It begins.

rakem

So I have been putting it off for a while now, but my last bunch of text books arrived in the mail last night so i have no excuse to put it off any longer.... Officially began studing for my CCIE last night.

My only goal at the moment is to have the written exam passed by the time my CCNP expires in mid 2012. A long way away i know but i don't get to much spare time.

So, diving straight into routing, BGP first as its on the cards at work at the moment. Will use the office exam cert guide, along with TCP/IP routing 2, and also the BSCI book which has some great info.

Last night was a bit of a review on basic routing, principals, Classless vs Classfull, VLSM, CIDR, summerization etc etc. I'll be using dynamips fairly heavily for the routing stuff, as its just nice and easy and i can run around 10 routers on my PC.

So thats it for now, I'm at work and should probably stop laughing so hard at this youtube clip 'the day the routers died'

YouTube - The Day The Routers Died...

Turgon

rakem wrote: »

So I have been putting it off for a while now, but my last bunch of text books arrived in the mail last night so i have no excuse to put it off any longer.... Officially began studing for my CCIE last night.

My only goal at the moment is to have the written exam passed by the time my CCNP expires in mid 2012. A long way away i know but i don't get to much spare time.

So, diving straight into routing, BGP first as its on the cards at work at the moment. Will use the office exam cert guide, along with TCP/IP routing 2, and also the BSCI book which has some great info.

Last night was a bit of a review on basic routing, principals, Classless vs Classfull, VLSM, CIDR, summerization etc etc. I'll be using dynamips fairly heavily for the routing stuff, as its just nice and easy and i can run around 10 routers on my PC.

So thats it for now, I'm at work and should probably stop laughing so hard at this youtube clip 'the day the routers died'

YouTube - The Day The Routers Died...

Good luck. Keep this blog updated and log your hours in the signature. I find it keeps me motivated and its nice to have people drop by!

rakem

Turgon wrote: »

Good luck. Keep this blog updated and log your hours in the signature. I find it keeps me motivated and its nice to have people drop by!

Will do, i like stats so it will probably keep me motivated as well.

SysAdmin4066

Nice to have some more company, welcome to the officers mess as Turgon would say

rakem

Thanks guys.... looking forward to getting back into it!

networker050184

If you are studying BGP definitely read Internet Routing Architectures. The book is full of useful info and not just tech specs. Its a lot better read than any of those resources you have listed.

notgoing2fail

Good luck! Let's hope you can pull this thing off....

It always gives me energy to hear people studying the CCIE...I know I am going to have to face that battle one day myself....


Right now I live vicariously through everyone else.......

gorebrush

There is quite a collection of us all with the CCIE in mind nowadays

burbankmarc

I'm right there with you. But I think I'm going to do QoS first since I have a project at work that calls for it. Just waiting for my Odom book.

Good luck, I hear this thing is tad hard...

chmorin

notgoing2fail wrote: »

Good luck! Let's hope you can pull this thing off....

It always gives me energy to hear people studying the CCIE...I know I am going to have to face that battle one day myself....


Right now I live vicariously through everyone else.......

I second this. Watching you guys pursue such events is amazing and inspiring even. I can only hope that I will someday be able to join the 'officers mess'

jason_lunde

rakem wrote: »

So thats it for now, I'm at work and should probably stop laughing so hard at this youtube clip 'the day the routers died'

YouTube - The Day The Routers Died...

Just wanted to wish you good luck man, and thank your for the link. Made my f'in day.

CCIEWANNABE

good luck bro, and don't give up. been studying for the lab since last Sept and will be taking the lab in a few months. The journey is long, but the outcome is definitely worth it.

Turgon

rakem wrote: »

Will do, i like stats so it will probably keep me motivated as well.

Good man. Start a counter in the signature. Knuckle down at the weekends on the studies. It's where the CCIE is won.

rakem

Turgon wrote: »

Good man. Start a counter in the signature. Knuckle down at the weekends on the studies. It's where the CCIE is won.

Well its been a pretty bad start, i've come down with one hell of a cold so not doing to much study this weekend.

