1. The broadcast parameter is necessary for multicast to work towards a particular set of addresses. You are correct that the link-local is really the only one that needs this parameter. On the other hand, you still need to have a MAP for ALL of your other addresses otherwise you'll get encaps failed. The broadcast parameter does not give you L3-L2 lookups for all addresses! So it's two different things. 2. BGP/MBGP is a tcp session. It's a tcp session that happens to carry routing information. So you need to first establish the tcp session. Whether you do that via link-local or global unicast addresses is entirely up to you as long as it's reachable! If you use link-local, you may run into some problems in ibgp because no next-hop information is changed in updates. And when you send YOUR link-local reachable address to someone else on a different interface, that doesn't do any good. :) 3. Both will work. Passive interface is easer IMHO. HTH,
Scott Morris, CCIE4 (R&S/ISP-Dial/Security/Service Provider) #4713, JNCIE-M #153, JNCIS-ER, CISSP, et al. CCSI/JNCI-M/JNCI-ER VP - Technical Training - IPexpert, Inc. IPexpert Sr. Technical Instructor [EMAIL PROTECTED] Telephone: +1.810.326.1444 Fax: +1.810.454.0130 http://www.ipexpert.com _____ From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Roger Sent: Saturday, April 05, 2008 7:44 AM To: [email protected] Subject: [OSL | CCIE_RS] R&S v9 section 11 -->>some general ipv6 questions Hi all experts out there, 1) Is it correct that while configuring IPv6 over Frame-Relay I need the "broadcast" keyword, used in the frame-relay map command, only on the link-local address mapping? There is no broadcast in IPv6, so it's used for multicast. This would mean that for all communication for example in OSPF (updates.), the link local address is used? As far as I can see, everything would work with link local addresses; neigbor relationship, population of routing tables., but I still get an Frame-Relay encapsulation failure, so I must use a global address as well. Can someone explain that? 2) In Multiprotocol BGP I can use both as neigbor statement, the global address or the link-local address. This confuses me somehow, is the approach here really different? 3) On Task 11.4 we are using passive interfaces in OSPFv3 in order to prevent the two routers R5 and R6 to become neighbors over the same segment. Well, I just used the instance command under the related interface config and to me it works fine, no neighborship has been build. Is that a correct configuration way, alternative to passive-interface or did I miss something to think about? Here the config for the two connected interfaces: R5: ! interface GigabitEthernet0/0 no ip address duplex auto speed 100 ipv6 address 2000:57:57:5757::5/64 ipv6 address FE80::5 link-local ipv6 ospf 1 area 57 instance 11 ! R7: interface FastEthernet0/0 no ip address duplex auto speed auto ipv6 address 2000:57:57:5757::7/64 ipv6 address FE80::7 link-local ipv6 ospf 2 area 57 instance 22 Thanks in advance Roger No virus found in this outgoing message. Checked by AVG. Version: 7.5.519 / Virus Database: 269.22.6/1360 - Release Date: 04.04.2008 18:02
