Salman Asadulah (CCIE# 2240) is a Cisco Distinguished Engineer and also serves as IPv6 Forum Fellow, Broadband Forum Ambassador, and Co-chair of the IPv6 Education Certification Program. He has been working with large-scale IP and multiservice network and technologies for more than 15 years. He is a coauthor and contributor to IETF RFCs/IDs and has written three Internetworking books: Cisco CCIE Fundamentals: Network Design & Case Study, PDIO of the IPT Networks, and Deploying IPv6 in Broadband Access Networks.
The following experts were helping Salman answer a few of the questions asked during the session: DSE Philip Remaker, Jim Bailey, Rama Darbha, Srinivasa Neppalli, and Pete Lumbis. All of them are top Experts on IPv6.
You can download the presentation slides in PDF format here. The Complete Recording of this live Webcast can be accessed here.
IPv6 General Questions
Q. How do I motivate higher management to drive process?
Q. Do the servers or routers need any additional resources in order to fulfill the IPv6 requests?
A. Routers, servers, switches, firewalls, etc, just need to be running the right software in order to handle IPv6 traffic and support the features that you need. Aside from that, there are no additional resources needed in order to process IPv6 traffic.
Q. Are there any ISPs in India?
A. India is pushing hard for IPv6, but mostly in the government and enterprise areas. ISPs have been slower than we would like. Refer to http://en.wikipedia.org/wiki/IPv6_deployment#India and call your provider. Mobile providers will probably lead here.
Q. Does anyone know exactly what Cisco IP6FD is and where to go in order to get documentation about it?
Q. Can I use Global ARIN NA IPv6 Allocation for my company in APAC or Europe? Will ISP in APAC and Europe route ARIN?
A. There is no policy that prevents you from using ARIN space in RIPE, APNIC, LACNIC, AFRINC, etc. You do need to ensure that your ISP's routing policy will support what you intend to do. Most ISPs today are more concerned about prefix length and not RIR origin. That policy could change. You can use BGP looking glasses and route servers in order to check what is happening in the IPv6 Internet regarding prefix lengths that are being routed.
Q. Will I need to have IPv6 in order to access sites participating in the World IPv6 Launch?
A. No, you will fall back to IPv4.
Q. What should I do if I cannot get to Cisco/Google/etc on the World IPv6 Launch?
A. In most cases, you either just fix your IPv6 issue or disable IPv6 in order to fix it. The troubleshooting sites (especially the ARIN wiki) can help. If your browser is using Happy Eyeballs, then there is no need to disable IPv6; it will do the magic for you.
Q. What does Happy Eyeballs do for you?
A. The Host simultaneously attempts initial connection using both IPv4 and IPv6. There is a fast response, even if the IPv6 (or IPv4) path is down. Since users are no longer trying to disable IPv6, it reduces IPv4 usage (reduces load on CGN).
Q. Who supports Happy Eyeballs?
A. Happy Eyeballs became RFC 6555 and is widely adopted. It is enabled by default in Firefox 13, Chrome, Apple iOS 4, and OSX 10.7. Everything interesting, except Internet Explorer, already supports Happy Eyeballs, or will do so very soon.
Q. How will my devices get IPv6 addresses assigned? Do I need a DHCP server?
A. SLAAC – if there is an IPv6 capable router, they WILL get an address! There is no DHCP needed except to pick up DNS server info, but it will share that info as learned from DHCPv4.
Q. What is the easiest way to get IPv6 connectivity set up for my company?
A. Use your ISP if they have it. Otherwise, use a free tunnel from a Tunnel broker. LISP is also an option. See the note on the support forums.
Q. Has Cisco designed any certification for IPv6?
A. No, there are no specific IPv6 certifications. However, we have integrated IPv6 as part of most existing certifications (CCIE, for example).
Technology Specific Questions
Q. Is HSRP IPv6 global address supported? I tried to enable Global unicast as an IPv6 address. Unfortunately, I do not have that option. However, according to the Cisco site, I should be able to set the global unicast as an HSRP address. Any reasons why?
A. At this time, only the Catalyst 6500 platform supports global IPv6 addresses for HSRP (supported as of 12.2(33)SXI4).
Q. Is OSPFv3 (OSPF implementation for IPv6) supported in Cisco Firewalls?
A. Unfortunately, it is not yet supported. Contact your account team in order to make sure that they are aware that you need this feature.
Q. What is the difference between IPv6 Multicast Address and Anycast Address?
A. A multicast packet goes to multiple destinations and is part of a reserved block of addresses in order to indicate special handling. Any anycast address is a unicast address that is routed to the nearest of a group of equally qualified devices. DNS is a common application of anycast. There are also IPv4 anycast addresses. For example, Google's 18.104.22.168 DNS server is an anycast address. The IPv6 multicast address block is FF::0/8. Anycast addresses are assigned from your regular unicast address space.
• Deploying IPv6 in Broadband Access Networks- Adeel Ahmed, Salman Asadullah, John Wiley & Sons Publications®
• Deploying IPv6 Networks- Ciprian Popoviciu, Patrick Grossetete, Eric Levy-Abegnoli, Cisco Press®
• IPv6 Security- Scott Hogg, Eric Vyncke, Cisco Press®
Hi, right now as i type this i'm currently on the Cisco DevNet sandbox using the SD-WAN sandbox, the main reason why i'm using it at the moment is to play around with REST-API calls to it. However no matter what call i try to do i always receive this back...
Hi All,I have a device that need to SSH at the other end. inbetween a CISCO router. Inside router i have done the NAT. All work fine and I can do ICMP ping from my end to device. It NAT inside CISCO router and reach the device then return the reply. ...
Hi,I am new to REP and working on a critical site that has multiple REP Segments. Currently the REP Admin vlan is vlan 1. For network improvement/security reasons this needs to be changed with the Admin vlan command. If I start at the end...
Hello, We have a 3750 stack and an ip on it when we MTR to that ip we dont see any packet loss but once we get past that to a server for example connected directly to that 3750 we are seing packet loss of around 10% on the 3750. There are no CRC...