Google warns of World IPv6 Day

Google is warning users of tomorrow’s test of IPv6, and more importantly of the fact that current IPv4 addresses have been depleted. I was only able to see the yellow banner in Linux running Firefox4, it never appeared on my Windows 7 machine. Google warning of IPv6 testing on June 8th.

While the banner is sure to cause some discussion among the non-networking crowd, I wish Google had included a link to more information. Instead they only include a link to test a users internet connection for IPv6 readiness. I don’t think the average user understands that their ISP is responsible for providing IPv6 connectivity, or of the problems that currently face IPv4.

I will give Google credit for starting the conversation. Hopefully, tomorrow there will be a lot of companies asking themselves what they must do to be ready for IPv6. Enterprise must lead IPv6 adoption, because as we all know, carriers are more than happy to sit on their butts as long as no one complains. The fact that so many ISP are considering CGN is a perfect example of that.

The velociraptor died after choking on a rib bone, so creating IPv7 is out of the question

OK, I admit it. I’ve had my head stuck firmly in the sand for almost 11 years. 11 years ago, to the month, I was sitting in my first TCP/IP class. I had fought through the first two days of class feeling mentally exhausted. I was finally beginning to wrap my head around IPv4 and variable length subnet mask. In fact, I was understanding IPv4 well enough that I could help my fellow students decipher the statements coming from our newly minted (and very proud of it) CCIE.
I was feeling pretty good about myself, and may have started to strut, just a little, as I moved from desk to desk, helping other students.
I should mention now, that I’m fairly quick on the up-take. I’m not bragging, simply stating that I meet the minimal requirements to be a geek. For some reason, I had really struggled with IPv4, so once I felt like I had a firm grasp of the concept, I was feeling pretty good.
My CCIE instructor, from his seat of power, saw a little pride develop in his class as more people caught the basics of VLSM. He, in the ultimate wisdom which comes with that coveted CCIE number, decided it was time to strangle those good feelings until they were most certainly dead. He did so, by launching into a 30 minute diatribe of how IPv4 would die a “quick death” and how IPv6 would take its place.
I’m sure you can imagine the look of horror on the faces of the students in the room. He certainly saw it, and fed off the fear as he blew through the broad topic that is IPv6. He delighted in mentioning that every device would have multiple IP’s, that each IP would be part of a different subnet. He threw out new words like anycast to a group of people who barely understood muilticast and unicast.
Wait, what?
In 30 minutes, he convinced three students that IT was not really the field they wanted to pursue, and the rest that IPv6 was EVIL. I was so affected and confused by that 30 minute rant, it took me five years before I had a practical understanding of subnetting IPv4 networks again.
Since that time, I have done my best to ignore the existence of IPv6. I used the fact that vendors were still releasing new products without IPv6 support as a reason to keep my eyes and ears firmly closed.
<My IPv6 Rant>
I believe that when IPv6 was being created someone said, “Yes, we COULD do that, but SHOULD we do that”. The rest of the attendees sat silently as he was taken from the room, and forced to watch his organs being fed to a genetically engineered, but very bored, velociraptor. The group then hired a soothsayer to read the velociraptor droppings, which gave us IPv6, reality TV, and the song “Friday”. The velociraptor died after choking on a rib bone, so creating IPv7 is out of the question.
</My IPv6 Rant>
With that said, IPv6 is here to stay, and it’s time for us, as Network Engineers, to get on board. We can’t complain about NAT64, without being willing to make the commitment to IPv6. When new protocols like TRILL are brought up for discussion, it’s easy to get excited. TRILL takes something that we already know (IS-IS, L2, etc) and simply builds on it. It is also transparent to layers 4-7, so it doesn’t affect non-network types.
IPv6, causes us to backtrack. It changes all of the rules. It’s not just IPv6, it’s new routing protocols, DNS, application stacks, etc. We have to forget what we learned in IPv4, and relearn it for IPv6. Server admins and developers will also have to update their skills. It’s painful.
With that acknowledged, we can’t put off learning to subnet, route, and filter IPv6. It’s time to begin examining IPv6 routing protocols, and buying equipment or ordering circuits which don’t support IPv6 should be out of the question. Yes, it does feel like starting from scratch. Yes, you will have to learn every protocol that you thought you knew all over again. Yes, IPv6 makes everything more complicated.
System Admins and developers can’t support IPv6 until we do. We must move forward, so that they can move forward.
Most network engineers agree that NAT is a poor solution to the problem staring us down. There are only a few other options. We can upgrade our skills, beginning the long arduous task of becoming experts in IPv6. We can ignore the change, until we are required to upgrade; then deal with entire IT teams being unprepared, learning on the fly, while implementing poor solutions in the near-term. Finally, we can make the same choice that those three classmates of mine did. “Maybe networking isn’t for me, I’ll go do something easier, like lion taming.”

