Tuesday, July 24, 2007

IE Lab 4 complete

I completed IE lab 4 today. The remaining parts consisted of BGP, IPv6, QoS, and other miscellaneous stuff. Unfortunately, the router that I needed to run IPv6 OSPF wouldn't handle it (2501 running 12.3 IP Plus). So, I implemented it on another router but wasn't able to watch it peer with the border router. I also reviewed the configuration guide heavily to ensure I didn't miss anything.

There were a couple of shockers at the end. I never knew there was a menu command in IOS. I also had no idea about the "ip alias" and busy/refuse message commands.

All in all, I felt pretty comfortable once again with most of the lab. I still need to work on my understanding of QoS and multicast, and I need to get more comfortable with IPv6.

Now that Lab 4 is complete, per the guidelines I'm going to (re)do the core workbooks labs 1-3 before continuing on to Lab 5. I actually did the core labs 1-3 a few months back because I felt so overwhelmed by the workbook 1 labs. Thanfully, around workbook lab 2, combined with what I learned from the core and the class on demand, I finally got route redistribution (mostly) figured out and haven't had much trouble with it since. I just hope that stays.

Before I continue on with Lab 5, I'm also intending on doing the technology labs for Multicast, QoS, and the peering portion of BGP. With a little more experience with those I should really start tearing through the rest of the labs.

Sunday, July 22, 2007

Daily Update

Okay, so it's been a few days since I've studied. I actually had to take a break for my own health. You see, over the years I've gotten a case of tech neck. Back in January I purchased a sleep number bed and I haven't had a single incident since.

That was until last Tuesday night. I like a sleep number of about 65. My cat apparently does not. On that night, he preferred a sleep number of 0. In the middle of the night he managed to hold the remote down to deflate my bed completely. I woke up effectively sleeping on the floor. About two hours into work, my neck popped and I knew what was coming. Throughout the day my neck got progressively more sore, to the point that when I awoke the next morning I couldn't move my head at all. One day off work and countless Ibuprofen later and I'm almost back to normal.

I hope to be back to studying Monday night, if more important matters don't come up. It's hard enough to study for my CCIE when life doesn't interfere. But as bad as I want it, some things just make it pale in comparison.

Tuesday, July 17, 2007

Daily Update

I finished IE lab 4 up to IGP tonight. Things certainly get more comfortable and faster every day. There were only three issues I ran into.

First, I showed I need to reread udld, specifically the spanning tree loop guard functionality.

Secondly, I need to make sure to read every single subtask. This is a trend that I continually see. I intend to put more effort into ensuring I complete every task once I complete lab 4, where it is recommended that I start seriously timing myself and preparing as if it were the real lab.

Finally, I fat fingered an IP address on a tunnel interface, which ate up about 20 minutes of troubleshooting. Then, to top it off, one of my frame-relay interfaces was acting up and, even though the show frame-relay maps were fine. I still needed to do a mass reboot to get everything responding fine. I wasted way too much time troubleshooting before I initiated the reboot, eating up another 20 minutes.

All in all though, I was happy with the results. I caught on to all the trickery and had no issues knowing what needed to be done, but some mistakes hurt me in the time to execute.

I'm looking forward to hitting the BGP and other portions tomorrow night.

Monday, July 16, 2007

Daily Update

I didn't do a whole lot today. I wiped the configs on all of my equipment and loaded the IE lab 4 base configs. There was very little studying completed over the weekend as well. I was having some work done on my house and studying would have been nearly impossible with all of the noise.

So, I thought this would be a great chance to explain a little about what the CCIE means to me. In a nutshell, the CCIE is the most demanding IT certification. It requires not only the passage of a computerized exam, but also one must pay Cisco a visit to demonstrate his or her knowledge. In this eight hour lab, Cisco requires you to set up an extraordinarily complex network that would probably never be seen in the real world. They throw the book at you and give you very little time. On top of that, each attempt costs over $1000 and the average person to pass--and many never do pass--takes three attempts.

