Hi Eric, Please see my response in-line below.
I add the pwe3 mailing list since this document is related to the following draft submitted for pwe3. ICCP Application TLVs for VPN Route Label Sharing http://www.ietf.org/internet-drafts/draft-zhang-pwe3-iccp-label-sharing-00.txt Thanks, Mingui >-----Original Message----- >From: Eric Rosen [mailto:[email protected]] >Sent: Monday, July 22, 2013 11:10 PM >To: Mingui Zhang >Cc: [email protected]; [email protected] >Subject: Re: FW: New Version Notification for >draft-zhang-l3vpn-label-sharing-00.txt > > >As far as I can tell, your scheme only handles the case where the two egress >PEs connect to the exact same set of CEs. If so, this applicability >restriction should be clearly stated. > >In Figure 2.1, you show CE2 connected to PE3 and PE4. Now suppose we also >have CE4 connected to PE3 and PE5. If PE3 fails, one might want use PE4 as >a backup for CE2, while using PE5 as a backup for CE4. Your scheme doesn't >seem to handle this. Not exactly. If PE3 and PE4 allocate (1100, vNH1) for CE2 while PE3 and PE5 allocate (1101, vNH2) for CE4. The scheme works well. This also answers your first question. > >Also, even in the case of Figure 2.1, it seems entirely possibly that one >might want PE3's primary route to CE2 to be via the directly attached >interface, while wanting PE4's primary route to CE2 to be via PE3. Does >your scheme handle this case? Interesting point. Yes, it handles. Please see the right part of eq1: Sxy3+M. If the backup tunnel goes through Pxy->PE4->PE3->vNH, then Sxy3+M < = Sxy4+S. That means eq1 does not hold.