Openflow, Merchant Silicon, and the end of the reign of King John.

Early this morning, I finally had an opportunity to listen to the latest episode of Packet Pushers Podcast.

In the podcast, the guys discuss Openflow and the impact it could have on the networking industry. One of the points mentioned in the podcast was that Cisco is apparently using merchant silicon in the latest 10GB Nexus switch, the 3000. I was shocked when I heard this, and had to do a little research to verify. Sure enough, it seems that Cisco’s latest Nexus switch is built on Broadcom chipsets. Wow.

Let me say that again…Wow. To recap, here is my favorite Cisco blog post regarding Cisco and merchant silicon by Douglas Gourlay, an ex-Cisco Senior Manager of Product Marketing.

http://blogs.cisco.com/datacenter/on_merchant_silicon_and_mowing_my_yard/

To quote the post:

Do major automobile manufacturers outsource engine design and development to other firms? Of course not, they design and build their engines. Do manufacturers of more consumer goods like lawn mowers outsource their engines? Absolutely, they go to specialized engine manufacturers because the core value of what they offer is either a certain price point, or the value is not tied to the engine. So the question then – is do you want to ride to work or school in a car, or on a lawnmower?

Ok, so if Cisco is using merchant silicon in their Nexus line, it seems to me that the course adjustment that Big John emailed his employees about last week wasn’t the beginning. Maybe John was trying to answer rumors that had already started within Cisco’s ranks. Change was in the air, questions were being asked, and it all had to be addressed.

What would cause such a shift in Cisco? Is it possible that Cisco already realizes that being faster is no longer relevant in an age of Openflow, TRILL, IPV6, etc. etc.? There is no doubt that Cisco has felt the pressure from HP, Juniper, and other vendors. In fact, my current role is in a company that made that jump from Cisco to HP and Juniper when Cisco tried to sell Nexus 7K’s when 4507’s  or 6509’s would have been the better solution. Cisco didn’t just lose a customer here, Cisco made enemies. (I get scowls when I mention Cisco.)

Is it possible that Cisco realizes that the days of huge profit margins on every device it sells are coming to a close? Is it possible that maybe, just maybe, Cisco realizes that it’s not the only game in town?

For years, people bought Cisco for the additional features that Cisco offered. PAGP, ISL, EIGRP, LWAPP were all answers to problems that no one else had addressed. They were good answers at the time, and all led the industry standards by a couple of years. Now, the alternatives 802.1Q, LACP, OSPF, and CAPWAP have replaced those proprietary Cisco protocols. Looking at the environment now, I don’t see any areas where Cisco has a unique answer. Either the networking community has a solution (Openflow, TRILL), or each vendor has their own unique solution to the same problem (Qfabric, Unified Fabric).

Let’s look ahead 3 years. If an engineer has the option of buying products from Cisco which cost a lot more, and must be managed individually, or buying products from a range of vendors that all must compete in a cost effective manner, and all of which support unified management through Openflow, and all of which have the same features, which would he choose?

Two closing thoughts:

Apple is trying to teach the tech world a lesson: specs alone doesn’t make a better product. For Cisco to compete, they have to focus on features that answer real world problems, not imaginary scenarios. IPv6 and TRILL vs. Who really uses an ASA for deep packet inspection on a regular basis?

Cisco is a very big ship, and it will take a long time to turn. Watching from the shore, we have only begun to realize that it is turning, and have no idea where the new heading points.

Cisco is SCARED! Why Cisco won’t release an emulator.

Greg Ferro posted on his blog another plea to Cisco to play nice and give network engineers tools for testing, verifying, and learning new technology. If you’ve missed the recent debate on the matter, it’s OK. Crawl back under that rock, you won’t miss a thing.

I generally read Greg’s posts while nodding my head like some sick bobble headed doll, with an occasional grunt in agreement. However today, my head stopped bobbing when I realized something…

Cisco is AFRAID of the virtual switch/router.