One thing that I have discovered that I thought was pretty cool, is a public available route-server where you can see the entire internet BGP table, run some show commands, trace routes etc. Telnet to route-server.ip.att.net

Turgon

rakem wrote: »

Well its been a pretty bad start, i've come down with one hell of a cold so not doing to much study this weekend.

One thing that I have discovered that I thought was pretty cool, is a public available route-server where you can see the entire internet BGP table, run some show commands, trace routes etc. Telnet to route-server.ip.att.net

Sorry to hear that. Good to see the study count begin!

peanutnoggin

rakem wrote: »

Well its been a pretty bad start, i've come down with one hell of a cold so not doing to much study this weekend.

One thing that I have discovered that I thought was pretty cool, is a public available route-server where you can see the entire internet BGP table, run some show commands, trace routes etc. Telnet to route-server.ip.att.net

Yeah, there is a bunch of looking glass servers available on the interwebs. Check out Traceroute.org

It's pretty cool stuff...

-Peanut

EDIT:

Hope you feel better!!!

rakem

Still feeling like **** but did a few hours tonight.

Went over the BGP basics. Message types and BGP states mainly.

I Didn't know that you could actually set the BGP keepalives to 0, so they will never be sent.... can't think of any reason why anyone would actually want it configure it that way.

Turned on GNS3 and just set up a simple two router lab so I could so some debugs and watch the routers go through their different states. Did some incorrect configs so i could see the different types of errors and notifications.

Example Below - Message type 1 was sent (Open Message) which identifies the router and specifies the BGP settings (timers, AS number, etc)

Message type 3 received (notification) which sent the router into the Active state because I configured the neighbour statement with the wrong AS number

*Jun 28 22:02:43.391: BGP: 10.1.1.2 send message type 1, length (incl. header) 45
*Jun 28 22:02:43.479: BGP: 10.1.1.2 rcv message type 3, length (excl. header) 4
*Jun 28 22:02:43.479: %BGP-3-NOTIFICATION: received from neighbor 10.1.1.2 2/2 (peer in wrong AS) 2 bytes FDE8

I think this is a good way to study.... read the theory, take notes, the hammer it home with some labs.

rakem

BGP path attributes tonight.

Well known mandatory
- Next Hop - self explanatory
- AS Path - list of AS to a destination
- Origin - IGP, EGP, incomplete - redistribution normally

Well know discretionary
- Local Pref - influence outbound routes
- Atomic Aggregate - alerts downstream routers about potential loss of path info because of aggregation

Optional Transitive
- Community - used for destination routes that share common properties and policies
- Aggregrator - provides info as to where aggregation was performed.

Optional Nontransitive
- MED - influence inbound routes
- Originator_ID - loop prevention (more reading to come on this)
- Cluster_List - loop prevention (more reading to come on this)

AS_Set & AS_Sequence (Loop avoidance when aggregate routes are sent, well AS_Set anyway)

Administrative Weight (Cisco Specific used for route preference - local router only)

gorebrush

rakem wrote: »

I think this is a good way to study.... read the theory, take notes, the hammer it home with some labs.

This has been my proven, tried and tested method since December 2004

18 Certification exams later can't be wrong

Luckycharms

I would suggest to follow turgon's advise of keeping this blog updated and log your hours in your signature. It will be eye opening when you get done. But best of luck, consistency and determination will get you that number.

rakem

Played around with a simple 3 router lab in GNS3. Mucked around with aggregating routes and using the as-set command.

Also did a lot of reading on the BGP decision making process, the CCIE cert guide has a really good section on every step, with config examples (Page 460 - 476).

My next step is to lab these examples up and manipulate BGP to use different decision making steps. Should be fun.

Lunch time now, then off to see some houses (trying to buy a new house at the moment)
Hopefully will get time for the above lab exercises later today, if not tomorrow.

rakem

got a decent sized lab going in GNS3.... established basic BGP peering between all routers, starting to play around with PAs.

lab screenshot attached

rakem

Did a bunch of reading on managing large scale BGP implementations.....