The typical estimate is that an individual who is already a professional network engineer takes about 2000 more hours of studying to pass the CCIE. This does not mean a year of work. This means 2000 hours of weekends and time after work spent dedicated directly toward CCIE material. Unless perhaps your job description precisely matches the CCIE blueprint. Since mine does not, it means a lot of time studying.

I've achieved a number of computer certifications over the years. The MCSE took me about a year to complete back in 1998. The CCNA, MCDBA, and CCNP were also completed with little effort--perhaps a month of studying for each exam. But, I have been working toward the CCIE off and on since about 2004. The first year was spent just reading every related book I could get my hands on. I needed to learn topics such as IPv6, Multicast, BGP, OSPF, and others much, much deeper than what was required to slide through the CCNP exams.

Last December I was in the fortunate position where I had a little extra money to spend and InternetworkExpert was throwing in a PS3 to boot, so I couldn't pass up the deal. Since I purchased the end-to-end training package, I have been, for perhaps the first time, consistently dedicated to the CCIE. I'm not nearly as afraid of many of the topics as I used to be, but I still have a lot to learn. I'm optimistic I can take a stab at the CCIE lab this year, if I can stay focused.

And that brings me to why I started this blog. I figure if I imagine others out there are reading my progress, I'll have to be accountable to somebody. I know this can be done. It really isn't brain surgery. It's just a matter of coming to terms with the large volume of information that's out there. I'm determined to pass the CCIE this year. Perhaps another time I'll explain why.

Friday, July 13, 2007

Lab 3 Complete

IE Lab 3 completed tonight. I learned I know nothing about RSVP and TCP intercept and could use some work on Multicast filtering. All in all, I feel lab 3 went pretty well though. My confidence gets a little higher with each lab. Unfortunately I was feeling a little burned out tonight.

My weekend is full so I probably won't get to lab 4 until Monday. Per the recommendations, I'm just going through lab 4 and spending a lot of time reading the docs and experimenting. Once lab 4 is complete, it's time to start taking things more seriously.

And then there's the issue that I still need to re-pass the CCIE written. I'm not in a hurry to retake this. For starters, the info I'm learning and finally understanding by going through the labs will help make the written studying go much better. Also, the written was recently changed to add IPv6 and MPLS. IPv6 is part of the written, so this shouldn't be a big deal. But I haven't even looked at MPLS since taking the Routing exam nearly 5 years ago. I'll probably just review and hope there aren't many questions on it.

Lab 3 BGP

Last night I completed IE Lab 3 BGP. The section was relatively easy until the last part. I was led down a path where I had to override BGP weight on an inbound path. Huh? I didn't think this was possible. Remote outbound always gets precedence over local inbound. That's why weight and local preference are above AS path and med. I was going to go down a path of advertising two /25's subnetted from the /24--longest match would definitely take precedence over weight. I know that would have worked, but if this were the real lab I'd definitely have to ask the proctor if that were allowed or not.

The more I thought about it, the more I assumed it would not be allowed, so I caved and checked the answers. Sure enough, it was something I did not know--BGP conditional advertisement. There's a command in BGP which allows you to only advertise a route when another route is not in the network. VERY HANDY! I can't wait to start using this in the real world. I always have ISPs choosing the wrong path even when the AS-path is longer. Usually because the remote host's ISP is the same as the low bandwidth link.

Wednesday, July 11, 2007

Daily Update

I'll go through my history and a little bit more on the CCIE when I have some time. For now, I really want to get started on a progress report. Today I completed the Internetwork Expert Lab 3 IGP section after work. I felt really comfortable about the time it took and the requirements. I did lose some time with reachability to the border IGP routes. It turned out that EIGRP external routes were being redistributed into OSPF external routes, which had a lower administrative distance, thereby creating a routing loop. Increasing the OSPF external AD to 171 took care of that issue.