IP address. Remember,
TCP/IP doesn't know names divorced from IP addresses. That was a windoze
NETBEUI concept, and we all know that microsoft abhors security.

>I don't want to care about their IPs, but I want to control them enough.

Then don't use IP. Use a non-routable protocol like NETBEUI, IPX, or
Banyan Vines. You then have one box that knows this crap, and has a
network connection to the IP networks and masquerades for you.

>I want a way to fit my 4 computers (and maybe afterwards add up to 60
>laptops, network printers, etc.) without caring who, when, why, ...
>assigns IPs. I want simple server configuration. I want to be able to
>simply rip of the above 4 machines (all NS severs) and add 15 NS servers
>more. I don't want to configure them again by assigning 15 static IPs
>- that is boring, error-prone and certainly not DYNAMIC.

Then don't use IP. Use a non-routable protocol like NETBEUI, IPX, or
Banyan Vines.

> More questions
>- Is my thinking bad or good? I.e. do you agree in common?

Try to guess why microsoft has invented TCP/IP to replace NETBEUI. Try to
think why NOVELL switched from IPX to IP. Try to find someone who even
remembers what Banyan Vines is.

>- Is there anything working in practice like this? Is Dynamic DNS what
>I am looking for? Can DHCP do the above?

Possibly.

>- If not, any ideas why not - what are the problems for it not being
>like this?

Security. You may not care - the rest of the world does.

>- How do you manage (if you do) large networks? Static DNS is also
>very demanding, as I see.

Piece of cake. The crews that are acceptance testing/configuring the
new hardware can do this in their sleep. The luser's boss does all the
paperwork about ordering the hardware. It's delivered to the IT shop
who do all the setup. The registrar (even with 1700 systems, that's a
small part time job that can be done by a student intern), gets a mail
with the requested hostnames (3 in case the first two chosen have already
been allocated), the location, user name, serial numbers, and MAC address
He chooses an appropriate address, and sends word to the IT crew. The
data gets put into the network and security files, and all is gravy.

Old guy

Is there a problem for a server having dynamic IP? If you KNOW that
IP, you can access it. It doesn't matter if it changes over time. The
only problem is that the server's CURRENT IP address is known to the
one who needs to access it, right?

Static DNS doesn't allow this - you need to type the static address in
the DNS database. Theoretically, however, this wouldn't be a problem.
You can use MAC broadcast to find any server and it's IP. DHCP
server's address is NOT known to a DHCP client, but it works anyway.
Why restrict yourself to put a server on a static IP when a client
will find it regardless of it's address? Why restrict an administrator
to setup DNS statically, while it can be a simple database which any
host on a given local network would just update when it turns on? The
benefit of all these would be: setup Linux (or any other system, there
should be no restriction), say: I have a local network named ABCD, it
has IP 192.168.0.0, netmask 255.255.255.0, you have a range of IPs
192.168.0.10 - 192.168.0.60 (51 addresses) to give to any host that
registers. Save this registration information and when any other host
wants it's address, you will provide it. If 51 hosts connect, with
their MACs, disconnect, then anoter 51 hosts connect, they will get
the same addresses from the given range, but will be different hosts
and it will work.

Assume you have 10 servers on 192.168.0.10 - 192.168.0.19 that do the
above, configured with no static IPs. You just plug 9 and all of them
would do it. If you unpluged one (192.168.0.19), then plug another
one, you still would do it. You wouldn't need to be reconfigured to
have IP 192.168.0.19 - it will get it automatically from the available
servers (192.168.0.10 - 192.168.0.18) and it can be any from
192.168.0.19 - 192.168.0.60. If no servers were available, it will
configure itself with any IP from 192.168.0.10 - 192.168.0.60 range.

I suppose that if the computers could configure themselves, why bother
yourself? That's why the computers are anyway - help humans by doing
something for them, not bother them with setup, configuration,
installation, ... Do you agreee?