Peer groups
Route Reflectors
Communities
Confederations.

Looking forward to labing up confederations.

My study so far has been 90% from TCP/IP Volume 2, Chapter 2 'Introduction to BGP'
Once finished with that, chapter 3 looks good for a lot of practical exercises to help compound the theory.

The CCIE cert guide will be very heavily used during the practical exercises as well.

rakem

Did some practice BGP questions tonight..... got lots wrong! lol

rakem

hmmm never seen this one before...

On one of our BGP routers at work, i was just having a bit of a snoop around, and typed the following

Primary_RTR#sh bgp 

% NOTE: This command is deprecated. Please use 'show bgp ipv6 unicast'
Primary_RTR#
Primary_RTR#

whats the deal there?

Firstly the router is not even running IPV6 so not sure why its telling me to use an ipv6 command... But to show the normal output of the 'sh bgp' command i had to do:

Primary_RTR#sh bgp ipv4 unicast summary 
BGP router identifier 10.8.0.8, local AS number 65500
BGP table version is 742, main routing table version 742
89 network entries using 10413 bytes of memory
179 path entries using 9308 bytes of memory
12/6 BGP path/bestpath attribute entries using 1488 bytes of memory
2 BGP AS-PATH entries using 48 bytes of memory
0 BGP route-map cache entries using 0 bytes of memory
0 BGP filter-list cache entries using 0 bytes of memory
BGP using 21257 total bytes of memory
BGP activity 182/93 prefixes, 716/537 paths, scan interval 60 secs

Neighbor        V    AS MsgRcvd MsgSent   TblVer  InQ OutQ Up/Down  State/PfxRcd
10.8.2.14    4 65500  150450  150480      742    0    0 14w6d          89
83.xx.xx.xxx   4  5466       0       0        0    0    0 never    Idle
83.xx.xx.xxx   4  5466 1152573 1800485      742    0    0 2w1d           85
% NOTE: This command is deprecated. Please use 'show bgp ipv6 unicast'
CoLo_Primary_RTR#
CoLo_Primary_RTR#

Router is a 2811 running 12.3(14)T5.... anyone seen this before?

stuh84

Is it not sh ip bgp summary for the same information?

billscott92787

I'm another in the 'officers mess' I got all my materials and busted open the CCIE Cert Guide this morning to take a look around. Started reading just the basic overviews of the 1st chapter. Not quite sure where I want to begin my studies, I haven't set a structure, I was just going to start at the beginning and work my way through. I got about half an hour or so this morning reading, tomorrow morning I'll be up at 5 AM bright and early hitting the book again. My goal is to pass the CCIE Written by December 2010.

rakem

stuh84 wrote: »

Is it not sh ip bgp summary for the same information?

either command. they show the same thing

rakem

So had to get my wisdom teeth pulled out today, got 3 days of work though!

Trying to muck around with a simple BGP lab and i'm getting a bit confused as to why a particular route is being chosen. (This should be simple, its probably just the pain killers i'm, currently on)

Anyway so i have a simple 3 router lab, connected in a triangle config.
R0 and R1 and iBGP peers
R0 and R1 and eBGP peers with R2

R2 has advertised a route to network 90.90.90.0/24

Both R0 and R1 have two routes to this destination (through the eBGP peer, or iBGP)
The local_pref for the iBGP routes is 100, but the routers have chosen the eBGP route which has a local pref of 0.

Local_pref is high in the BGP decision making process, i know eBGP routes are prefered over iBGP but local_pref should take preference(?)

