OSPF Webinar Link Plus Questions Answered
Pearson Education, publishers of the Cisco Press products, hosted me for a CCNA-level OSPF web cast back in February 2012. (You can go here to watch the video of the webcast.) Anyway, during the webcast, anyone watching could send in a question, and there was a set time for me to answer. I answered some of those in the webcast, but not all. The folks at Pearson sent me the list of questions, so I figured it was potentially helpful to list the answers in a blog post. So… here it is!
Some of these may not make sense without watching the video of the webcast, but some will. Also, here’s a link to a PDF that shows the network topology and area designs used in the webcast. (Also, i’ll wrap up the OSPF lab answers here in the next week or so.)
These are just listed in the order I got them from Pearson, which pretty much means no order at all…
1) “What proportion of the current CCNA exam questions are OSPF based?” “Are the topic proportions consistent over multiple exams?”
a) It’s a good question, but only in that I have to say that it’s squarely against Cisco’s Non-Disclosure Agreement rules for all of us to not discuss the particulars of the exams. The answer would be very useful, but totally off limits due to the NDA, unfortunately. Sorry about that…
2) “what is the significance of process ID in OSPF??” “but i guess DR-BDR thing does not apply here right?..”
a) That’s a 2-parter. Part A: the process ID is local on a router. It’s possible to run multiple OSPF processes on one router; that’s just not a CCNA level topic. It’s somewhat common when you get into MPLS designs, for instance. But the process ID doesn’t flow in messages, so it doesn’t have to match amongst the routers. Part B, if you care to remember back to the webinar, yep, the DR/BDR thing does not matter in that design. You need at least two routers on the same multi-access LAN or WAN for that to be an issue.
3) “Why use area’s?” “Why logical reasons are the area’s used for?””got it, thanks…”
a) I think I got this one on the video, so I’ll leave it at that. 😉
4) “how can ospf deal with loops, does it have any rules like EIGRP”
a) Likewise, I think we got it on the videos as well. Short version, it builds detailed topology information, so any router can do the equivalent of us humans looking at a map and choosing the best current route.
5) Is this CCNA level OSPF? Will this be available for later viewing too?
a) Well, I tried to make it CCNA level OSPF. And yes, it’s already available for viewing as of the time I got around to answering the rest of these questions; see link with this blog post.
6) “What’s the difference between using “”summary-address”” and “”area range”” on an ABR? (CCNP, I know…)”
a) No problem… summary-address summarizes routes that a router grabs from another source of routing information during redistribution. EG, when redistributing routes from EIGRP, the summary-address command’s logic would be applied to the routes being pulled into OSPF from EIGRP. The area range command applies to routes in one OSPF area, as they are injected into another OSPF area. So, one’s for external routes, and the other is for inter-area routers. (OSPF does not allow route summarization just at any point in a network; it only allows summarization at the ASBR, where redistribution happens, and at ABRs, at the edge between multiple areas.
7) “Question for Wendell: What is the terminal software used in this presentation?”
a) It’s called “terminal”, and it’s built in to Mac OS X. It works pretty well. (Edit – well, I didn’t. I instead used iTerm for the preso, not the built-in terminal. Whoops, sorry about that. iTerm is free, at http://iterm.sourceforge.net/index.shtml.)
8) “Why did you use proce 4 in the command router ospf rather then proces 1 on R4”
a) You did ask “why”… I did it just to show that neighboring OSPF routers did not have to use the same process ID, so anyone who didn’t know that would notice/learn/ask. No other reason. Looks like it worked! 😉
9) “what about the DR BDR ??” “i mean behaviour of OSPF on serial link” “is that compulsory to connect to area 0”
a) I think we covered all these either already here, or as part of the event…
10) “? from Richard: What is the best way to join multiple networks that might have duplicated OSPF Router ID’s (conf’d with Lo interfaces)? Convergence of Organisational Structures…”
a) I probably answered this one a little too quickly, and maybe incorrectly, live in the video. First and foremost, don’t do that (duplicate OSPF RIDs.) Of course, merging two OSPF networks may cause it to happen. The best option is to change the OSPF RIDs so there are no duplicates. What I left out in the video was to not only prevent the routers with duplicate RIDs from being neighbors, but also keep them from being in the same area, which also means to avoid having those routers as Area Border Routers (ABRs).
11) “? from Gufran: what about the DR BDR ?? >>KS>>sorry wendell I don’t know what this may mean .. skip the ? if you don’t know it’s association?”
a) From Wendell – just to make sure it’s clear, in the topology I used for the session, there are no DR/BDRs. In fact, there’s no DR election until a router recognizes a 2nd router in the multi-access LAN or WAN, and there are none. So, a non-issue here.
12) “? from Jeff: Wendell, thank you for doing this webcast. I noticed a lot of updated versions of the CCENT and CCNA study material. Have there been changes to the CCNA that we will have to consider before taking the test anytime soon?”
a) Cisco did not announce changes to the exam that I’ve heard of. For the first time since Cisco announced CCNA in 1998, we revved the books without a concurrent rev of the exams by Cisco. I wrote up a series of blog posts about it. I’ve listed the links here. But the short answer is that there’s no announced content change. I did not add new topics to the new editions. I did re-write some sections, plus the publisher had other motivations for revving the books as well. Check out the “News” tab at the CCENT Skills blog, or just click here to do the same, to see those posts.
13) “? from Vakhtang: i was thinking that R1 must redistibute OSPF areas or somesing like this, Now network is fully converged yes ?”
a) We got this one on the video, but yes, there is no redistribution required in what we showed in the video.
14) “? from Majid: What are the common causes of OSPF routers not exchanging routes properly or at all?.”
a) I’ve got a list in the books, by the way, for easy reference/study. But the more obvious: they must be in the same subnet, must be in the same area, and they must pass the authentication checks.
15) “Why exactly du you use zero subnets in ospf configuration for R2 and R3? sry if i dont get it earlier…”
a) An interesting question… the R2 and R3 configs don’t use zero subnets. There are no zero subnets in the entire example/preso. I’m guessing we have a terminology issue? Or maybe you’re thinking that “network 0.0.0.0 255.255.255.255 area 0” implies a zero subnet? (It doesn’t.) The reason I used those network commands was just to match all interfaces, and put them in area 0, and those work. That’s all.
16) if you have a single OSPF area design, do you still need area 0?in the rtg table for ospf routes, the 1st value [say 110] is for adm cost of OSPF. What’s the 2nd value for? it’s diff for diff ospf routes. e.g., [110/65] or [110/74] on R1
a) Two parts here. The first part we covered on the video, I believe. On part B, the 2nd number in brackets is the OSPF metric for the route.
17) Why is the advantage to have multiple areas?
a) I think we got this one on the video as well…
18) Are you likely to host any CCNP level routing webinars?
a) I don’t know if this one was for the Pearson folks or not. I’ll probably do more of these webinars in the long term, in between book schedules. Suggestions?