----
Davide Bianchi <(E-Mail Removed)> wrote in message news:<(E-Mail Removed) n.net>...
> On 2004-10-27, Iceman <(E-Mail Removed)> wrote:
> > When any host (even one of the 5 Linuxes running the server) turns on,
> > it tries to get it's IP.
>
> Servers never ever have dynamic IPs.
> Davide

Iceman wrote:
> Is there a problem for a server having dynamic IP? If you KNOW that
> IP, you can access it. It doesn't matter if it changes over time. The
> only problem is that the server's CURRENT IP address is known to the
> one who needs to access it, right?
>

servers should never have a dynamic IP!! why have a server if it's
going to be dynamic? not very secure either.

i blame M$ for this DDNS bastardization w/ ActiveDirectory making people
think it's OK for server to have dynamic addresses... even tho in very
small print they explicitly say the server should have a static IP in
DNS. and even M$ has a secure method for making sure everything is in
order for that server in DNS when using DDNS.

> Static DNS doesn't allow this - you need to type the static address in
> the DNS database. Theoretically, however, this wouldn't be a problem.
> You can use MAC broadcast to find any server and it's IP. DHCP
> server's address is NOT known to a DHCP client, but it works anyway.

huh? the DHCP server's address *IS* known by the client. how else
would it renew the lease?

> Why restrict yourself to put a server on a static IP when a client
> will find it regardless of it's address? Why restrict an administrator
> to setup DNS statically, while it can be a simple database which any
<....snip....>

again... a M$ bastardization. i bet you if your server's IP address
changed and you had a mission critical FTP job that had to get to that
server it's not going to work very well. too easy to get screwed and up
too manual for a simple database (whatever that is, why not DNS?). how
hard is it to go into a zone file and add a static entry and never
change it again? i must be doing something wrong?

>
> Assume you have 10 servers on 192.168.0.10 - 192.168.0.19 that do the
> above, configured with no static IPs. You just plug 9 and all of them
> would do it. If you unpluged one (192.168.0.19), then plug another
> one, you still would do it. You wouldn't need to be reconfigured to
> have IP 192.168.0.19 - it will get it automatically from the available
> servers (192.168.0.10 - 192.168.0.18) and it can be any from
> 192.168.0.19 - 192.168.0.60. If no servers were available, it will
> configure itself with any IP from 192.168.0.10 - 192.168.0.60 range.
>

something is wrong if you are unplugging and plugging in servers at whim?

> I suppose that if the computers could configure themselves, why bother
> yourself? That's why the computers are anyway - help humans by doing
> something for them, not bother them with setup, configuration,
> installation, ... Do you agreee?

DHCP is for clients and workstations, not for servers. should a DHCP
server configure itself from nowhere? someone has to conigure it and
forget it.

>
> ----
> Davide Bianchi <(E-Mail Removed)> wrote in message news:<(E-Mail Removed) n.net>...
>
>>On 2004-10-27, Iceman <(E-Mail Removed)> wrote:
>>
>>>When any host (even one of the 5 Linuxes running the server) turns on,
>>>it tries to get it's IP.
>>
>>Servers never ever have dynamic IPs.
>>Davide

Iceman wrote:

> Is there a problem for a server having dynamic IP? If you KNOW that
> IP, you can access it. It doesn't matter if it changes over time. The
> only problem is that the server's CURRENT IP address is known to the
> one who needs to access it,

1) DHCP addresses do not normally change, unless a system has been off for
more than the lease time. Even then it may still be able to get the same
IP.

b) A DHCP server can reserve an IP for a corresponding MAC, so that one
machine and only that one can get a specific address.

--

(This space intentionally left blank)

(E-Mail Removed) (Iceman) writes:

> Is there a problem for a server having dynamic IP? If you KNOW that
> IP, you can access it. It doesn't matter if it changes over time. The
> only problem is that the server's CURRENT IP address is known to the
> one who needs to access it, right?
>
> Static DNS doesn't allow this - you need to type the static address
> in

So use dynamic DNS.
--
Joseph J. Pfeiffer, Jr., Ph.D. Phone -- (505) 646-1605
Department of Computer Science FAX -- (505) 646-1002
New Mexico State University http://www.cs.nmsu.edu/~pfeiffer