IS-IS Protocol Extensions for Path Computation Element (PCE) Discovery
RFC 5089

Note: This ballot was opened for revision 08 and is now closed.

(Ross Callon) Yes

(Jari Arkko) No Objection

(Ron Bonica) No Objection

(Lisa Dusseault) No Objection

Comment (2007-10-03)
No email
send info
I still haven't memorized routing terminology so I exercised the terminology section.  FYI:

 - The terminology expansion of IS-IS LSP confused me.  Why is LSP "Link State PDU" here and "Label Stitched Path" everywhere else?.  
 - I assume PCED is PCE-Domain -- only the latter is expanded in terminology
 - TLV not defined or referenced (though I realize it's a very well-known TLA in the field)

Section 3.2
 - flooding scope through "L1 area" and "L2 sub-domain" -- should this be part of terminology or an explanation referenced?

Section 4.1
 - If two PCE-ADDRESS sub-TLVs appear, " only the first occurrence is processed and any others MUST be ignored".   With this requirement, what use is it to allow two of PCE-ADDRESS?  If one can be IPv6 and one IPv4 but only the first one MUST be processed, then the other one is useless.

thx -- Lisa

Lars Eggert No Objection

(Russ Housley) No Objection

(Cullen Jennings) No Objection

(Chris Newman) No Objection

(Jon Peterson) No Objection

(Tim Polk) (was Discuss) No Objection

Comment (2007-10-03)
No email
send info
In section 4.2:

Is consistent computation of the PrefL, PrefR, PrefS and PrefY field values important?  If so, are
we depending upon common administration of all PCEs?  Without a more detailed algorithm, it
seems likely a PCC could discover a set of PCEs that used very different algorithms for setting
the preference values.

(Mark Townsley) No Objection

(David Ward) (was Discuss) No Objection

(Magnus Westerlund) No Objection