[IPv6]Re: [v6ops] New draft: "The IPv6 Loopback Address Prefix"

[Ole Trøan <[e-mail]>]

Thanks for doing this!

Suggestions:

- say less about loopback mechanics. E.g. drop ingress queues etc. 

- ::/96 will likely still be treated as IPv4-compatible by address parsers and formatters. Resulting in ::127.0.0.1 f.ex. Don’t think that is a problem.

- SAS/DAS policy table needs updating as mentioned 

- DNS local use

- RFC4007 talks about the scope of ::1 quite a bit. Not sure if it needs updating. The prefix is having same scope as the address. Should be ok. 

- rfc4291 forwarding rules. This will require forwarding implementations to change. Which currently should have a drop-check for forwarding packets with a ::, ::1, … source. I presume we want the loopback prefix to have the same properties?

- probably other documents needing update?

Cheers,

Ole

On 25 Nov 2025, at 18:39, Warren Kumari <[e-mail]> wrote:



Dear 6MAN and V6OPS,

Geoff Huston and I have just submitted https://datatracker.ietf.org/doc/draft-kumari-ipv6-loopback/" rel="noopener noreferrer nofollow" class="">draft-kumari-ipv6-loopback - "The IPv6 Loopback Address Prefix"

We believe that it is within the 6MAN charter ("The 6man working group is responsible for the maintenance, upkeep, and advancement of the IPv6 protocol specifications and addressing architecture."), but I have CCed V6OPS as well, as it is clearly operational as well.

Abstract: 

"This document updates the IP Version 6 Address Architecture to define the IPv6 address prefix ::/32 as the Loopback address prefix."

Basically, this document expands the single loopback address ::1/128 into a prefix.

Yes, we are aware that there have been some previous discussions[0] on the need (or lack thereof!) of a loopback prefix in IPv6, but we believe that they are worth revisiting. 

There are a number of situations in which having more than a single address is helpful; an obvious example of this is Dockers/k8s use of 127.0.0.11 for the DNS resolver, SPAM RBL use of the last octet on 127.0.0.x to encode the type of SPAM. It is also relatively common it use this for inter-service communication in container environments. 

It is also a common practice to bind different services to different addresses in the IPv4 loopback space to allow for scaling (avoiding the "Port already in use" issue), testing, etc.  Yes, these can be somewhat emulated with ULAs and / or additional interfaces and scopes, but they are all more complicated, and much more likely to result in leakage or collision. 

Another, more recent example is the ICANN Public Comment on "Name Collision IPv6 Research Study" and proposed use of ::ffff:7f00:3535 [1] - if there was a loopback prefix this would have been a better option[2]

We expect a fairly robust discussion :-), 

W

[0]: I know I've seen them, but I quick search of my mail was unable to find these — the authors are more than happy to link to previous documents, etc. 

[1]: See long threads on 6MAN https://mailarchive.ietf.org/arch/msg/ipv6/-HrYFMwHhsUWYxSXsFIkLpF_Qgk/" target="_blank" rel="noopener noreferrer nofollow" class="">https://mailarchive.ietf.org/arch/msg/ipv6/-HrYFMwHhsUWYxSXsFIkLpF_Qgk/ and V6OPS.

[2]: Solving the technical concerns, but not necessarily the policy ones.

_______________________________________________
v6ops mailing list -- [e-mail]
To unsubscribe send an email to [e-mail]