I finally finished Narbik Kocharians’ BGP Advanced Lab Workbook last night. I have been having a tough time getting rack hours in. The BGP book was good for establishing a sound understanding of BGP fundamentals. When you’re done with that particular book in Narbik’s advanced workbook series, you’ll have covered the following topics:
- Establishing neighbor adjacency
- Route reflectors
- Conditional route advertisement & BGP backdoor
- Route dampening
- Route aggregation
- Community attribute
- The as-path attribute
- The local-pref attribute
- The weight attribute
- The MED attribute
- Filtering routes using access-lists and prefix-lists
- Regular expressions
- Advanced BGP configurations
- Administrative distance
I thought I’d seen every insane, strange thing you could do with BGP doing the NMC DOiT labs, but I still learned some things I did not know going through Narbik’s exercises. Narbik’s approach to BGP (like his approach to all the blueprint technologies) is to make as simple a lab setup as possible, and then demonstrate how to use the specific components of the technology.
For example, when doing the community attribute lab, there were 5 routers, configured in 4 different AS’s. All of the well-known communities were demonstrated in various lab tasks. Then the lab was reconfigured to be 3 routers in 3 different AS’s. Then manual manipulation of the community attribute was demonstrated. When you were done with all of the lab tasks, you know pretty much what you’d need to know for BGP communities, at least for R&S.
All of Narbik’s BGP topics above worked like that. Build as simple a lab as possible, then flog the tech to death so that you know what you can do with it.
I’ve said this before, but I’ll say it again now for people just starting out in their prep: don’t pass the written exam and then start doing full-scale labs or mock labs. Everybody’s different, but my strong opinion is that you need a workbook series like Narbik’s, or the InternetworkExpert.com IEWB volume 1. You need to spend lab time focusing on individual technologies first. THEN you can start putting it all together in full-scale lab scenarios. I can’t tell you how many hours I frittered away on NMC DOiT’s saying to myself, “They want me to do WHAT?!? You can DO THAT?!? I had no idea…” That’s a rough way to learn the tech.