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

[Brian E Carpenter <[e-mail]>]

On 27-Nov-25 07:19, Michael Sweet wrote:
> Ole,
> 
>> On Nov 26, 2025, at 12:52 PM, Mark Smith <[e-mail]> wrote:
>>
>> Hi Ole,
>>
>> On Thu, 27 Nov 2025 at 00:18, Ole Trøan <[e-mail]> wrote:
>>>
>>> Mark,
>>>
>>> Since the loopback address is of link-local scope you can already do this with multiple loopback interfaces.
>>> ::1%loop0
>>> ::1%loop1
>>>
>>
>> That's interesting.
>>
>> One thing I think worth avoiding is having to use loopback addresses
>> with zone IDs to minimise the amount of typing since these addresses
>> may be commonly typed in manually.
> 
> One other thing, and something that has been a sore spot for IoT devices like printers forever, is that relying on using zone/scope IDs doesn't work for services that use URIs.  The old URI WG and somewhat more recently 6man have tried extending URIs to support IPv6 addresses with zone/scope IDs three different ways (mainly for link-local addresses but really for any address whose routing depends on it) and have failed.
> 
> There isn't a lot we can do for link-local addresses (there we have to rely on mDNS hostnames), but for loopback addresses things *should* Just Work if every loopback interface has a unique address and not "::1%ZONEID".

None of which is helped by:

(a) the confusion I already mentioned between loopback meaning "loop back" and loopback meaning "virtual interface";

(b) the fact that Linux fails to support a default zoneid;

(c) Linux seems to think that %lo is interface 0, which should in fact be the default interface;

(d) as far as my experiments went, mDNS on Linux has messed up support for link-local addresses, which all need a zoneid because of (b). Which is probably the real reason that ::/96 would work on Linux virtual interfaces but fe80::/10 wouldn't.

    Brian

> 
> ________________________
> Michael Sweet
> 
> --------------------------------------------------------------------
> IETF IPv6 working group mailing list
> [e-mail]
> List Info: https://mailman3.ietf.org/mailman3/lists/[e-mail]/
> --------------------------------------------------------------------