Gunter –

I was not involved in writing RFC 6119 – but I don’t believe what you reference 
is a typo. The section from which you quote starts with:

“To prevent an SRLG TLV and an IPv6 SRLG TLV in the same logical LSP
   from causing confusion of interpretation, the following rules are
   applied.”

So it seems clear to me that this was done intentionally.
The authors are in the best position to explain why (if they remember after all 
these years 😊 ).

In regards to 
RFC8919<https://datatracker.ietf.org/doc/html/rfc8919#section-4.3>, as we were 
defining a new TLV that supported both IPv4 and IPv6, we saw no reason to limit 
the per link advertisement to a single TLV – especially given the potential for 
exceeding the 255 byte TLV limit. So we intentionally did not impose this 
restriction.

Note that RFC5307<https://datatracker.ietf.org/doc/html/rfc5307#section-1.4> 
does not make an explicit statement regarding TLV 138 – so you are correct in 
thinking that it is possible to have multiple TLV 138 for the same link. This 
has been made explicit in 
https://www.ietf.org/archive/id/draft-ietf-lsr-multi-tlv-19.html#name-mp-tlv-for-is-is-top-level-
 .
Given the explicit statement in RFC 6119 forbidding multiple TLVs/link for TLV 
139 we could not alter that in the MP-TLV draft w/o introducing backwards 
compatibility issues.

HTH

   Les


From: Gunter van de Velde (Nokia) 
<[email protected]>
Sent: Tuesday, October 28, 2025 7:49 AM
To: [email protected]; [email protected]; 
[email protected]; lsr <[email protected]>
Subject: [Lsr] clarification question about IPv6 SRLG TLV procedure documented 
by RFC6119

[writing as individual contributor without AD hat]

Hi folks, LSR WG,

I am trying to figure out if RFC6119 has a typo or if it was the intent. In 
RFC6119 we see the following procedure:

“There MUST NOT be more than one IPv6 SRLG TLV for a given link.”
https://datatracker.ietf.org/doc/html/rfc6119#section-4.4

So, this forbids more as a single IPV6 SRLG TLV for a single link. The RFC is 
very clear and formal in this procedure. No issue understanding this.

But this seems to deviate from prior SRLG TLV procedures.

According RFC8919<https://datatracker.ietf.org/doc/html/rfc8919#section-4.3>, 
its possible to have multiple Type 238 TLVs for a particular link.
“Multiple TLVs for the same link MAY be advertised.”

Also according to 
RFC5307<https://datatracker.ietf.org/doc/html/rfc5307#section-1.4>, for TLV138 
the following statement is made.
   “The SRLG TLV MAY occur more than once within the IS-IS Link State
   Protocol Data Units.” – So for me its possible to have multiple type 138 
TLVs as well.

So not sure why this restriction is there only for Type 139 TLVs? Was this 
intentional or a accidental typo that slipped into RFC6119? Thoughts?

Brgds,
G/
(as individual contributor)
_______________________________________________
Lsr mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to