Re: [spring] SRv6 Network Programming and Link Local Source Addresses

["Darren Dukes (ddukes)" <[e-mail]>]

Hi Ron, I agree with Bob here.

Section 4.2 pseudocode simply says an implementation would use a predetermined egress adjacency instead of performing a FIB lookup to find one.  
It specifies the SID processing, not the entire IPv6 data path.

It has no text that would indicate RFC4291 text on link-local addresses and routers would not apply.

As a side note, every routing header currently defined (even those now deprecated) do not re-state the RFC4291 text.

Thanks,
  Darren


> On Dec 2, 2019, at 10:58 AM, Ron Bonica <rbonica=[e-mail]> wrote:
> 
> Bob,
> 
> Before we debate presentation too much, we should let Pablo answer the original question. Will the packet be dropped or forwarded?
> 
> If the packet will be dropped, how is the reader of Section 4.2 to know this? Normally, pseudocode is taken literally, and the pseudocode in Section 4.2 suggests that the packet will be forwarded.
> 
> One way to wiggle out of this problem is to include a sentence at the beginning of Section 4 saying, "When the following pseudocode contradicts RFC 4291 or 8200, RFCs 4291 and 8200 take precedence.
> 
>                                                                                                              Ron
> 
> 
> 
> 
> Juniper Business Use Only
> 
> -----Original Message-----
> From: Bob Hinden <[e-mail]> 
> Sent: Monday, December 2, 2019 10:47 AM
> To: Ron Bonica <[e-mail]>
> Cc: Bob Hinden <[e-mail]>; Mark Smith <[e-mail]>; SPRING WG <[e-mail]>
> Subject: Re: [spring] SRv6 Network Programming and Link Local Source Addresses
> 
> Ron,
> 
>> On Dec 2, 2019, at 7:36 AM, Ron Bonica <[e-mail]> wrote:
>> 
>> Bob,
>> 
>> Take a look at Section 4.2. The pseudocode is pretty specific.
> 
> Please explain.  I don’t see that.
> 
> Thanks,
> Bob
> 
> 
>> 
>>                                           Ron
>> 
>> 
>> 
>> Juniper Business Use Only
>> 
>> -----Original Message-----
>> From: Bob Hinden <[e-mail]>
>> Sent: Sunday, December 1, 2019 5:56 PM
>> To: Ron Bonica <[e-mail]>
>> Cc: Bob Hinden <[e-mail]>; Mark Smith <[e-mail]>; SPRING WG <[e-mail]>
>> Subject: Re: [spring] SRv6 Network Programming and Link Local Source Addresses
>> 
>> Ron,
>> 
>>> On Dec 1, 2019, at 2:47 PM, Ron Bonica <[e-mail]> wrote:
>>> 
>>> Mark, Bob,
>>> 
>>> Yes, I agree that routers should not forward packets with link local source addresses.
>> 
>> or Destination addresses.
>> 
>>> 
>>> Pablo,
>>> 
>>> Maybe we should update section 4.2 of the network programming draft to reflect this?
>> 
>> I was thinking that unless network programming has text that might cause one to think it overrides the defined behavior from rfc4291 for link-local addresses, I am not sure it has to be mentioned.
>> 
>> Bob
>> 
>> 
>>> 
>>>                                                                Ron
>>> 
>>> 
>>> From: Mark Smith <[e-mail]>
>>> Sent: Sunday, December 1, 2019 5:31 PM
>>> To: Bob Hinden <[e-mail]>
>>> Cc: Ron Bonica <[e-mail]>; SPRING WG <[e-mail]>
>>> Subject: Re: [spring] SRv6 Network Programming and Link Local Source Addresses
>>> 
>>> 
>>> 
>>> On Mon, 2 Dec 2019, 08:35 Bob Hinden, <[e-mail]> wrote:
>>> Ron,
>>> 
>>>> On Nov 30, 2019, at 12:36 PM, Ron Bonica <rbonica=[e-mail]> wrote:
>>>> 
>>>> Pablo,
>>>> 
>>>> 
>>>> 
>>>> Consider the packet (SA,DA) (S3, S2, S1; SL) where:
>>>> 
>>>> 
>>>> 
>>>>     • SA is link-local (fe80)
>>>>     • DA, S3, S2, and S1 are all END.X
>>>> 
>>>> 
>>>> Section 4.2 suggests that this packet will be delivered over multiple hops to its destination, regardless of its link-local source address.
>>> 
>>> I would think that RFC2460 Section 2.5.6. "Link-Local IPv6 Unicast Addresses” covers this:
>>> 
>>>  Link-Local addresses are for use on a single link.  Link-Local
>>>  addresses have the following format:
>>> 
>>>  |   10     |
>>>  |  bits    |         54 bits         |          64 bits           |
>>>  +----------+-------------------------+----------------------------+
>>>  |1111111010|           0             |       interface ID         |
>>>  +----------+-------------------------+----------------------------+
>>> 
>>>  Link-Local addresses are designed to be used for addressing on a
>>>  single link for purposes such as automatic address configuration,
>>>  neighbor discovery, or when no routers are present.
>>> 
>>>  Routers must not forward any packets with Link-Local source or
>>>  destination addresses to other links.
>>> 
>>> I think that's RFC4291.
>>> 
>>> RFC4007, "IPv6 Scoped Address Architecture" does too, more generally and probably more formally, in particular section 9, "Forwarding".
>>> 
>>> Regards,
>>> Mark.
>>> 
>>> 
>>> 
>>> Bob
>>> 
>>> 
>>>> 
>>>> 
>>>> 
>>>> Is this the case?
>>>> 
>>>> 
>>>> 
>>>>                                                            Ron
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>>> Juniper Business Use Only
>>>> _______________________________________________
>>>> spring mailing list
>>>> [e-mail]
>>>> https://www.ietf.org/mailman/listinfo/spring
>>> 
>>> _______________________________________________
>>> spring mailing list
>>> [e-mail]
>>> https://www.ietf.org/mailman/listinfo/spring
>>> 
>>> Juniper Business Use Only
> _______________________________________________
> spring mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/spring