Friday, December 11, 2009

EIGRP Queries and SIA's


Today I learned a little more theory behind DUAL and it's process, along with a few ways to manipulate the way DUAL queries are sent. The EIGRP DUAL process is very efficiant and can be very scalable in larger networks but only with proper planning. When there are houndreds of routers in a given network, the convergence time for EIGRP can be much longer and more complex than needed. To eliminate convergence time as much as possible, it is smart to plan networks in such a way that there are no more than 7 hops between two routers on a network. One good way to cut down on convergence time is to summarize networks along with using default routes on remote routers that connect to the core. I also learned a lot how SIA (Stuck-in-Active) works and how it can increase convergence time quite a bit on redundant networks. The next few days I will be finishing up my EIGRP study and will be going through mainly labs since most of the theory is finished for this chapter.

Monday, December 7, 2009

EIGRP Bandwidth and Authentication


Today I spent most of my time learning the EIGRP Bandwidth theory along with authentication. Math isn't one of my most favorite subjects but in order to understand route manipulation you have to understand the mechanics behind it. EIGRP is setup to use only 50% of available bandwidth for it's routing by default. So as the picture shows, you will have to manipulate the EIGRP topology in believing that a link is slower than it really is in cases such as Frame Relay multi-point and point-to-point. I spent the last part of my studying time configuring MD5 authentication between two routers by enabling MD5 on the interface and configuring my key chains on both sides of the routers. I won't have time to study like I want to due too my shift beginning tomorrow but I should be wrapping up my CCNP EIGRP studies next week. It's hard to believe that I'm really working on obtaining such a prestigious certification but I'm focused so i don't ponder to much on it. So far the BSCI is doing a very good job of answering a lot of the questions I had while studying the CCNA.

Sunday, December 6, 2009

EIGRP DUAL and Basic Configuration


Today I spent time learning exactly how DUAL works within EIGRP by reading through the CCNP Study Guide along with creating lab scenarios and watching EIGRP debugs. Seeing it in action through the debugs and reading the study-guide book examples gave me a way better understanding of exactly what's going on when the EIGRP AS topology changes. Using the EIGRP Packets it sends EIGRP Queries, Updates, and Replies when the topology changes. I have a fuller understanding of why this protocol is considered a hybrid protocol .Because it does distant-vector tendencies like split-horizon and hold-down timers while it does link-state tendencies such as using the metrics too determine the best path rather than hop count. I finished up tonight by going through one quick simple lab regarding manual route summarization based off the examples that were provided in the study guide. I've also been practicing setting up IP host translations for telnetting between devices. It's a little bit of a pain setting it up but once you do it's a beautiful thing being able to quickly telnet to the device you want rather than having 10 console screens open!

CCNP EIGRP Theory

Yesterday I spent most of my time in the CCNP Study Guide book learning about EIGRP in a little more detail. Such things as the metric used for calculating routes:
\bigg [ \bigg ( K_1 \cdot \text{Bandwidth} + \frac{K_2 \cdot \text{Bandwidth}}{256-\text{Load}} + K_3 \cdot \text{Delay}                       \bigg )          \cdot \frac {K_5}{K_4 + \text{Reliability}} \bigg ] \cdot 256

EIGRP associates five different metrics with each route:

K1 = Bandwidth modifier

  • Minimum Bandwidth (in kilobits per second)

K2 = Load modifier

  • Load (number in range 1 to 255; 255 being saturated)

K3 = Delay modifier

  • Total Delay (in 10s of microseconds)

K4 = Reliability modifier

  • Reliability (number in range 1 to 255; 255 being the most reliable)

K5 = MTU modifier

I also spent time learning about the neighbor table, topology table, and routing tables EIGRP use to determine the best route or Successor route. I spent time going through the 5 types of packets EIGRP uses for making this routing protocol works. Hello's, Updates, Queries, Replies, and ACK (Acknowledgements) are all used in conjunction to make EIGRP work. Today I'm going to be going through many different EIGRP configuration scenarios along with learning more about the DAUL algorithm.

Tuesday, December 1, 2009

CCNP BSCI Chapter 2 Lab


I spent a total of 4 hours configuring, troubleshooting, and manipulating what has to be the biggest Cisco lab I've worked so far. I used a total of 11 routers, 1 switch, and a connection to the outside world. Following the book instructions I spent a good portion of my time configuring two back bone routers and a Cisco router as a Frame-Relay switch. I implimented a lot of commands I wasn't to familar with including some BGP commands and even some multi-cast commands. Everything some how came up as expected with not to many real hitches. The purpose of this lab was to pretty much bring everything I learned in Chapter 2 together. I had 8 routers that I ended up setting RIPv1, then moved to a default route, then enabled ip classless, enabled RIPv2, and last but not least I practiced scalibility by manually summarizing routes on my edge routers. All in all it was a good learning experience and I'm slowly but surely starting to see my knowledge grow.

I have 7 more chapters in this first book and I'm estimating that it's going to take between 2-3 weeks per chapter. Once I'm done with this book then I'm going to buckle down and prep for the exam by going through the Cisco Exam-guide along with the Lab manual. Hopefully by then I should be prepped enough to take this exam, then repeat this process times 3! As you can see there's a long way to go and a lot of dedication needed for any Cisco certification, especially the high level certs.

Sunday, November 29, 2009

RIP Protocol and Floating Static Routes


Today I read up on topics related to the two different RIP versions 1 and 2 which was mostly review for me. There was some good stuff such as using both RIPv1 and RIPv2 on a network which i of course created a lab for. I also went over topics related to classful and classless routing over a network and how routing updates are summarized over Dis-contiguous or over different major classful networks. It was a good read and confirmed my suspicion that you can subnet within a classful network using RIPv1 as long as that subnet number is consistent throughout that network. However the network will still be summarized to a /16 when crossing major network boundaries.

I finished up today by learning about Floating Static Routes which was A LOT simpler then I imagined it to be. A floating static route is primarily used when a dynamic routing protocol fails or when you have a WAN link fail and would like to have an automatic fail-over route say through anISDN/DSL backup connection. A floating static route is configured by setting the administrative distance (AD) on a static route higher than the dynamic protocol (AD). By doing this the static route "floats" above the dynamic protocol since best routes when using mutli protocols uses AD to determine the best protocol to use. By default a static route has an AD of either 0 or 1 depending on the configuration. The higher the AD the less likely it is to be used over another protocol as indicated in the picture above.

Saturday, November 28, 2009

Static Route and ODR Overview


Well today was my first real study session on CCNP, from the get go I could tell there is a higher level understanding needed if I'm going to pass this certification. The BSCI goes a lot deeper into routing than anything I've came across so far and just through the first few pages I already set up a few labs pertaining to the information I was reading. I went through setting up a simple static router network which was easy enough followed by setting up a static default lab which wasn't bad at all either. I then finished up by learning and creating a lab for On-Demand Routing (ODR) which is really a Cisco Prioperty way of setting up a cisco network by using CDP rather than a true Dynamic Routing Protocol or setting up Static Routes. All in all not to bad, I didn't get as much as I wanted done today because I came down with a cold and I'm sitting here hoping it doesn't turn into a fever or the flu. Tomorrow I am going to go over some RIP topics and labs which shouldn't be too bad hopefully but we'll see!