BGP MED

BGP MED is an optional non-transitive attribute meaning its not propagated throughout the whole internet but just to adjacent AS. The word "optional" means that this is not necessarily by default sent with the BGP updates. The purpose of MED is to influence how other autonomous systems enter your AS to reach a certain prefix. If the other attributes are set to default, MED will be the attribute used for path selection however, if Weight or Local preference is configured on the adjacent AS router, then MED will not be selected. The lower the MED the more preferred the path will be.

Acronym for Multi Exit Discriminator and otherwise known as "Metric" in the BGP table.
The lower the MED the more preferred.
It is an optional non-transitive attribute.
Can dictate how other AS enter your AS.

Configure R4 so that it will advertise a MED value of 30 to R3 and 20 to R2. Afterwards,
tweak the route map to set a MED of 10 for network 144.144.144.144/32 in R4 towards R3. The end
result should have traffic from AS123 to 4.4.4.4/32 and 44.44.44.44/32 take R2 but traffic for
144.144.144.144/32 should take R3. At the start BGP has been established on all routers.

To start lets configure 2 route maps and set the MED as required.

R4(config)#access-list 20 permit any
R4(config)#access-list 30 permit any
R4(config)#route-map R2SETMED20 permit 10
R4(config-route-map)#match ip address 20
R4(config-route-map)#set metric 20
R4(config-route-map)#route-map R3SETMED30 permit 10
R4(config-route-map)#match ip address 30         
R4(config-route-map)#set metric 30

Access-list matches all routes. Now, lets apply the route map to the neighbors.

R4(config-router)#neighbor 24.24.24.2 route-map R2SETMED20 out
R4(config-router)#neighbor 34.34.34.3 route-map R3SETMED30 out

Its in the "out" direction because R4 is the one advertising the routes. Let's check what R2 and R3 see in their BGP table.

R2#sh ip bgp
BGP table version is 31, local router ID is 24.24.24.2
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
             r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete

  Network          Next Hop            Metric LocPrf Weight Path
*> 4.4.4.4/32       24.24.24.4              20             0 4 i
*> 44.44.44.44/32   24.24.24.4              20             0 4 i
*> 144.144.144.144/32
                   24.24.24.4              20             0 4 i

R3#sh ip bgp
BGP table version is 10, local router ID is 34.34.34.3
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
             r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete

  Network          Next Hop            Metric LocPrf Weight Path
*> 4.4.4.4/32       34.34.34.4              30             0 4 i
*> 44.44.44.44/32   34.34.34.4              30             0 4 i
*> 144.144.144.144/32
                   34.34.34.4              30             0 4 i

Ok, its clear that the Metric now is changed to 20 and 30 for R2 and R3 respectively. Let's see how R1 sees the routes.

R1#sh ip bgp
BGP table version is 34, local router ID is 13.13.13.1
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
             r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete

  Network          Next Hop            Metric LocPrf Weight Path
*>i4.4.4.4/32       24.24.24.4              20    100      0 4 i
* i                 34.34.34.4              30    100      0 4 i
*>i44.44.44.44/32   24.24.24.4              20    100      0 4 i
* i                 34.34.34.4              30    100      0 4 i
*>i144.144.144.144/32
                   24.24.24.4              20    100      0 4 i
* i                 34.34.34.4              30    100      0 4 i

R1 sees two paths, but notice that the paths with ">" are with those with lower MED. Therefore it is going to take those paths. Now, lets configure R4 so that the path R1 will take to 144.144.144.144/32 in R4 will be through R3 then clear the BGP session after.

R4(config)#access-list 33 permit 144.144.144.144 0.0.0.0
R4(config)#route-map R3SETMED30 permit 10
R4(config-route-map)#match ip address 33
R4(config-route-map)#set metric 10
R4(config-route-map)#route-map R3SETMED30 permit 20
R4(config-route-map)#match ip address 30
R4(config-route-map)#set metric 30
R4#clear ip bgp *

Now, lets see the outputs.

R2#sh ip bgp | beg Network
  Network          Next Hop            Metric LocPrf Weight Path
*> 4.4.4.4/32       24.24.24.4              20             0 4 i
*> 44.44.44.44/32   24.24.24.4              20             0 4 i
*> 144.144.144.144/32
                   24.24.24.4              20             0 4 i

R3#sh ip bgp | beg Network
  Network          Next Hop            Metric LocPrf Weight Path
*> 4.4.4.4/32       34.34.34.4              30             0 4 i
*> 44.44.44.44/32   34.34.34.4              30             0 4 i
*> 144.144.144.144/32
                    34.34.34.4              10             0 4 i

R1#sh ip bgp | beg Network
  Network          Next Hop            Metric LocPrf Weight Path
* i4.4.4.4/32       34.34.34.4              30    100      0 4 i
*>i                 24.24.24.4              20    100      0 4 i
* i44.44.44.44/32   34.34.34.4              30    100      0 4 i
*>i                 24.24.24.4              20    100      0 4 i
*>i144.144.144.144/32
                    34.34.34.4              10    100      0 4 i
* i                 24.24.24.4              20    100      0 4 i

We can see a change in R3 that shows a MED of 10 for 144.144.144.144/32, while R1 shows the same and will now take R3 to reach the subnet.

BGP Local Preference

Local Preference is one of the ways to alter the path taken by one AS to reach another AS. The difference between Local Preference and Weight is that Weight is just locally signifant in the router while Local Preference is what I call "Local AS significant". What I mean by that is that Local Preference is being propagated Intra AS but not outside the AS.

