1st CCIE lab attempt blog and help for candidates.

Turgon

Ok I finally got MP-BGP VPNv4 working It has a dependency on VRF Lite. I configured that and everything started working properly. Now on to OSPF..

Turgon

CE-PE routing using OSPF worked, mostly. The OSPF Sham link exercise was partially successful. The sham link came up but the OSPF routes are not appearing on SW2. A break now as my wife needs the PC to do some household things so I shall be spending time with the youngster. Im halfway through the Vol 1 MPLS exercises which is pleasing even though there is still much I don't understand properly at this point.

I will need to spend some hours researching PIX/ASA and IPT very soon. The new job requires it. I haven't logged on to a PIX for over six years and I have never worked with those ugly things Cisco tries to pass off as a telephone. Oops

For the moment though it's toys with the boy now.

Turgon

Ok I had a brief read through the remaining exercises for MPLS in Vol 1. We have a busy day tomorrow and have things to do in the city but I will try to get the rest of the exercises configured during the course of the afternoon if I can. MPLS is a very interesting subject. Of particular note is the way metrics from IGPs are transported in the MP-BGP extensions. I have PE-CE routing using EIGRP next up. Once I have completed the first pass of the Vol 1 MPLS exercises I will need to look at another topic. While there are clearly new things sprinkled around the old blueprint topics that need to be looked at closely I don't think there is a new major subject like MPLS that needs such a focused review. So I think I can just pick any topic from Vol 1 and just look out for what features have been added and which ones have been dropped. On that basis it seems to make sense to tackle the subjects I suck on first.

These will be Security, QoS, Multicast, Optimise Network, Network Services, IPv6, IGP/BGP, Layer 2. Plenty to be going on with there! Then it's the Vol 2 workbook again taking on some multiprotocol labs. Throw in Vol 4 troubleshooting practice, OEQ practice and some mock exams I imagine the struggle will continue far into the summer. Many long evenings of studies lie ahead. For the moment though, patience, faith and constancy of purpose are critical.

One could easily feel overwhelmed. Might be time to blow that horn

http://www.youtube.com/watch?v=ZpwsKRpKS_M&feature=related

Turgon

Off to the shops then back on with Vol 1 MPLS exercises. Will grab a rack slot now.

Turgon

More wasted racktime. By the time we got back home and I got logged on to the end of one slot and set up a single router it ended. Par for the course. Still, I have the next one starting in a few minutes so on to PE-CE using EIGRP and as much MPLS config as I can stomach this evening

Turgon

INE Vol 1 CE-PE using EIGRP - DONE

With a real struggle and admittedly some of the routes in the given show output did not show up. Dependencies on previous labs again. This time it was VRF-Lite, perhaps LDP (which I added), MP-BGP and finally the loopbacks and network injection into BGP in the *previous* lab CE-PE OSPF that really got things working better. Have swapped emails with Anthony Sequeira about these dependencies and Petr Laprokov is on the case. I think this is all by design but perhaps needs to be better spelled out in the workbook. At the same time the MPLS-VPN config in graded labs is a barebones so you will need to save your configs as you go and load them up before starting your next exercise. Some 31.x.x.x routes in the CE-PE RIP lab dont show up either so I think there may be topology issue with backbone routers in the workbooks verses the rack hire physical topology. Anyway, progress. Message to all CCIE candidates, *really* get stuck into MPLS or you are toast come lab day. OEQ, TS, Config await!

Turgon

I will be purchasing the MPLS fundamentals Cisco Press book. Meanwhile I will skiddoo through exam D in the Boson test for MPLS and see if I can steal some more time away from the family today to plod through some more exercises in Vol 1.

Turgon

Study plans abandoned for the day! An afternoon out in the City with the family awaits. Should be fun

yuriz43

Turgon, what do you use for your labs? Do you use dynamips? Rent racks? Both?

Turgon

yuriz43 wrote: »

Turgon, what do you use for your labs? Do you use dynamips? Rent racks? Both?

Remote racks these days. I have a stalwart rack of 2500's named after female Tolkien characters that served me very well the last couple of years for basic technologies but I needed the remote racks for the features. Not a dynamips man, Im too old school for all that messing about

Turgon

INE Vol 1 MPLS

EIGRP Site-of-Origin - DONE
PE-CE Routing with BGP - DONE
BGP SoO Attribute - DONE
Internet Access - DONE
AToM - DONE
L2TPV3 - DONE
MPLS VPN Performance Tuning - DONE

All exercises completed today with mixed results. Far too much to learn on MPLS for Vol 1 to be anything other than a tutorial so I just put in the solutions and observed. Not everything worked out. A long way to go on MPLS then and I will be reading the whole section over in detail again in the New Year, but at least it is the end of the beginning so far as making a start on configuring CCIE R&S MPLS blueprint topics is concerned. Better understanding will come in time. I wanted to get to the new blueprint before the year ended and I have managed to accomplish that. There will be no more configuration practice in 2009 as we are now getting ready for the New Year celebrations with family, so that's my last rack session for this year. Some impressions on plans for 2010 in my next post, but for the moment 2009 CCIE prep is finally over.

