IPv6 II: Addressing Types

DIFFERENT (aka, not the same!)

As is the case with anything new and unfamiliar, IPv6 can appear  strange, hard to understand, and downright confusing!  Often, when learning new content, it helps to compare or even anchor the concepts in something more familiar, as depicted above with the photo of two very different dog breeds.  In this example, the contrast is plainly visible; one dog towers over the other, and each have variances in coloring, build, and a dozen or more differences.  Even so, both are still dogs, with four legs, a keen sense of smell, ability to learn, and even “man’s best friends.”  Even though the differences may appear vast, the similarities are much greater in number.  This is precisely the case with IPv6, as it represents a different “breed” or “species” of the Internet Protocol, but with many similarities to its predecessor; think of IPv6 as IP 2.0!  One are, which we shall consider now, has to do with addressing types in this new version of the Internet Protocol, namely the types of multicast, anycast, and unicast.

Multicast

In IPv4, multicast addressing was encompassed the Class D space of 224.0.0.0 – 239.255.255.255, and used to address multiple hosts.  Broadcast addressing, on the other hand, sent messages to every host on the LAN/VLAN segment in question.  As mentioned previously, IPv6 eliminated the use of broadcasts, in favor of multicast functionality.  Formatting of multicast addresses in IPv6 requires use of the prefix FF, and the basic mechanics are the same as IPv4.

Anycast

To even seasoned engineers, the concept of anycast addressing can sound confusing and rather contradictory, as it calls for the use of identical addressing on multiple devices.  To be frank, I myself struggled with this for years before inadvertently stumbling on an explanation that made the idea much clearer in my mind.  Anycast is often used on servers to provide resiliency, such as is the case on DNS servers, and often used by service providers for this every purpose.  IP routing directs the requesting device to the nearest server using standard routing metrics which can provide load-sharing as well as failover capabilities.  To better understand this, think of the process by which an end-user utilizes a SmartPhone or GPS device.  The user enters the name of a familiar grocery store, for example (Safeway, Albertsons, etc.), and may get half a dozen responses, along with the distance to reach each one.  ALL the stores have the same name, but each is a different distance from the standpoint of the user, who normally selects the CLOSEST.  This is precisely how anycast addressing works.

Unicast

Unicast addressing in IPv6 operates the same way as it does in IPv4, but has a number of subtypes, which will be addressed in the next article.

– Joe (Twitter: @jjrinehart)

CCNA Challenge Lab!

As I wrapped up another class on ICND2, I created a fairly comprehensive lab that covers many of the major topics on the ICND2 exam.  Take a look, I certainly hope it helps with the preparation process for the CCNA!

ICND2 Course Labs – Comprehensive Lab Packet Tracer Version

– Joe

Twitter: @jjrinehart

NEW Lab Scenarios

I recent have been leading another set of students through the ICND1 & 2, and created a series of labs designed to reinforce the learning concepts in the course.  These have been created in Packet Tracer, which is intended for students in the Cisco Networking Academy program, though they can be easily created in other simulators and on actual equipment.  I am posting the lab guides for your use, I am unable to post the configuration files.  If you have any questions, let me know.

Joe (@jjrinehart on Twitter)

ICND2 Course Labs – Spanning-Tree Protocol

ICND2 Course Labs – RSTP Labs

ICND1 Course Labs – Switch Configuration

IPv6 I: Networking’s Senior Citizen, the IPv4 Protocol!

How OBSOLETE Looks!

I grew up during the 1970’s, a period well before the advent of many of the technologies that we virtually take for granted today.  Back then, you often shared a phone line with another neighbor (called a party lin), there were no answering machines, and personal computers simply didn’t exist.  In addition, if you missed a movie at the theater or on television, you were simply out of luck altogether.  Today you carry your telephone with you, can stream movies on that same device, and carry on real-time video conversations with someone literally on the other side of the planet!  The point of this “flashback” is simply to highlight that time and technology marches on, leaving some things in the dust of obsolescence.

This is precisely the situation in which the beloved Internet Protocol (referred hereafter as IPv4) finds itself in today.  Granted, it has not passed off the technology world stage (nor will it anytime soon), but is displaying signs of its definitive sunset years.  In a lot of ways, IPv4 became a victim of its own success, after beating out other routable protocols such as IPX and Appletalk, among others.  Globally routable address space was becoming rapidly depleted, Internet routing tables were ridiculously large, and yet the proliferation of devices continued unabated.  As a result, in the 1990’s, short-term solutions such as Classless Interdomain Routing (CIDR), Private Addressing, and Network Address Translation (NAT), were implemented, but the long-term solution was a newer version altogether.

