3 ms·

There is one big problem I see with IPv6 deployment: Hosting providers don't get it. Yes, you can increasingly get IPv6 connectivity, but it's common that you

There is one big problem I see with IPv6 deployment: Hosting providers don't get it. Yes, you can increasingly get IPv6 connectivity, but it's common that you get a single /64, or even just a single address. Now, I want to run a bunch of virtual machines on my server, with a virtual network or two between them ... and already, I have to use ugly hacks to get by instead of just assigning addresses using the standard way of doing things with IPv6. In the best case, I just have to ask for more address space, but even that is completely pointless overhead that only works as an incentive to use fragile network constructs instead of just assigning addresses where they are needed. People seem to have internalized that addresses are scarce and therefore precious and therefore have to be conserved at all cost - they don't even notice that that was the problem of IPv4 (or at least the most important problem of IPv4) that IPv6 was supposed to solve, and which it indeed does solve very well, and that their efforts at conserving IPv6 address space cause more harm than they help. Lots of hosters around here are RIPE members who have their standard /32 allocation (which they could extend to /29 without any further documentation required), so if they chose to assign /56 networks to customers, they would have space for about 16 million customers before they would need to request more address space from RIPE, and for 128 million before they need to justify any further need. But apparently they consider it better use of that address space to just leave it mostly unused and make it difficult for their customers to use IPv6, not even noticing that there is a reason why RIPE assigned them a /32 ...

mappu 12y ago
Correct me if i'm wrong, but doesn't an IPv6 /64 give you 2^64 globally routeable addresses? How is that not sufficient?
- simoncion 12y ago
  You can't subnet a /64. This RFC [0][1] describes the IETF's thoughts on why allocating a block larger than /64 is to be strongly encouraged. [0] https://tools.ietf.org/html/rfc6177 https://tools.ietf.org/html/rfc6177 [1] Particularly, see page 3
- matteotom 12y ago
  AFAIK, it's customary to assign a /64 to each machine, and the last 64 bits are used for other purposes (usually the MAC address I think?). https://en.wikipedia.org/wiki/IPv6_subnetting_reference https://en.wikipedia.org/wiki/IPv6_subnetting_reference
  - simoncion 12y ago
    Perhaps we're using different terminology, but it's absolutely not customary to assign a /64 to each machine. IME one advertises one or more /64's on a given network, then (either through DHCPv6, or SLAAC [or both]) machines attached to that network acquire addresses within those spaces. These machines will have a XXXX::XXXX/64 address, but that just means that they're attached to a network that advertises a /64, not that that machine has been assigned an entire /64 for its use.
  - ghshephard 12y ago
    With IPv6 you assign a /64 to the network. All networks in IPv6 are 64 bits long, with 64 bits assigned to the host's interface ID. There are lots of ways for a host to generate an Interface ID - (http://technet.microsoft.com/en-us/library/cc736439(v=ws.10).aspx http://technet.microsoft.com/en-us/library/cc736439(v=ws.10)...) is a nice reference.
- zAy0LfpBZLC8mAC 12y ago
  Yes, it does, but individual IPv6 addresses are not directly comparable to individual IPv4 addresses, because protocol mechanisms assume that every network ("LAN") segment has its own /64. For example, IPv6 stateless auto configuration (for automatic stateless address assignment, another central innovation of IPv6) only works with a /64, so you can't use it if you have to subnet a /64. As far as the RFCs are concerned, a /64 is the minimal allocation that is used for a network segment, anything else is a hack.
- ghshephard 12y ago
  IPv6 doesn't work like IPv4. In IPv4 Parlance, you would be given a CIDR block which you could then subnet as you saw fit. I.E. The ISP might give you a /19, which you could then turn into 32 /24s (or 64 /25s, etc...) With IPv6, /64s are the basic unit, and you don't further subnet the network. So, instead of thinking of "addressable space", you should now think of "deployable networks." . An ISP that hands out a /64 is basically just giving the customer a single network. The smallest network any ISP should hand out to a customer is a /48, which lets the customer deploy up to 2^16, or 65K networks as they see fit.

mappu 12y ago

Correct me if i'm wrong, but doesn't an IPv6 /64 give you 2^64 globally routeable addresses? How is that not sufficient?

simoncion 12y ago
You can't subnet a /64. This RFC [0][1] describes the IETF's thoughts on why allocating a block larger than /64 is to be strongly encouraged. [0] https://tools.ietf.org/html/rfc6177 https://tools.ietf.org/html/rfc6177 [1] Particularly, see page 3
matteotom 12y ago
AFAIK, it's customary to assign a /64 to each machine, and the last 64 bits are used for other purposes (usually the MAC address I think?). https://en.wikipedia.org/wiki/IPv6_subnetting_reference https://en.wikipedia.org/wiki/IPv6_subnetting_reference
- simoncion 12y ago
  Perhaps we're using different terminology, but it's absolutely not customary to assign a /64 to each machine. IME one advertises one or more /64's on a given network, then (either through DHCPv6, or SLAAC [or both]) machines attached to that network acquire addresses within those spaces. These machines will have a XXXX::XXXX/64 address, but that just means that they're attached to a network that advertises a /64, not that that machine has been assigned an entire /64 for its use.
- ghshephard 12y ago
  With IPv6 you assign a /64 to the network. All networks in IPv6 are 64 bits long, with 64 bits assigned to the host's interface ID. There are lots of ways for a host to generate an Interface ID - (http://technet.microsoft.com/en-us/library/cc736439(v=ws.10).aspx http://technet.microsoft.com/en-us/library/cc736439(v=ws.10)...) is a nice reference.
zAy0LfpBZLC8mAC 12y ago
Yes, it does, but individual IPv6 addresses are not directly comparable to individual IPv4 addresses, because protocol mechanisms assume that every network ("LAN") segment has its own /64. For example, IPv6 stateless auto configuration (for automatic stateless address assignment, another central innovation of IPv6) only works with a /64, so you can't use it if you have to subnet a /64. As far as the RFCs are concerned, a /64 is the minimal allocation that is used for a network segment, anything else is a hack.
ghshephard 12y ago
IPv6 doesn't work like IPv4. In IPv4 Parlance, you would be given a CIDR block which you could then subnet as you saw fit. I.E. The ISP might give you a /19, which you could then turn into 32 /24s (or 64 /25s, etc...) With IPv6, /64s are the basic unit, and you don't further subnet the network. So, instead of thinking of "addressable space", you should now think of "deployable networks." . An ISP that hands out a /64 is basically just giving the customer a single network. The smallest network any ISP should hand out to a customer is a /48, which lets the customer deploy up to 2^16, or 65K networks as they see fit.