Heres the config from the routers

R0#sh run | section  bgp
router bgp 10
 no synchronization
 bgp log-neighbor-changes
 redistribute connected
 neighbor 10.1.0.1 remote-as 10
 neighbor 10.2.2.2 remote-as 2
 no auto-summary
R0#sh ip int br
Interface                  IP-Address      OK? Method Status                Protocol
FastEthernet0/0            10.2.2.1        YES manual up                    up
FastEthernet0/1            10.1.0.2        YES manual up                    up


R0#
   Network          Next Hop            Metric LocPrf Weight Path
*> 10.1.0.0/24      0.0.0.0                  0         32768 ?
* i                 10.1.0.1                 0    100      0 ?
*>i10.1.2.0/24      10.1.0.1                 0    100      0 ?
*                   10.2.2.2                 0             0 2 ?
*  10.2.2.0/24      10.2.2.2                 0             0 2 ?
*>                  0.0.0.0                  0         32768 ?
* i90.90.90.0/24    10.1.2.2                 0    100      0 2 ?
*>                  10.2.2.2                 0             0 2 ?
R0#

R1#sh run | section bgp
router bgp 10
 no synchronization
 bgp log-neighbor-changes
 redistribute connected
 neighbor 10.1.0.2 remote-as 10
 neighbor 10.1.2.2 remote-as 2
 no auto-summary

R1#sh ip int bri
Interface                  IP-Address      OK? Method Status                Protocol
FastEthernet0/0            10.1.2.1        YES manual up                    up
FastEthernet0/1            10.1.0.1        YES manual up                    up
R1#


   Network          Next Hop            Metric LocPrf Weight Path
* i10.1.0.0/24      10.1.0.2                 0    100      0 ?
*>                  0.0.0.0                  0         32768 ?
*  10.1.2.0/24      10.1.2.2                 0             0 2 ?
*>                  0.0.0.0                  0         32768 ?
*  10.2.2.0/24      10.1.2.2                 0             0 2 ?
*>i                 10.1.0.2                 0    100      0 ?
*> 90.90.90.0/24    10.1.2.2                 0             0 2 ?
* i                 10.2.2.2                 0    100      0 2 ?


R2#sh  run | section bgp
router bgp 2
 no synchronization
 bgp log-neighbor-changes
 redistribute connected
 neighbor 10.1.2.1 remote-as 10
 neighbor 10.2.2.1 remote-as 10
 no auto-summary
R2#
R2#sh ip int br
Interface                  IP-Address      OK? Method Status                Protocol
FastEthernet0/0            10.1.2.2        YES manual up                    up
FastEthernet0/1            10.2.2.2        YES manual up                    up
Loopback10                 90.90.90.90     YES manual up                    up
R2#

   Network          Next Hop            Metric LocPrf Weight Path
*  10.1.0.0/24      10.1.2.1                 0             0 10 ?
*>                  10.2.2.1                 0             0 10 ?
*  10.1.2.0/24      10.1.2.1                 0             0 10 ?
*                   10.2.2.1                               0 10 ?
*>                  0.0.0.0                  0         32768 ?
*  10.2.2.0/24      10.1.2.1                               0 10 ?
*>                  0.0.0.0                  0         32768 ?
*                   10.2.2.1                 0             0 10 ?
*> 90.90.90.0/24    0.0.0.0                  0         32768 ?
R2#

So the routers that R0 and R1 have chosen are the best routes, since they are just one hop away, but why is local_pref being ignored here?
Something do with with the AD of iBGP being 200 and the AD of eBGP being 20??

Remember, im on lots of pain killers and feel a bit stupid at the moment... lol

Thanks

kalebksp

It's been a while since I was messing with BGP, but I believe that a blank Local Pref doesn't mean that it's zero, but rather that it's equal to the default, which is 100. Since the local pref is equal it ends up selecting the path based on eBGP vs iBGP.

billscott92787

kalebksp wrote: »

It's been a while since I was messing with BGP, but I believe that a blank Local Pref doesn't mean that it's zero, but rather that it's equal to the default, which is 100. Since the local pref is equal it ends up selecting the path based on eBGP vs iBGP.

+1 for this answer. IF you don't set the local preference it is defaulted out at 100. You can confirm this here.

Prefer the path with the highest LOCAL_PREF.
Note: A path without LOCAL_PREF is considered to have had the value set with the bgp default local-preference command, or to have a value of 100 by default.


So the local preferences are matching, which leads to the route being chosen as preferring the eBGP path, just as kalebksp said.