Let that sink in for a minute.

I know what you’re thinking. “They don’t have anything to be afraid of. That’s crazy talk.”  I’m sure that people said the same about Dell and HP when ESX was first announced. “They don’t have anything to worry about. No data center could ever virtualize all of their servers. That’s just crazy.” Only, it did happen. Right now I am sitting just a few hundred feet from 100 servers that would be over 500 servers if it wasn’t for vmWare. Think of the lost revenue to Dell and HP.

But, you say, “what about the Nexus 1000v”. What about it? Cisco had already lost sales because all of those virtual servers didn’t need individual switchports. That was Cisco’s way of getting some of that revenue back. It wasn’t about extending network engineer’s control into the virtual environment. It was about lost port revenue.

Imagine with me for a moment. What would happen if you could virtualize the Edge and Core layers of your network all onto a single HA cluster. (Maybe a couple of Dell or HP servers.)

Firewalls, Check
Routing, Check
IDS, Check
VPN, Check

Where is the need for 10GB, 40GB, 100GB, TRILL, or Fabric Path? What about all of the other technologies that Cisco will sell us over the next 10 years, forcing us to replace existing hardware?

Outside of the HA cluster, you would need a couple of switches for Distribution, and you would need your normal Access layer switches, but how many components of the network would be cut? Not only routers, firewalls, and switches, but adapters, redundant power supplies, wireless controllers.

It’s already been done. Look at Cisco Call Manager. A router, switch, and server that do the work of racks and racks of PBX equipment.

“But, I just want them to release it so that I can test.”

Cisco has three choices: 1. Stick fingers in their ears and hum loudly. (Current tactic) 2. Release a good virtual network platform, and wait for everyone to ask, “wait…why can’t we virtualize this for real?” 3. Release a crippled, barely working virtual platform, and then get derided for their poor product.

No matter how Cisco looks at it, they lose.

Suddenly I am asking myself. After IPv6, what is the next big thing to happen in networking? Could virtualization change networking the way it changed servers?

Is Cisco getting back on track?

Cisco’s big-man-in-charge, John Chambers, sent out an email to all employees this week, which outlined a few important things:

-Cisco has lost focus
-Cisco was caught off guard by certain movements within the Networking community (openflow, new products from other vendors, etc)
-Cisco makes it difficult for new product to make it to market
-Cisco has to focus on the core business components, rather than continuing to diversify into low margin consumer markets
-Most importantly, Cisco shareholders, employees, and customers are not happy with the current direction that Cisco has taken

The message is a great read, and gives me hope that Cisco can get back on the ball, and address some of it’s core issues. Kudos to the Cisco team for taking a hard look at where they are, and making decisions to correct their wandering trajectory. Here’s hoping they follow through!

http://blogs.cisco.com/news/message-from-john-chambers-where-cisco-is-taking-the-network/

Microsoft meets the first snag in plan to purchase IPv4 addresses

As you should now be aware of, Microsoft is planning on hoarding purchasing a large huge block of IP addresses from Nortel. Now ARIN chief, John Curran, has made it clear that if the plan does not meet the current ARIN requirements for transfers, the IP address space can be reassigned. Here are a couple of relevant quotes:

Companies that are allocating their address to a third party can ask for compensation if they want to, he said. However, the acquiring party is required to show an immediate and appropriate need for the addresses, he said.

Existing transfer policies allow up to 12-months worth of address space to be transferred from one entity to another, he said.

So, that brings up the question, can Microsoft show a need for 666,000 in the next 12 months?

Link: http://www.computerworld.com/s/article/9215091/IPv4_address_transfers_must_meet_policy_ARIN_chief_says

Microsoft sets the price of IPv4 space post IPocalypse.

The resellers market for IPv4 address space has just heated up. Microsoft has puchased 666,624 IPv4 addresses from the burned out shell of Nortel for the tidy sum of $7.5 million dollars. That works out to a little over $11 per IP address. That’s pretty expensive when you consider how much Nortel paid originally. This also make me worry about how quickly prices will inflate on the remaining address space that is floating around out there.

On another note, I guess Microsoft realized that since they had built poor, if any IPv6 support into their products for so many years, they better make certain they have enough v4 addresses.

via http://downloadsquad.switched.com/2011/03/25/microsoft-buys-666-624-ipv4-addresses-from-nortel-for-7-5-milli/