Enter IP Version 6, typically just called IPv6, designed from the ground up to address all of the shortcomings of its predecessor, the venerable IPv4 protocol.  You may immediately wonder why the version number jumps from version 4 to 6, instead of simply 5, and the answer is simply that IPv5 was experimental and never actually released, similar to versions 1-3.  While not exhaustive, here are some major improvements brought to you exclusively by IPv6:

• Increased Address Space: Expanding the addressing from 32 bits to 128 bits allows for 340 Trillion addresses!
• Address Assignment Features: Hosts can calculate their own addresses, as well as take advantage of new DHCP features
• No More NAT: Due to the vast availability of addresses, having to implement private addressing and Network Address Translation is simply no longer necessary
• Elimination of Broadcasts: Broadcasts are a necessary evil in IPv4, but no longer used in IPv6, instead using multicast-driven protocol mechanisms
• Integrated Security: When the now-familiar IPsec protocol first came out, protocol analyzers would label the packets as IPv6, partly because the mechanism was originally designed for it, and is native in IPv6.

As you can see, there are many new mechanisms and features in this up and coming version of the Internet Protocol, which is certainly and inevitably in our collective future.

Next time we will consider address types!

– Joe

 

WANs IV: Route Globally, Act Locally

GLOBAL Route-Sharing

As we talked about in our last discussion, it’s really amazing that the Internet functions as well as it does.  Anyone with access on one side of the planet can instantly send information to someone on the other side, including email, voice, video, and many other types of data.  Ever wonder how it all manages to work so well most of the time?  The answer, which is not usually covered in CCENT/CCNA discussions, is the Border Gateway Protocol, currently at version 4.

When I first started reading up on networking protocols, I was impressed at how OSPF could communicate vast amounts of accurate reachability information to devices in its domain.  The problem, though, is that comparing even a large, 2000-node network, for example, is far different from the Internet.  Communicate the intricate details about every conceivable route (including host routes) would make Internet routers capable of cooking eggs from the sheer heat, because of the massive routing tables and computations involved.

This is where we can talk about an important but sometimes overused networking term—scalable.  Scalability is the ability for something to grow in a controlled, measured fashion, rather than haphazardly or too rapidly.  To allow the Internet to be scalable, routing has to be simplified in some form or fashion, and this is the unique ability of BGP as an exterior gateway protocol (between autonomous systems, namely devices under a common administration).

Routing across the Internet is a lot like the old line you hear from military personnel in most movies—“That information is on a need-to-know basis” (and usually implies that the person being told that does not need to know).  For example, the Internet Service Provider that I first worked at had a block or range of Class C addresses from 216.145.0.0 to 216.145.31.255, and were allocated to various customers that they serviced.  Advertising out all 31 separate routes, or even worse, even smaller subnets, would have created a minimum of 32 entries if not more.  Instead, because of the beauty of BGP, they advertised a single entry, 216.145.0.0/19.  This greatly reduced the size of the potential routing table, and if you multiply that out across the world, you can see why this works so well.

Another important concept is peering, where service providers interconnect and exchange their routes.  Originally this took place at public exchange points, but companies like AT&T and Verizon connect directly to one another to accomplish that, called private peering.  Most of these entities also have rules about the sizes of network advertisements that they will accept as well.

Next time we will start exploring the newest version of the Internet Protocol, IP Version 6!

– Joe

WANs III: The ULTIMATE WAN (aka, the Internet)!

The Internet (GREATLY Simplified)!

As promised, now we want to consider the most extensive WAN ever created–the Internet!  Begun as a public project, the Internet is not actually a single monolithic network, but rather a collection of INTER-Connected NETworks (notice the way the terms break out).  If you think about all of the logistics involved with integrating the various components, geographies, devices, and access methods, it’s a wonder that it even works at all!  When I started in the networking business back in 1998, the hierarchy was much simpler, with large backbone carriers (AT&T, UUNET/Verizon, etc.), and regional providers, with smaller Internet Service Providers connecting to smaller customers. Needless to say, a lot has happened with the public Internet, with many more developments on the horizon.

In the “old days” there were very limited ways to access the Internet, most often through dedicated access, or through dial-up.  Today there are many ways to get to Internet resources, including cellular (typically 3G), 4G, DSL, cable, and so forth.  For the purpose of simplicity, we will narrow the types of access to the four most common types, as follows:

1. Dedicated Internet Access: Probably still the “gold standard”of Internet access, dedicated access uses private-line connectivity of some type between the customer location and the provider’s Point of Presence (POP).  If you remember our discussion about private lines, this involved a telecommunications circuit, usually terminated by a CSU/DSU and can have speeds from T1/E1 up to insanely huge optical connections.

2. Dial-Up: Archaic by today’s standards, dial-up networking once ruled the information world, with big names like America Online, CompuServe, and Earthlink once considered the “heavyweights.”  An analog device called a modem (modulator-demodulator) turned digital information into analog tones from transmission over standard phone lines, usually at very slow speeds.  Originally it was terribly slow (I remember 2400 bps being the top speed), but newer techniques helped promised 56 Kbps speeds.  Dial-up was eclipsed by the introduction of broadband technologies (considered next), and is difficult to find today.