Certainly the toughest but most productive year of all for me, involving study however little almost every single day of the year. Hundreds of hours have gone in, even during my family holiday in Russia back in the summer and I haven't stopped at all since we returned to the UK in September. Many workbook labs completed in Ipexpert and INE Vol II and Vol III. Many technologies worked hard in isolation. Also a written pass attained with all the accompanying reading. This has given me a shot at a lab pass next year but I had to put all this work in to *earn* a chance of being able to prepare properly for v4. All that work lies ahead of me in 2010 and it will be fierce.

christino

Congrats Turgon, you done mentored peeps you dont even know already. Wish you all the best, following in your footsteps soon. Regards to the family. Happy New Year in advance.

Turgon

christino wrote: »

Congrats Turgon, you done mentored peeps you dont even know already. Wish you all the best, following in your footsteps soon. Regards to the family. Happy New Year in advance.

Same to you my friend and all at TE. Happy New Year when it comes around folks! Hotel and Flight booked, off to Barcelona on the 3rd!

Turgon

Flew into Barcelona this evening. Checked into hotel. First day tomorrow. Priority is a good meet and greet first week and to execute any work given without problems. Business dynamics may prevent a renewal of contract regardless of expectations or performance OTJ. At the least non renewal will not be down to effort on my part. Let's see how it goes here.

On the study front. Evenings and weekends free to make progress although I miss my family. Will try the spiral approach to CCIE topic revision, returning to any subject covered in 7 days.

GiddyG

All the very best with this. I can empathise on the family front, having had to spend months away from mine a few years back.

I look forward to reading about your Spanish exploits.

laidbackfreak

Barcelona is a fantastic city, I keep half an eye on the job market there too, so if anything else comes up give me a shout lol

Good luck with that gig and hope it rolls on for you

Turgon

Thanks guys. First day over. Intense. Happy to report nothing there technically to phase me. Will now concentrate in settling in and provisioning what our sites need worldwide.

A busy but interesting city. I find the weather rather mild

veritas_libertas

The big questions is, do you speak Spanish?

Turgon

veritas_libertas wrote: »

The big questions is, do you speak Spanish?

You must be joking. Will try and learn it

veritas_libertas

Turgon wrote: »

You must be joking. Will try and learn it

I would really like to learn it myself. I am glad the job is working out for you, hopefully it will be a long term contract.

Turgon

veritas_libertas wrote: »

I would really like to learn it myself. I am glad the job is working out for you, hopefully it will be a long term contract.

Certainly an experience. But before the whole family can commit to the move out here and the expense some longivity in the contract is essential. Im covering for someone going on leave in 3 days time so not much time to learn everything to fill the hole there. Opinions will be forged at the end of my first week when I have been covering for his absence. So no pressure there then

Turgon

Ok so we turned a page and we are now studying for the CCIE evenings and weekends in Spain.

I have had some time to work out a plan of action which I will now execute. Essentially prep falls under one of two umbrellas..

1. Basic config

The least demanding of the two. Know enough to light up the necessary configuration for a given protocol or feature. This can be learned and gets better over time with repetition.

2. What I very loosely refer to as 'Traffic Engineering'. This is a far from adequate description as I am using it to carpet bag so many strictly non TE areas, but essentially Im talking about the fine tuning of things to meet a *specific* requirement.

This really requires knowledge of how things actually work or behave and separates the men from the boys. It could be timers, it could be constraints, it could be prefering one path over another. But every subject area on the blueprint has things like this and mastery of them is essential. No lab pass without it.

Turgon

Time for a skype call with the folks back home. Then I will skiddoo on the command memoriser for an hour. Fancy looking at QoS today.

Turgon

That was a good session on command memoriser. I just went over the first of two batches of exercises for QoS. I held together quite well just needing to reorientate my selection of various commands and recall a few I had forgotten.

