RE: Google wants to be your Internet
We also see this with extranet/supply-chain-type connectivity between large companies who have overlapping address space, and I'm afraid it's only going to become more common as more of these types of relationships are established.
Fortunately, IP addresses are not intended for use on the Internet. Rather, they are intended for use with Internet Protocol (IP) implementations. That's why the RIRs, in alignment with RFC 2050, section 3(a), do give out IP address allocations to organizations who are connected to extranet-type networks. If you read RFC 1918, section 2, category 3, you will see that this is consistent. So if the power companies want to assign a unique network address to all power meters then there is no good reason to stop them. After all, it is consistent with the goals of the original IP designers to address every light switch and toaster. Just remember, IP addresses are *NOT* Internet addresses. They are Internet Protocol addresses. Connection to the Internet and public announcement of prefixes are totally irrelevant. --Michael Dillon
On Jan 24, 2007, at 12:33 AM, <michael.dillon@bt.com> wrote:
Just remember, IP addresses are *NOT* Internet addresses. They are Internet Protocol addresses. Connection to the Internet and public announcement of prefixes are totally irrelevant.
Of course I understand this, but I also understand that if one can get away with RFC1918 addresses on a non-Internet-connected network, it's not a bad idea to do so in and of itself; quite the opposite, in fact, as long as one is sure one isn't buying trouble down the road. ----------------------------------------------------------------------- Roland Dobbins <rdobbins@cisco.com> // 408.527.6376 voice Technology is legislation. -- Karl Schroeder
On Wed, 24 Jan 2007 02:07:06 -0800 Roland Dobbins <rdobbins@cisco.com> wrote:
Of course I understand this, but I also understand that if one can get away with RFC1918 addresses on a non-Internet-connected network, it's not a bad idea to do so in and of itself; quite the opposite, in fact, as long as one is sure one isn't buying trouble down the road.
The problem is that you can't be sure that if you use RFC1918 today you won't be bitten by it's non-uniqueness property in the future. When you're asked to diagnose a fault with a device with the IP address 192.168.1.1, and you've got an unknown number of candidate devices using that address, you really start to see the value in having world wide unique, but not necessarily publically visible addressing. -- "Sheep are slow and tasty, and therefore must remain constantly alert." - Bruce Schneier, "Beyond Fear"
On Jan 24, 2007, at 4:58 AM, Mark Smith wrote:
The problem is that you can't be sure that if you use RFC1918 today you won't be bitten by it's non-uniqueness property in the future. When you're asked to diagnose a fault with a device with the IP address 192.168.1.1, and you've got an unknown number of candidate devices using that address, you really start to see the value in having world wide unique, but not necessarily publically visible addressing.
That's what I meant by the 'as long as one is sure one isn't buying trouble down the road' part. Having encountered problems with overlapping address space many times in the past, I'm quite aware of the pain, thanks. ;> RFC1918 was created for a reason, and it is used (and misused, we all understand that) today by many network operators for a reason. It is up to the architects and operators of networks to determine whether or not they should make use of globally-unique addresses or RFC1918 addresses on a case-by-case basis; making use of RFC1918 addressing is not an inherently stupid course of action, its appropriateness in any given situation is entirely subjective. ----------------------------------------------------------------------- Roland Dobbins <rdobbins@cisco.com> // 408.527.6376 voice Technology is legislation. -- Karl Schroeder
I hear you on the double, triple nat nightmare, I'm there myself. I'm working on rolling out VRFs to solve that problem, still testing. The nat complexities and bugs (nat translations losing their mind and killing connectivity for important apps) are just too much for some of our customers, users, etc to deal with. Some days it kills me that v6 is still not really viable, I keep asking providers where they're at with it. Their most common complaint is that the operating systems don't support it yet. They mention primarily Windows since that is what is most implemented, not in the colo world but what the users have. I suggested they offer a service that somehow translates (heh, shifting the pain to them) v4 to v6 for their customers to move it along. Roland Dobbins wrote:
On Jan 24, 2007, at 4:58 AM, Mark Smith wrote:
The problem is that you can't be sure that if you use RFC1918 today you won't be bitten by it's non-uniqueness property in the future. When you're asked to diagnose a fault with a device with the IP address 192.168.1.1, and you've got an unknown number of candidate devices using that address, you really start to see the value in having world wide unique, but not necessarily publically visible addressing.