3. Cable: If you have ever had cable television, you know first-hand the amount of information that is possible to squeeze through that fairly narrow coaxial cable it is known for.  Hundreds of channels with specialized content is available at the click of a remote-control button.  In many ways, delivering Internet access across this connection is similar to just adding another “channel” of sorts into the lineup.  Boasting great speeds, it is a popular option where available.

4. Digital Subscriber Line: For years it was possible only to transmit voice conversations across analog telephone lines, including dial-up networking (which used audio tones for transmission).  Of the number of possible frequencies to transmit across an analog line, a fairly narrow amount is used for voice, leaving the rest open for, yes, you guessed it, transmitting data.  The benefit of this is being able to support voice calls and data transmission at the same time.

While this covers Internet access, there is much more to say about how the Internet communicates, which we will look at next time!

WANs III: “Pack-It” In: Frame Relay Networks

Sample Frame-Relay Network

As you can observe by the evolution of cellular phones and the Internet, technology never stands still; rather, it continues to morph, change, and improve, and often at a dizzying pace.  Such was the case with leased lines, especially with the introduction of the personal computer.  Even before the advent of the public Internet, business entities needed efficient connectivity between multiple locations, but disliked the mileage-based charges involved with private line based WANs.  To complicate matters, these links were not in constant use, meaning that when they sat idle (such as during the night), they were paying for bandwidth they were not using.  Well, why not share bandwidth, and make the whole process less costly and more efficient?  The solution was packet-switched networks, including such technologies as SMDS, ATM, and of course Frame-Relay; due to becoming obsolete, we will skip over the first two and concentrate just on Frame-Relay.

When I started out in the industry, Frame-Relay was “all the rage” because it solved several problems with private lines right away.  First, the only charge for mileage was between the customer premise and the service provider Point of Presence (or POP).  Usually this was less than thirty miles (at least in my area), which trimmed the cost substantially.  In addition, you no longer had to be restricted to only joining pairs of sites, instead you could put many sites on the network and only consume one physical port on equipment).  It also allowed customers to share bandwidth in the network, which also drove down costs.  For a time, this was considered cutting-edge connectivity.

Now on to the mechanics of how all of this works.  First, connections across the network are logical rather than physical, which is how the flexibility is achieved to begin with.  Equipment connects to the service provider, to a logical entry point called a port, sold at speeds in increments of T1 (1.544 Mbps).  Every site needs a port to communicate, but the logical point-to-point connections are created using Permanent Virtual Circuits, or PVC’s, and identified using Layer 2 WAN addresses called Data Link Connection Identifiers, abbreviated DLCI. Multiple PVC’s can terminate on a port, such as in the diagram above.  Notice that the Denver location has three PVC’s defined between itself and the other three sites, and the DLCI’s identify the specific PVC’s in use.  All of this information is sent by service provider equipment using a protocol called the Local Management Interface, which acts as a keepalive mechanism, as well as DLCI and other information.

Frame-Relay is a complex topic with many more nuances than this, but it gives you a good start.

Next time we will look at the ultimate WAN, the Internet!

– Joe

WANs II: A New “Lease” on Life!

Private/Leased Line Network

One of the sayings I tend to use when explaining network concepts is a twist on the opening words of the book of Genesis, “In the beginning was the mainframe…” because so many things we take for granted today started with that piece of technology.  In terms of WAN connectivity, the original form of connectivity took place across specialized telephone lines that carried data rather than voice conversations.  Because the end-user/customer paid the IXC for the exclusive use of the line, they were referred to as private lines or point-to-point leased lines.  As mainframes were replaced by personal computers (and networks) these lines connected to devices that convert bits of data to electrical impulses that can be transmitted across these lines for long distances.  The technical term for this device is a Channel Service Unit/Data Service Unit, or CSU/DSU, which used to be an external device but are now integrated on interface modules on routers.  This is actually a good time to introduce two more terms, namely DCE and DTE.  A Data Terminal Equipment (DTE) device is the terminal or end-point sending and receiving information, usually a computer or router.  DCE devices, on the other hand, perform the conversion between raw data and the format needed for transmission, such as a CSU/DSU or modem.  The acronym stands for either Data Communications Equipment or Data Circuit-Terminating Equipment, depending on the publisher.  Cisco prefers the latter term as a general rule.

