ciberkot

Members
  • Content count

    36
  • Joined

  • Last visited

Community Reputation

-1 Poor

About ciberkot

  • Rank
    Newbie
  1. ok, if you redistribute the connected loopbackon R2 into the EIGRP domain, your OSPF domain will not have any route to loopback bacause you don't redistribute EIGRP into OSPF somehow (don't you?), the same is if you redistribute your Loopback into OSPF only. in this particular case you should redistribute your loopback into both routing domains.
  2. if you use dinamips try to use 12.3-Txx IOS
  3. Thanks a lot to you all!
  4. Could you explain what is the difference then between physical and multpoint subinterface in your case? I did some tests and for me it works with physical interfaces as well. the only problem can arise later on, in the routing table you have /24 netwoks and /32 host routes, and if "no peer neighbor-route" configured then hub will not know via which interface the spokes are reachable, because it has 2 routes like C 148.1.0.0/24 is directly connected, Virtual-Access1 is directly connected, Virtual-Access2 but as the spokes are on the different DLCIs hub will not be able to reach one of the spokes, or it will loos 50% packets. actually in that lab, if I remember, there is a full-mesh frame-relay, means that spokes can reach each other directly.
  5. There are many ways to get spokes to talk to each as Beavis points out. What is special about this Narbik scenario is that it states (directly or indirectly): 1. You should not run a routing protocol. 2. You should not use point-to-point sub interfaces on the spokes. 3. You should not use frame-relay map statements. 4. You should not use PBR. But the spokes still need to reach each other. The only way to do that without using any of the other options is to use PPPoFR because then you do not need to rely on any map statements. If you had bought the workbook you would have seen that in the solution. I don't think PPPoFR is possible here, the spoke will not reach each other, from the requirements there two solutions: 1. inverse-arp - there is requirements to disable it 2. ODR - from the commn sence it'S not a routing protocol, and I believe it was the correct solution in Narbik's book. Upd. hmm, PPPoFR should be the right solution, you're right, but I remember there are some issues with that setup if you enable later on a IGP protocol there - the hub use wrong virtual-access interfaces to reach the spokes.
  6. As I posted a couple days ago I successfully passed the CCIE R&S lab from the first attempt. Here I'll try to summarize my experience to CCIE (but please don't expect much and fluent story ). I'll start from the end - on 21st of January in Brussels on 7:30 we met wiht other 9 candidates (4 or 5 RS, 2 VoIP, 2 Security or SP and one Wireless (didn't know before that there is the Wireless CCIE track ) ) on the lobby. Two of the candidates were CISCO employees. at ~8 o'clock we were gathered and followed to the lab-room, where we were given short instructions. Room with some drinks, coffee, tee and snacks was opposites the lab room, one could go outside so often as needed. Some people says that in reality the lab exam much easier as expected, but I should say here, if one really spent one year to prepare for the lab he should know about 70% what to expect there, you're not going to meet there something new, if you did, it simply mean that during the preparation you missed something. For me it was exactly so difficult as I expected - I didn't meet there any new technologies or approaches, it's more about interpretation the questions and solving the issues, which are generated for you specially during the lab. and I'm pretty confidential - I'm not genius, and if I would meet the lab with difficulty like 9-10 (according IE scale) I would fail, with level 8 I think I'd be right above of the 75%. Ok now to my approach and experience. Actually it's quite difficult for me to give the exactly amount of time which I used for preparation, because during last 3-4 years I was preparing for different Cisco certifications: CCNA - CCNP, then CCDA - CCDP and CCSP. it means that for the written exam it took me about 2 months without big stress. I started my first part of lab practice (IE Advanced Technologies lab and may be 4-5 lab from the Main part) on March 2007, but during this time I was mostly busy with building my test lab. I started with real-hardware (I had a few cisco's routers on my own ) but then I realized then dimamips gives me exactly what I need + flexibility. I took Dell 2650 with Double-DualCore 2 GHz and 4 GB RAM, install there all necessary tools and it was enough to run 13 routers + FRS. On my working place we have a test lab and I used it for the practice on the switching tasks. Unfortunately I didn't have any 3550 or 3560 but enough 6500s . I played around with my test lab until I managed it to a very flexible and stable envenoming and then middle of May I started to prepare for the written part. It was for me the easiest part, after all certifications which I was preparing for, I didn't see or learn anything new, may be only QoS was for me a "dark" area. I read the official Cisco's CCIE written lab guide