That's what I meant by the 'as long as one is sure one isn't buying trouble down the road' part. Having encountered problems with overlapping address space many times in the past, I'm quite aware of the pain, thanks.
;>
RFC1918 was created for a reason, and it is used (and misused, we all understand that) today by many network operators for a reason. It is up to the architects and operators of networks to determine whether or not they should make use of globally-unique addresses or RFC1918 addresses on a case-by-case basis; making use of RFC1918 addressing is not an inherently stupid course of action, its appropriateness in any given situation is entirely subjective.
----------------------------------------------------------------------- Roland Dobbins <rdobbins@cisco.com> // 408.527.6376 voice
Technology is legislation.
-- Karl Schroeder
-----Original Message----- From: owner-nanog@merit.edu [mailto:owner-nanog@merit.edu] On Behalf Of Jason LeBlanc Sent: Wednesday, January 24, 2007 8:40 AM To: Roland Dobbins Cc: NANOG Subject: Re: Google wants to be your Internet
I hear you on the double, triple nat nightmare, I'm there myself. I'm working on rolling out VRFs to solve that problem, still testing. The nat complexities and bugs (nat translations losing their mind and killing connectivity for important apps) are just too much for some of our customers, users, etc to deal with. Some days it kills me that v6 is still not really viable, I keep asking providers where they're at with it. Their most common complaint is that the operating systems don't support it yet. They mention primarily Windows since that is what is most implemented, not in the colo world but what the users have. I suggested they offer a service that somehow translates (heh, shifting the pain to them) v4 to v6 for their customers to move it along.
Windows XP SP2 has IPv6. It isn't enabled by default, but it's not difficult to do. Apparently Vista does do IPv6 by default out of the box, but I don't have a Vista system to play with yet to confirm this. Jamie Bowden -- "It was half way to Rivendell when the drugs began to take hold" Hunter S Tolkien "Fear and Loathing in Barad Dur" Iain Bowen <alaric@alaric.org.uk>
On 24-Jan-2007, at 10:01, Jamie Bowden wrote:
Some days it kills me that v6 is still not really viable, I keep asking providers where they're at with it. Their most common complaint is that the operating systems don't support it yet. They mention primarily Windows since that is what is most implemented, not in the colo world but what the users have.
Windows XP SP2 has IPv6. It isn't enabled by default, but it's not difficult to do.
Apparently Vista does do IPv6 by default out of the box, but I don't have a Vista system to play with yet to confirm this.
I might argue that, legacy systems and hardware aside, the main reason that v6 might be considered non-viable these days is the lack of customers willing to pay for it. I don't think the viability of v6 has been blocking on operating systems or router hardware for quite some time, now. It's still a problem for many operational support systems, but arguably that would change rapidly if there was some prospect of revenue. Joe
On Wed, Jan 24, 2007 at 05:23:10AM -0800, Roland Dobbins wrote:
RFC1918 was created for a reason, and it is used (and misused, we all understand that) today by many network operators for a reason.
I used 10/8 for my LAN a while back until my ISP's routers advertised in DHCP suddenly started using the same 10.0.0.0/24... so I ended up having to renumber, or else I wouldn't route out the WAN link for them. What a pain. I'd name the ISP if I had the Time. -- Kill dash nine, and its no more CPU time, kill dash nine, and that process is mine. -><- <URL:http://www.subspacefield.org/~travis/> For a good time on my UBE blacklist, email john@subspacefield.org.
participants (7)
-
Jamie Bowden -
Jason LeBlanc -
Joe Abley -
Mark Smith -
michael.dillon＠bt.com -
Roland Dobbins -
Travis H.