Random (and not so deep) Thoughts by Some Clown
I haven’t been writing a lot lately, mostly due to a combination of my work and study schedule. I thought, however, that it would be useful to just toss down a few random thoughts on the proverbial paper to wrap up 2010. I’ll try to keep it somewhat cohesive, but I can’t really guarantee anything.
Having made the decision last year at Cisco Live to finally buckle down and pursue the CCIE Routing and Switching certification, I have been as busy as you might imagine with studying. As I’ve gone down this road I’ve noticed a couple of things:
(1) In the office I’m used to studying large white papers, documents, manuals, command references, etc., quickly to get to the answers I need for either deployment or break-fix. This is not the best way to study for the CCIE qualification exam, however, as I tend to just as quickly forget that information past the point of it being immediately useful. I’ve had to change my habits now to include taking notes, reviewing portions over and over, and cross-referencing with multiple sources. Nothing earth shattering to be sure, but a change for me.
(2) As alluded to above, I do a lot of cross-referencing on my study material. I have material from CCBootCamp that I consider to be my primary source (by virtue of being enrolled in the Cisco 360 program through them). I have also been reading the CCIE Routing and Switching Certification Guide, 4th Edition, as well as the CCIE Routing and Switching Exam Quick Reference Sheets–both by Cisco Press. I think it helps me quite a bit to read different perspectives on the same material; to see it put a different way on the page. I have a Cisco Live Virtual account as well, and so have been pulling some presentations–notably on QoS–from that site.
(3) I have over 16 years of professional experience in this industry, and while I am by no means an expert, I am confident in things that I know. To that end I would say that at some point in your studies you will be almost guaranteed to come across information, answers to practice questions, etc., that you just know are wrong. I’ve had to learn not to be afraid to challenge my study material. I don’t do it blindly, but I do go out and research in other sources to verify what I think I know. I have found many instances of incorrect information in several sources–more often than not in the Cisco IOS example configurations. Sometimes using commands that won’t work on that platform, other times referencing non-existent class-maps or access-control-lists. Less often have I found blatantly incorrect explanations of how a thing works, but even there I have found a couple of examples. I take this as a good sign, actually; it’s a sign that I am becoming more aware of the details of what I am studying.
Interesting Design Decisions
It always fascinates and bewilders me to see some of the design decisions that other engineers make when putting together a network. Much of what we do is subjective, and even the most experienced experts disagree on a good many things. With that said, certain things just don’t strike me as particularly useful and it’s my prerogative to complain about them. My top complaints from recent experience, in no particular order are:
(1) My predecessor who built our main datacenter using 4503 switches exclusively: access, distribution and core (mostly, but we do use a collapsed core model). The 4500 series is great but my general argument is that they’re under-powered, or at least under-featured for the core (Sup II-plus) and just a bit overpowered for the access layer. We use PoE 1‑Gig to every port in the building, but the access layer is still barely running (less than 1 percent utilization ever, on any metric). I think someone got a deal or something. We’re now replacing the core with a pair of 6506, 720 supervisor, 10-gig, etc.
(2) A main distribution point had a single 3845 with a 100-meg Internet connection, and two full DS3 links. Considering the 3845 maxes out at 45 Meg of throughput, this seems a particularly egregious violation in my mind. We’ve now moved that to a 3945, which if under full load is probably still a tad over-subscribed, but much better and the price was right.
(3) Who was it at Cisco that decided that the ASA-5510 would only have two Gig links available, and only with the right license? Why only two? Why not three or all five? This might be a backplane issue, I don’t know, but it just bothers me.
(4) My own stupidity in setting up the aforementioned ASA-5510 pair (failover) with the inside and outside interfaces on the gig links, when I should have had the two trunk links that handle much more traffic on those interfaces. This will be changed soon, but I should have done it right the first time.
2010 has been a good year overall, with a lot of interesting projects, experiences, and solid learning had by all–or at least me. I’m looking forward to 2011 and all of the continued successes and experiences to come. I’d also like to give a special shout-out to all of my Twitter colleagues, friends, followers, and various clingers-on and lurkers. I have found the Twitter community to be an invaluable source of support, wisdom, and occasionally respite from the rigors of the daily grind. If you’re not on Twitter, I’d highly encourage you to give it a look.
Happy New Year everyone!