From the blog.

Managing Digital Racket
The more I tune out, the less I miss it. But that has presented me with some complex choices for a nuanced approach to curb
Complexity – My Friend, My Enemy
Over my years of network engineering, I've learned that the fewer features you can implement while still achieving a business goal, the better. Why? Fewer

InternetworkExpert.com Mock Lab #2 Summary – UPDATED: 12/16/2007

1,163 Words. Plan about 7 minute(s) to read this.

I just got through an InternetworkExpert.com mock lab. First, some comments about the product.

  • I have no complaints thus far. I haven’t seen the score sheet yet (won’t be for up to 2 business days). I’ll add comments about that when I get it.
  • Signing up was easy. Buy a mock lab for a timeslot of your choice (depending on availability) in their online store. Pay with a credit card.
  • Upon receipt of payment, you get an e-mail back from IE telling you what rack you’re on, how to telnet to it, how to manage the console server if you need to, etc.
  • You have 11 hours of rack time total. That gives you plenty of time to print out the mock lab documents, take a half-hour break for lunch, and even putter around after you should be done, if you want.
  • IE makes the PDFs of the lab itself, lab network diagrams, and initial configurations available on the day of your mock lab. When I logged into my IE account at 6:30am, all my mock lab PDFs were a click away.
  • I had no equipment problems, other than one router had 2 fast ethernet interfaces, when the documentation said gig. No biggie.
  • Connecting to the rack was via the Internet using telnet. Nothing magical about it at all. I was able to keep my telnet sessions live all day without getting kicked off of the equipment due to idle timeouts.

And now some comments about the mock lab itself.

  • I don’t know what level of difficulty this mock lab was. Based on my experience with the NetMasterClass.com practice scenarios, I’d call today’s lab about a 7 out of 10. Maybe an 8, not because the tasks were that hard, but because of the volume of work. There was a LOT of work to do. A lot. The sheer task count was part of why it was as hard as it was for me. Given enough time, I could have completed more tasks than I was able to.
  • I opted to restrict myself to 8.5 hours (which included a half-hour for lunch), to make it just as close to reality as possible.
  • I did not finish QoS. I did not even start multicast.
  • I was unable to bring up a PPP over frame-relay circuit. This was an absolute killer – I lost 2 points not being able to get this working. What’s worse, this connection was to a pre-configured backbone router that I had no access to, so I couldn’t fudge the connection just to get the circuit up. That cost me big, because that meant there was an EIGRP task I could not complete (3 points), and an IGP redistribution task I could not complete (3 points). So that was a whopping 8 point deduction. Plus, it ate up a lot of time while I was troubleshooting.
  • I burned more time than I’d planned on what they called a “backup circuit” configuration. I decided to go with weighted static routes on one side, and PBR using “set ip default next-hop” on the other. I’m still not sure if I broke any lab rules doing it the way I did it. It works, but I may have broken a rule by using PBR.
  • I also got stuck on some of the BGP tasks. Getting the infrastructure together was easy enough – neighbors, a route-reflector, a confederation, etc. But then tweaking local pref took too long for me. And I killed a ton of time on a summarization task that I just couldn’t get working, and I’m still not sure why. And then I was drawing a blank on how to set MED, and whether “metric” meant “MED”, and then whether high or low values were preferred, and then whether or not local pref is more important than MED as far as BGP best-path is concerned. (MED can be set with “set metric”, smaller values are preferred, but remember that local pref trumps MED.) So setting MED and local pref are 2 things that were easy to do, 5 minute tasks a piece, that I stumbled through burning up all kinds of time.
  • I was able to gain a bunch of points quickly going through a lot of the little tasks like configuring SNMP, some security features, and so on.

So how do I think I did? If I earned the points for every task that I think I did right, I got a 73. Reality is that I probably got dinged all over the place for all kinds of stuff. So – maybe a 50? Maybe worse? Hard to say. I had almost no time to check anything. I did some reachability checks, but I think I might have had some challenges there I never got back to. I did manage to check my frame-relay maps towards the end, and they were all good. I gave up 7 points on the multicast, which I didn’t even start, and 7 more points on the QoS from tasks I didn’t get to. Plus 802.1x that I never quite completed, 2 points. And then the 8 I lost because of that PPP over FR circuit. And some other stuff I’m forgetting at the moment.

So was it worth $99? Oh, yeah. Absolutely. $249? Um, not sure. I guess it’s all what you want to get out of it. I didn’t find out anything I didn’t already know. PPP over FR has been on my list of “stuff I know I need to work on”. And it cost me 8 points. BGP best-path algorithm memorization is also on that same list. Details I forget, and they cost me time. I know I’m not quite as fast as I need to be. And I know there are some technologies that I need to be better with. We’ll see what the grade report looks like in a few days, but I think I’m right where I thought I’d be. Getting closer, but not ready for that first attempt.

I’d say my readiness for the actual lab is maybe a 6 out of 10.

UPDATE: 12/16/2007

I downloaded the solution PDF and started reading through it. I think I need some time alone to cry for a while. There were a number of little things I just didn’t pick up on while doing the lab. Many of my approaches were CLOSE, but not close enough, I have a feeling. Or my approaches solved the problem, but looked rather different than the solution. It will be interesting to see how they grade it, but it’s not going to go well at all, not at all. Oh, boy. The good news is, I have a lot of things I can review and learn from this whole experience. The bad news is that I’m not as far along as I’d been hoping, if this mock lab is comparable to the difficulty of the actual lab. If I tried to take the actual lab today and got an exam like this mock IE lab, the proctors would chew me up and spit me out. And then laugh.