Le 2012-03-09 à 17:56, Ole Trøan a écrit :
>>>> Sec. 4.4 of draft-mdt-softwire-mapping-address-and-port-03 says:
>>>> "Default Mapping Rule:
>>>> {2001:db8:0001:0000:<interface-id>:/128 (Rule IPv6 prefix),
>>>> 0.0.0.0/0 (Rule IPv4 prefix),
>>>> 192.0.2.1 (BR IPv4 address)}
>>>>
>>>> Example 3: Default Mapping Rule
>>>>
>>>> In most implementations of a routing table, the next-hop address must be
>>>> of the same address family as the prefix. To satisfy this requirement a
>>>> BR IPv4 address is included in the rule. Giving a default route in the
>>>> IPv4 routing table:
>>>> 0.0.0.0 -> 192.0.2.1, MAP-Interface0"
>>>>
>>>> I must admit I found this very confusing:
>>>> - It seems to be a purely internal matter, having therefore no place in a
>>>> proposed standard.
>>
>> Not answered.
>
> see two lines below.
They don't explain why one would need "0.0.0.0 -> 192.0.2.1, MAP-Interface0" in
a proposed standard.
>>>> - In an internal IPv4 routing table, any never-assigned address seems
>>>> sufficient (no need for dependence on the BR IPv4 address).
>>>
>>> the idea of the BR also being part of the domain has some benefits.
>>> e.g. it is possible to IPv4 ping the default gateway. it isn't strictly
>>> necessary as you say, but I think it has value in troubleshooting.
>>>
>>>> - How a MAP-T CE could get the BR IPv4 address isn't specified (no DHCPv6
>>>> parameter specified for its transmission).
>>>
>>> manual configuration.
>>
>> Are you really suggesting that MAP-T CEs would be manually configured?
>
> no, sorry, I thought you asked how the BR was configured.
No hurry is needed, but please answer THE question. Not only does it concern a
CE (not a BR), but also a -T (not a -E).
> the CE can extract the BR IPv4 address from the BR IPv6 address.
The draft says:
"The Rule IPv6 prefix is the IPv6 address or prefix of the BR. Which is used,
is dependent on the mode used. For example translation requires that the IPv4
destination address is encoded in the BR IPv6 address, so only a prefix is used
in the DMR to allow for a generated interface identifier."
Thanks,
RD
>
> Ole
_______________________________________________
Softwires mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/softwires