and one of the Cisco's QoS guides and I passed written test without big problem and stress. After written part we traveled with my family for one month to our parents, of course this month was completely out of the preparation plan. Ok, Middle of August we were back I started to plan my journey. I wanted to take the first attempt on December, but the first available date was 21.01.2009 and I said - OK, it's better then, because the X-mas and NewYear time are not going to be very productive anyway. After the first labs I realized (and I supposed to be) that my strengths are L2 switching and BGP (it's my work where I could master this topics), the weaknesses are QoS and Redistribution (sick!) The plan was following: - the first 10-12 weeks I had to finish the second part of IE completely - next 4-5 weeks part 3 of IE, - next 2-3 weeks rent some racks to play around with 3550/3560 and then to finish 4-5 Mocklabs. - If time is still there try other study materials (In my case it was I P Expert v9, I managed about 10 last labs, one lab a day). of course life and family make their own "tuning" to my plan, accidentally traveling, friends visits, X-mas and NewYear but at the end the plan was 100% complete. According to my plan I needed to spend 5-6 hours every day until the time X, and I think I was pretty much close to it. In the last 6 weeks before the lab we were surprised that our company is going to be closed in half a year but from the other side the exam fee and the travel was already paid J and thank to that fact I could spent my working hours completely for the lab preparation, I think last 4 weeks I spent around 8-10 hours per weekday for practice. The first 5-6 months (including the written part) was the months where I really leant some new technologies and approaches. As I said for my work I was pretty experienced in L2 switching and BGP by the rest should be yet mastered. I read a tons of materials from DocCD and some external sources, like IE blog (very very useful information hxxp://blog.internetworkexpert.com/category/ccie-routing-switching ) and IE forum (where they discussed the labs solution), and dozens of CCIE forums and blogs, including our favorite sadikhov The last 2 months was the time when I practiced my speed and accuracy. At the end all basic tasks (L2 switching, Frame-relay, basic IGP and BGP) where done by my fingers without the head. The IOS application/services are the topics which one should simply know that are exists somewhere and to know the way (from DocCD) how to configure them. QoS should have been practiced a lot (at least for me) in order to get really understanding Security was not so difficult for me because I have some security backgrounds, of cause the way of configuration should be practiced. IGP redistribution - sometimes I thought my head is boiling , but at the end I decided to follow the advices to not loose too many time and efforts to do the redistribution which is requitered by a task and simply do only the necessary redistribution to get the full reachablity IPv6 was pretty straightforward, without MP-BGP it's not so interesting ok, to my lab: I lost too much time reading the Task book and trying to solve the issues and the beginning, It was really enough just the look through it and notice some relationships between tasks. I found out, the L2-L3-IGP part was about 75% of all tasks regarding time and efforts. Services,IPv6, Security and BGP were pretty straightforward (but I got into pitfall there anyway ). At the lunchtime I didn't finish my IGP part completely, redistribution task with some really strange requirements was left, proctor couldn't help much, but at least he tried. He answered another couple of questions and I was pretty satisfied, there were not too much what I wanted to ask him anyway. After lunch in 2,5-3 hours I finished all tasks and for me left 1-1,5 hour to check everything again. I did a short break for coffee and some fruits. During the re-check phase I didn't find anything what I could improve , but in the last 5 minutes I realized (BGP section was the last one this time) that my neighbor AS don't see the prefixes from one of the BBs! Here I got some panic, but could very quick found out that one of iBGP neighbor missed next-hop-self option, after I put it on the placed and restarted the BGP session all prefixes were there! God.... Ok from my own calculation, if I take only really 100% tasks (where the interpretation and solution were correct from my point of view) I reached 85%, plus 8-10 points on the questions where my interpretation could be correct, and a couple of question where I didn't see the way how to meet requirements (one strange question about STP timers and shortest ACL). After the lab three of us went to the same and had a chat. For the one of them it was the third attempt for the second it was the first as well, we discussed a little bit the exam and then I went to my room to get a little bit rest. The results were available at 2:00AM but I was at this time completely sick I think I had 39C and couldn't sleep at all, I checked the status, called my wife and she was very proud of me. I was really satisfied and fill better
