It begins.

testh

IT certifications in modern society are critical to many people, especially those IT workers. But the fact is even if most people want to pass Cisco, CompTIA or HP certification, lacking of time to prepare the IT exam will always make the exams more difficult to pass. So that is why ********.com exists which aims to be your loyal support and help you pass the IT exams with ease.

rakem

jovan88 wrote: »

Good to see a CCIE candidate representing Sydney! Good luck

Thanks mate! Not to many aussies on this forum.

rakem

3 hours tonight, mostly on EIGRP.

When through the Boson exam questions on the CD which came with the Cert guide and got 100% correct on the EIGRP questions so i'm happy with that.

Decided to do the QoS based questions on the CD and got hammered. Some tricky stuff in those questions so a big review on QoS is necessary. Bit annoyed at that, i thought i had QoS down. So many little things to remember.

Turgon

QoS has a tendency to hammer people. Study hard!

rakem

Turgon wrote: »

QoS has a tendency to hammer people. Study hard!

Yep, i'm happy with EIGRP so going back to QoS before moving onto multicast. Will spend this comming weekend immersed in 'End-to-End QoS Network Design' (Cisco press - Tim Szigeti, Christina Hattingh)

rakem

A good two hours of classification and marking review this morning.
Moving onto congestion management now

rakem

Here's something that i miss-understood.

LLQ bandwidth is taken from the configured interface bandwidth NOT the max-reserved-bw.
bandwidth remaining percent is taken from the max-reserved-bw minus any LLQ value.
bandwidth percent is taken from the configured interface bandwidth only, even if LLQ is configured.

however all configurations can't use more than the max reserved bandwidth which is 75% be default.
and you can't mix and match bandwidth commands with bandwidth remaining commands in the same policy

rakem

An hour or so of LAN switching congesting management and avoidance.

rakem

Tried to study tonight but couldn't get into it. Only about 30 mins of reading.

reaper81

rakem wrote: »

Tried to study tonight but couldn't get into it. Only about 30 mins of reading.

These days do happen. Don't get beaten up. If you need to take a day off and come back with recharged batteries, most people burn themself out when studying for CCIE.

Turgon

rakem wrote: »

Tried to study tonight but couldn't get into it. Only about 30 mins of reading.

Keep pumping. Once you cross 100/50 in terms of Reading and Labbing you will be able to handle the 'wall' better.

rakem

So i found this interesting:

On page 581 of the CCIE cert guide it is talking about reducing the Bc to make Tc lower to cater for VoIP and it uses the example of a shaping rate of 96Kb.

So we want to get the Tc down to 10ms. So to do that we need to change the Bc as this directly influences the Tc. So the forumla is:

Bc = CIR/100
So if our CIR is 96000 bits (96Kb) that means
Bc = 96000/100
Bc = 960

So i tried to configure that on a test 1841 i have at home.

Router(config)#policy-map shape-96
Router(config-pmap)#class class-default
Router(config-pmap-c)#shape average 96000 960 ?
less than 1000 bits in an interval doesn't make sense

Gah? That is exactly what is on page 581 of the cert guide.

So i put Bc up to 1000

Router(config-pmap-c)#shape average 96000 1000

and it works as expected, the Tc is set at 10ms, even though the numers don't add up exaclty using the forumlas.

Router#show policy-map interface fa0/1
 FastEthernet0/1

  Service-policy output: shape-128

    Class-map: class-default (match-any)
      22 packets, 1337 bytes
      5 minute offered rate 0 bps, drop rate 0 bps
      Match: any
      Traffic Shaping
           Target/Average   Byte   Sustain   Excess    Interval  Increment
             Rate           Limit  bits/int  bits/int  (ms)      (bytes)
            96000/96000     250    1000      1000      [B]10[/B]        125

Perhaps different IOS versions manage that differently?

Also - on page 580 it mentions that if the CIR (or shaping rate) is less that 320Kb, the Bc is set to 8000 bits. Well on this 1841 again its not quite like that. For example i set the shaping top 128000. Notice the Bc in the show policy-map command is 7936 - not quite 8000.

Router(config-pmap-c)#shape av
Router(config-pmap-c)#shape average 128000
Router(config-pmap-c)#do show policy-map int fa0/1
 FastEthernet0/1

  Service-policy output: shape-128

    Class-map: class-default (match-any)
      111 packets, 6694 bytes
      5 minute offered rate 0 bps, drop rate 0 bps
      Match: any
      Traffic Shaping
           Target/Average   Byte   Sustain   Excess    Interval  Increment
             Rate           Limit  bits/int  bits/int  (ms)      (bytes)
           128000/128000    1984   7936      7936      62        992

rakem

Been a bit distracted this week. Been looking around a new work and have landed a new job!
Will be working as a senior network engineer for a Cloud Services provider here in Australia.
They run all Juniper - Junos based gear (which I'm very familiar with) so it's going to be a bit different.