For class maps, one can match on many things..

match on a URL with fast in it..
match protocol http url *fast*

match on IP precidence 3 or 4
match ip precidence 3 4

match on an input interface
match input-interface ethernet 0

match on an ACL
match ip address 1

match on mac address
match source-address mac 0000.1111.2222

match on COS
match cos 5



For policy maps, be careful to differentiate between marking, shaping, policing, WRED and giving a maximum bandwidth

gets 30K...
bandwidth 30

enable WRED on all remaining traffic
class class-default
fair-queue
random-detect

give 10% of bandwidth
bandwidth percent 10

police to a maximum of 64K
police 64000 conform-action transmit exceed-action drop

get average of 16000
shape average 16000

get peak of 16000
shape peak 16000

change to dscp EF
set ip dcsp ef

change ip precidence to 5
set ip precedence 5

enable WRED
random-detect

many things..

Turgon

Ok going quite well today. I think I will potter along and look at FRTS now. Is it QoS? I suppose that's a grey area, but not as grey as my definitive understanding of those nasty CIR/Bc/Be values which I need to brush up on.

frame-relay traffic-shaping

interface dlci_302
class FRTS

map-class frame-relay FRTS
frame-relay cir xxxxxx
frame-relay bc xxxx
frame-relay be xxxx

Oh yeah and there is the fragment thing too..

frame-relay fragments 960

Turgon

Ok command memoriser has some FRTS stuff in section two. It's ok but a bit lightweight with no exposition on the reasons why the bc and be values are so. DE eligibility was also covered.

I abandoned the section once FRF compression exercises came up as thats not the object of todays exercise. Im going to have to read up on the CIR/Bc/Be thing again so I have it down when a FRTS requirement comes up on lab day. I can configure the stuff well enough but those values may be implied in requirements and I may need to be able to figure them out or lose the marks.

Isolating all these horrible things is very necessary. There are many of them across protocols that I need to pin down, read up on and sort out so Im ready for them on lab day. On Tuesday I looked into using Variance to set up unequal path load balancing on EIGRP. We know that EIGRP will naturally load balance but only if the metrics for the alternate paths are equal. Given that the EIGRP metric is so complex it's unlikely that you will have equal metrics there for alternate paths. So variance gets things working by factoring paths that have a metric value within a multiplier of the best metric. Ok so far. Find your best metric, look at the others and stick in a value that allows the others to be used. But then EIGRP will use them in a way it sees best. If you want granularity of the ratio of traffic using different links then you need to get into metric calculation in detail and work out the delay parameter you need to add to various interfaces. Horrid. I can follow the maths, but its going to take lots of practice until 6 months down the line I can spit that sort of stuff out if it comes up.

That's just a landmine in EIGRP. There are many others like that. Reading is important but Im going to have to spend lots of time isolating these sorts of things and working them out.

If you blow these sorts of questions it shouldn't break the shop, but with a liberal sprinkling of them across your lab you will drop enough 3 point questions to fail. I need to get better at them across the piece. We all do.

laidbackfreak

Turgon wrote: »

If you blow these sorts of questions it shouldn't break the shop, but with a liberal sprinkling of them across your lab you will drop enough 3 point questions to fail. I need to get better at them across the piece. We all do.

Aint that the truth. I'm heading down VP path at the moment and know I'm skimming some of the stuff that irl I would grab the books to refer to BUT I know I need do some serious lab time to get it down solid

Turgon

laidbackfreak wrote: »

Aint that the truth. I'm heading down VP path at the moment and know I'm skimming some of the stuff that irl I would grab the books to refer to BUT I know I need do some serious lab time to get it down solid

Yeah theres a real tendency to clatter through a lot of material to pass tests. The approach has uses but lately Im finding a slow down to look at just something small in depth and reflect on it is reaping dividends. Read up on FRTS basics today. Makes more sense now. Had a look at the INE Vol 1 section on QoS. 400 pages of hell in the solutions guide. Truly awful, the prospect of studying that madness properly. It will require a great deal of time. Will concentrate on the FRTS bits first I think.

Turgon

A game of inches. The evening consumed sorting things out for my next hotel move this weekend and talking things over at home. So the rack session I scheduled tonight went nowhere. Still at least I can try logging in before I turn in. A busy day at work, hanging back later to carry out my first scheduled change. Still, at least I looked at FRTS theory today..

Turgon

The end of the first week in the contract. The guy is on leave now and Im holding the fort in his absence. I think it went well and I made decent impressions. Been informed that a Network Architect is leaving so I will be inheriting his dual DC project with failover and MPLS thrown in. Voice transport work. Another 7206VXR arrived today. Cool. Need to get it finished before handing over to the support team. Working with people all over the world. Lots going on, lots to do. Busy, busy.