  7. I'll write the story later on, right after my exam I got terribly sick and I can't now really concentrate. from the material it was IE and I P Ex (just to get another view) At the end the lab was exactly so difficult as during the preparation, but it's more stressfull because of new envinroment and the time pressure. I think I spent every other day at least 4-5 hours during last 6-7 months.
  8. Hi, First attempt and it was successful, thanks to half a year and about 1000 hours of preparation. Only a couple of tips - master your configure skills for the basic task - Switching, Trunk, Channel, VLAN, Frame-relay, all possible OSPF variants, basic BGP configuration (RR, confideration) Your finger should be able to do it alone without loosing of time and without using of your head. Your head comes in play when you will need to configure tricky tasks and troubleshoot the issues. many suggest to read the whole task book before, it's OK, but just don't spent too much time, you shouldn't try to solve all issues during this first read, just look through it, notice approximately the difficulty level (just for your scale), but not more then 10-15 mins. Then diagrams - actually they quite good and one can use them perfectly, but in order to get more feeling it'S good to draw your own diagram. it takes some time but for me it was esier to use meins. thanks to this forum, it gave me many good tips and helps.
  9. Hi, what I'm curios about it's the "tools" which are available during the lab. From the Cisco's description it's clear that CRT (single or multi - session ) and Calc are there. What else? Is there some kind of text editor available? if yes is it possible to use Search/Replace feature there? How the "Task book" look like? is it a hard copy or electronic format? How the diagrams look like - are there separate Physical and Logical digrams? Is it possible to draw something directly on the diagrams or the "Task book"? If we compare I.E. and IP.Expert. prepare materials which one of them is closer to the real lab? Or is it completly different? How the IP addressing looks like? Are there mostly /24 subnets or different length? Thank you for the answers!
  10. policy-routing, really? Allright then.
  11. Thank you all for the replay! Isn't ODR a dynamic routing protocol? Hmm then it's an option. the policy-routing is normally not an option on CCIE lab, but surely possible solution, but I'd use then route-map PEER-DEFAULT per 10 set ip default next-hop 1.1.1.1 in order to send only packets which are not matched routing table. Anybody has the Narbiks answer book? What Narbik says about it? Upd. Hmm, the very next topic in Narbik's book after this FR lab is ODR, I don't see any sence then to use ODR for Frame-relay lab. Actually I don't see any sence to use policy routing either, as I understand, in this Frame-relay topic, all labs should be solved with only FRame-relay features, on L2, without any L3 functionality.
  12. Hi, may be some of you tried to configure such a nework (Narbik Advanced-technology, Frame-relay Lab7): R2 | | R1-----R3 | | R4 R1 - Hub R2,R3,R4 spokes R1,'s loopback0 1.1.1.1/8 R1's frame-relay interface: Address DLCI unnumbered loo0 102 R2 unnumbered loo0 103 R3 unnumbered loo0 104 R4 R2,'s loopback0 2.2.2.2/8 R2's frame-relay interface: unnumbered loo0 201 R1 R3,'s loopback0 3.3.3.3/8 R3's frame-relay interface: unnumbered loo0 301 R1 R4,'s loopback0 4.4.4.4/8 R4's frame-relay interface: unnumbered loo0 401 R1 configuration should not use "frame-relay map" command, any static or dynamic routing, no subinterfaces. Routers should have full reachability to all loopbacks (!) How to get the connectivity between spokes?
  13. it's not always the the case, in some case you configure the parameters in bits , but the show commands show them in bytes, but anyway, may be in question: they simly ask to use Bc=16000 bytes and Be=16000 bytes? What does ask the question exactly? The Question : "On R8 limit the amount of inboud web traffic from R7 to 64k with a normal burst size of 16k and excess burst size of 16K". See the problem is that the bc and be show in bytes while configuring rate limit under serial interface. it is "k" as per the question the bc and be in 16k so if u convert in bytes then u need to do 16000/8. Question mention bc and be in "k" but how do you read or understand these "16k" - as 16000 Bytes or 16000 bits or 16000 bits/s? usually if one talk about Bc and Be or burst zize it means bytes. Then question should be undestood as "use burst size 16000Bytes and excess burst size 16000Bytes" you don't nee to convert them in Bytes again.
  14. it's not always the the case, in some case you configure the parameters in bits , but the show commands show them in bytes, but anyway, may be in question: they simly ask to use Bc=16000 bytes and Be=16000 bytes? What does ask the question exactly?
  15. *I think* : if it's not the Frame-relay policing, then the you should use the number bits/s in the command. For me it's the most unclear part of QoS - when you should use Bytes and when you should use bits and when bits/s. but you can always check yourself e.g. with command "sh int <> rate-limit" or "sh traffic-shap" for Frame-relay.