Private lines utilize electrical circuits to create the pathway between locations, as illustrated in the diagram above; this is in contrast to packet switched networks (which we will deal with later).  Depending on the part of the world you live in, there will be differing names and capacities, such as T1 (1.544 Mbps) or E1 (2.048 mbps), and even higher speeds.  These are typically copper connections, with very high speeds delivered on fiber optic connections (OC-X).  In North America these are the T1/DS1 and T3/DS3 standards, while the rest of the world utilizes the E1/E3 standards.  These lines are charged by mileage and often very expensive as a result, although very secure since a customer has private and full-time use of the connection, although if it ever sits idle that ends up being wasted bandwidth.  The structure of the framing and line coding ca be rather complex, and something you should definitely be familiar with as you pursue your certification studies.

Next time we will look at packet-switched WANs! Don’t you just love this stuff?

– Joe

WANs I: LANs, MAN’s and WAN’s Oh My!

A Wide Area Network

I started my career in the networking industry at a small Internet Service Provider in Seattle, Washington in the United States (this was in the late 1990’s).  At that time, most end-users accessed the Internet using dial-up connectivity, and modem speeds topped out at about 28 Kbps (yes, it sounds like ancient history).  From there I went to work for AT&T, where I spent the next five or so years assisting Fortune 500 businesses connecting multiple locations together; needless to say, I spent a lot less time dealing with Wide Area Networks than Local Area Networks.

Although I have since mastered LAN’s, I still have a great fondness for WANs and enjoy building labs that simulate them. LANs are suspiciously easy to recognize, first because they use Ethernet switches, but also because they occupy a fairly localized geographical area (hence the term LAN).  Wide area networks are also easy to recognize, as they almost universally depend on large telecommunications providers (AT&T, Verizon, British Telecom, etc.) and use an entirely different set of connections to provide services with.  A term that can be confusing, however, is that of a MAN or Metropolitan Area Network, and as such, needs some clarification.

The simplest way to differentiate MANs from WANs is to look at geography once again.  LANs connect computing devices with a floor, building, or campus, but no further than that.  WANs include networks that tie together sites across significant distances, such as nationally or internationally.  MANs are networks that lie within a smaller, more specific region; in a sense all MANs are WANs but not all WANs are MANS.  Think of the concepts like squares and rectangles; all squares are rectangles but not all rectangles are squares (using the classic shape definitions).

The more technical way of explaining Metropolitan Area Networks involves a little bit of US telecommunications history, specifically when AT&T was broken up in the 1980’s.  In the new system following this “divestiture”, new regions were created within states called local Access Transport Areas, or LATAs.  Local phone companies (e.g., Bell Atlantic, Pacific Bell, etc.operated in these LATAs, and Inter-Exchange Carriers (IXC’s) created connectivity between these areas; each had to operate separately.  In this arrangement, a MAN would be between locations within a LATA, while a WAN would be between them.  Usually this would encompass a city and its outlying suburbs and such, hence the term “metropolitan.”  With regulatory changes, these distinctions are not nearly as relevant, which explains why the term MAN is far less frequently used.

Next time we will look at some of the types of WANs that exist today!

– Joe

EIGRP III: Choosing the Winner of the Miss EIGRP Pageant!

The Winner! (Feasible Successor)

Even though EIGRP is a simpler protocol in many ways, there are still important concepts to understand, one of which how it chooses routes.  As we talked about before, the lowest metric (constrained bandwidth + cumulative delay) wins the contest for the best route, without the somewhat confusing exceptions that exist in OSPF.  However, because there are elements of Distance Vector routing in this hybrid protocol, there inevitably have to be some kind of loop prevention mechanisms.

EIGRP uses two forms of the calculated metric to a network to select the best loop-free route, and at first glance they might sound identical.  The first value is called the Feasible Distance (FD), which is the complete metric from the router to the destination network.  The second value is called the Reported Distance (RD), which is the metric from the point of view of the next-hop router.  If a path is loop-free, the value of the RD will be less than the FD, expressed mathematically as RD < FD.  Paths that could cause loops will have opposite values, which will make EIGRP discard it.  That may sound simple enough, but as you have probably heard on infomercials, “But wait, there’s MORE“!

Routes that meet the RD < FD test (called the Feasible Condition) are held in the EIGRP Topology Table and the best route gets installed in the IP Routing Table.  This best route is called the Successor, or Successor Route in EIGRP, and if other equal-cost routes exist they are also flagged as successors and installed in the routing table.  One of the truly amazing features of EIGRP, however, has to do with backup routes.  In just about every other protocol, if the primary route fails, the entire convergence process for that failed route starts over to select a new one.  In EIGRP, though, another loop-free path is held in reserve for immediate use if the successor fails; this route is called the Feasible Successor.

If you have ever watched any type of beauty pageant (Miss America, Miss Universe, etc.) then you have actually seen this in action!  After all of the contestants have proceeded through the judging process, one of them is crowned the winner (Successor)!  However, the next contest in line, usually called the first runner-up (Feasible Successor) automatically becomes the winner if anything happens to the original victor–no new pageant is required!

That covers the basic principles of IP routing, next time we will start considering Wide Area Networks!

– Joe