Dear All Expert,
I would like to seek for your expertise to answer my few questions for CCIE R&S LAB Exam.
1. Regarding the Frame-Relay part, is it all "Frame-relay Map IP X.X.X.X " should put the broadcast in the end. No need to care about redudancy broadcast traffic ??
2. About the Qos Question, if i configured MQC like below :
match protocol http url *(.gif|.jpeg)
service-policy input abc
My question is --- should we configure ip nbar protocol-discovery under the fa0/0 ??
3. OSPF part : If the question ask me to configure all interface loopback as prefix /24. But in the OSPF part there, it didn't require us to make the loopback which display /24 in the routing protocol, should we configure it -- ip ospf network point-to-point ??
4. During the LAB exam, we just need to complete the asking question ? if not specific then not do it ?? Or just over configure it due to not penatly for it.
5. Any other advice for me or other CCIE R&S candidate, your useful message will save us alot of power :)
Really appreciate all expert or even proctor can share share its experience and how its mark the question :)
Thanks alot @@
1. My policy is to put the broadcast keyword on only one map statement per DLCI. That is, suppose you have three map statements, two of which reference DLCI 201, and one that references DLCI 203. Put the broadcast keyword on the 203 statement, but only one of the 201 statements.
In the case of a spoke, I put the broadcast keyword on the spoke-to-hub statement, but not the spoke-to-spoke. If you put the keyword on multiple maps to a single DLCI, then any broadcasts will be replicated that number of times on the DLCI. This can play havoc with your multicast scenario.
2. I don't think so. You only need ip nbar protocol-discovery if you want it to collect statistics about NBAR-recognised protocols. You might do this, for example, if you want to design a service policy but you are not yet sure what traffic is going through the interface.
3. If in doubt, ask the proctor.
4. Again, if in doubt, ask the proctor, but phrase the question in such a way that he/she understands that you know what you are doing, that you are not fishing for "the correct answer", but you want clarification on the requirements.
5. Be prepared! I had to fail twice before I understood what level of preparation this exam needed.
If you feel able to, try blogging your studies (always keeping within any NDA of course). Recent experience has shown that sharing your technical knowledge is one of the most effective ways of enhancing your learning.
1. Whether or not you configure a frame-relay map with broadcast depends on the configuration, and whether or not the interface you're configuring is a hub or a spoke. Yes, if you just apply broadcast everywhere, you can end up with needless broadcast traffic in the frame cloud. The appropriate setting (broadcast or not) depends on the topology and required configuration.
2. You do not need "ip nbar protocol-discovery" in order for NBAR to function. "match protocol" starts NBAR all by itself, assuming ip cef is enabled.
3. "ip ospf network point-to-point" is not automatically required just because the interface is a loopback. Whether or not the statement is required depends on the requirements of the task. If the loopback is actually a /24, but shows up in the routing tables as a /32, that will be fine as long as all other lab requirements are met.
4. In my lab experience, I only did what was asked. I did not do any "extra" tasks. (Yes, I passed.) However, there are often *implied* tasks that you will have to do to successfully complete the task. Remember that when the lab is complete, you have to have a functioning network. You will not be penalized for configuring extra stuff AFAIK, but that can be time-consuming and might break something else later. I do not recommend extra configuration.
5. The best thing you can do to prepare for the lab is practice, practice, and practice. For example, if you're wondering about the implications of "broadcast" on a frame-relay map statement, spend enough time in a lab with it to understand what happens when you use it versus when you omit it. How does multicast function? What about OSPF if you're required to change the network type to broadcast? Etc. There is no substitute for practice time on the rack. You should walk into the lab reasonably confident. There shouldn't be any blueprint technologies you're worried about. If you're saying to yourself "I hope they don't give me protocol XYZ..." then you're probably not ready. In the context of the lab blueprint, most people I know who've taken the test recently have said that the lab exam is difficult, but fair.