Still looking to get CCIE as its been a goal of mine since i started stuyding for the CCNA all those years ago.
Starting on some Multicast study this weekend.

rakem

two hours of Muliticase this morning

Turgon

rakem wrote: »

two hours of Muliticase this morning

Keep reading. Start doing the Odom test engine, the end of chapter questions and final exam. Then the Boson. It will supplement your reading.

rakem

Neck deep in a pile of PIM
(my attempt at a multicast related joke)

On a serious note - anyone know of a way to generate some Multicast traffic so i can lab up this stuff?

Turgon

You can ping a multicast address or set up a streaming application.

jamesp1983

rakem wrote: »

Neck deep in a pile of PIM
(my attempt at a multicast related joke)

On a serious note - anyone know of a way to generate some Multicast traffic so i can lab up this stuff?

Multicast | Table

Simple, but good multicast labs...like Turgon said, ping a multicast address.

rakem

Any free applications that i can use?
If you just ping can you test IGMP leave and join messages?

Turgon

rakem wrote: »

Any free applications that i can use?
If you just ping can you test IGMP leave and join messages?

Dont get hung up on applications and labs and dont get hung up on multicast generally. Study enough and move on, few points..

Just add an interface to a group using ip igmp join and ping that address.

Sirsamon

i normally fire up VLC media player and stream something to the network, just remember VLC had or may still have a ttl bug, at the last part of streaming enter you own ttl it works fine.

rakem

Turgon wrote: »

Dont get hung up on applications and labs and dont get hung up on multicast generally.

I like to lab up everything! I learn by doing better than by reading...
From the CCIE blueprint though it doesn't look like Multicast is heavily tested.

Sirsamon wrote: »

i normally fire up VLC media player and stream something to the network, just remember VLC had or may still have a ttl bug, at the last part of streaming enter you own ttl it works fine.

I have tried VLC - what method do you use to stream? HTTP, RTP etc... it has many options.

rakem

home late from work... tired. Lots of F5 issues today.
Just an hour of reading.

rakem

Not the best week for study this week, been very busy.

Hopefully can make up for in over the weekend.

Turgon

rakem wrote: »

Not the best week for study this week, been very busy.

Hopefully can make up for in over the weekend.

Its probably best to start hammering the Odom and Boson practice tests for a few weeks supplemented with some reading. After a month of that you should be close and the labbing will add more value as the theory really clicks in.

rakem

Turgon wrote: »

Its probably best to start hammering the Odom and Boson practice tests for a few weeks supplemented with some reading. After a month of that you should be close and the labbing will add more value as the theory really clicks in.

The Boson one is the CD that comes with the cert guide right?
i do use that a fair bit already.

The Odom questions are the ones at the start of each the chapter?

rakem

A good 3 1/2 hours of study this morning.
Trying to finish up multicast this weekend.

Some of the multicast questions on the boson CD are very tricky, need to read them very carefully.

Turgon

rakem wrote: »

The Boson one is the CD that comes with the cert guide right?
i do use that a fair bit already.

The Odom questions are the ones at the start of each the chapter?

Yup. The CD that comes with the cert guide has a test engine including all the end of chapter questions and a final exam. I would also invest in the Boson for the written that you can buy from the Boson site. I think you get a discount if you register Odoms book.

jamesp1983

Turgon wrote: »

Yup. The CD that comes with the cert guide has a test engine including all the end of chapter questions and a final exam. I would also invest in the Boson for the written that you can buy from the Boson site. I think you get a discount if you register Odoms book.

Turgon is right about the discount.

rakem

Tired today - Had a late night last night due to a major outage in our primary DC (bloody junos bugs)

Anyway - been doing the questions from the CD and the book for mulitcast this morning. Doing pretty well on the CD, missed a couple from the book. Overall pretty good.

One question that is driving me crazy from the CD which comes with the CCIE cert guide. The question ID is CD-17-03-1 (if you start the full exam you can jump to this question by clicking exam > jump to then entering CD-17-03-1 in the question ID)

I don't undertstand why C is not correct.
There are no hosts wanting traffic on LAN4 so C would have pruned all its links. Right?
Even thou S0/2 would be its RPF link it can still send prunes out this link.

R1 wouldn't be sending prues out Fa0/0 since that interface connects to the source. The assert process would make R2 the forwarder, but i don't think that means R1 would send prunes out its Fa0/0 (please correct me if i'm wrong here.

So i believe the answers should be:

B - Just before the Join, R5 cannot list its S0/2 interfaceas an outgoing interfcae
C - For the past four hours R6 could have been sending prune message every three minutes for group 229.1.1.1 out S0/3

Would appreciate it if someone could have a look at this question and do a bit of a sanity check.

Thanks,