Local preference is to influence your own AS how to get or exit to another AS.
MED is to influence other AS how to enter your AS.
The higher the local preference, the more preferred.

In this lab, local preference will be configured and will be using route map for more flexibility. Check the diagram below for details.

R1, R2 and R3 belongs to AS 123. R4 is in AS4 and is advertising 4.4.4.4/32, 44.44.44.44/32 and 144.144.144.144/32
subnets.We need to set all routes learned from R3 to have local preference value of 300. After which, configure a
route-map that will assign a local preference of 500 in R2 for the network 144.144.144.144/32.

Here are the initial BGP configurations on the routers.

R1#
!
router bgp 123
no synchronization
bgp log-neighbor-changes
neighbor 12.12.12.2 remote-as 123
neighbor 13.13.13.3 remote-as 123
no auto-summary

R2#
!
router bgp 123
no synchronization
bgp log-neighbor-changes
neighbor 12.12.12.1 remote-as 123
neighbor 12.12.12.1 next-hop-self
neighbor 24.24.24.4 remote-as 4
no auto-summary

R3#
!
router bgp 123
no synchronization
bgp log-neighbor-changes
neighbor 13.13.13.1 remote-as 123
neighbor 13.13.13.1 next-hop-self
neighbor 34.34.34.4 remote-as 4
no auto-summary

R4#
!
router bgp 4
no synchronization
bgp log-neighbor-changes
network 4.4.4.4 mask 255.255.255.255
network 44.44.44.44 mask 255.255.255.255
network 144.144.144.144 mask 255.255.255.255
neighbor 24.24.24.2 remote-as 123
neighbor 34.34.34.3 remote-as 123
no auto-summary

Let's see what is the best path taken by R1 to reach the networks advertised by R4.

R1#sh ip bgp

BGP table version is 10, local router ID is 1.1.1.1
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
             r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete

  Network          Next Hop            Metric LocPrf Weight Path
*>i4.4.4.4/32       12.12.12.2               0    100      0 4 i
* i                 13.13.13.3               0    100      0 4 i
*>i44.44.44.44/32   12.12.12.2               0    100      0 4 i
* i                 13.13.13.3               0    100      0 4 i
*>i144.144.144.144/32
                   12.12.12.2               0    100      0 4 i
* i                 13.13.13.3               0    100      0 4 i

R1#sh ip route | beg Gateway


Gateway of last resort is not set

    1.0.0.0/32 is subnetted, 1 subnets
C       1.1.1.1 is directly connected, Loopback0
    4.0.0.0/32 is subnetted, 1 subnets
B       4.4.4.4 [200/0] via 12.12.12.2, 00:04:00
    144.144.0.0/32 is subnetted, 1 subnets
B       144.144.144.144 [200/0] via 12.12.12.2, 00:04:00
    12.0.0.0/24 is subnetted, 1 subnets
C       12.12.12.0 is directly connected, Serial1/2
    13.0.0.0/24 is subnetted, 1 subnets
C       13.13.13.0 is directly connected, Serial1/3
    44.0.0.0/32 is subnetted, 1 subnets
B       44.44.44.44 [200/0] via 12.12.12.2, 00:04:00

It's clear that it prefers to take R2 to reach the networks in R4. Let's configure R3 so that all routes received by R3 will have a Local preference of 300

R3#config t
R3(config)#router bgp 123
R3(config-router)#bgp default local-preference 300

R1#sh ip bgp
BGP table version is 13, local router ID is 1.1.1.1
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
             r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete

  Network          Next Hop            Metric LocPrf Weight Path
* i4.4.4.4/32       12.12.12.2               0    100      0 4 i
*>i                 13.13.13.3               0    300      0 4 i
* i44.44.44.44/32   12.12.12.2               0    100      0 4 i
*>i                 13.13.13.3               0    300      0 4 i
* i144.144.144.144/32
                   12.12.12.2               0    100      0 4 i
*>i                 13.13.13.3               0    300      0 4 i

Immediately, even without clearing the BGP process, R3 now became the more preferred path to reach R4. Remember that in Local Preference, the higher the value, the more preferred.

Let's configure in R2 a route-map so the network 144.144.144.144/32 will have a local preference of 500. This will make R2 the best path to reach the mentioned network.

R2(config)#access-list 1 permit host 144.144.144.144
R2(config)#route-map LOCALPREF500 permit 10
R2(config-route-map)#match ip address 1
R2(config-route-map)#set local-preference 500
R2#(config)# router bgp 123
R2(config-router)#neighbor 24.24.24.4 route-map LOCALPREF500 in
R2# clear ip bgp *

R1#sh ip bgp
BGP table version is 14, local router ID is 1.1.1.1
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
             r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete

  Network          Next Hop            Metric LocPrf Weight Path
*>i4.4.4.4/32       13.13.13.3               0    300      0 4 i
*>i44.44.44.44/32   13.13.13.3               0    300      0 4 i
*>i144.144.144.144/32
                   12.12.12.2               0    500      0 4 i
* i                 13.13.13.3               0    300      0 4 i

Firstly, why is that the route-map has an "in" direction. It's because we are receiving the route from another router, R2 is not the one advertising. You can see now that path to 144.144.144.144/32 will have R2 as the next hop. If you noticed also, why is that for 4.4.4.4/32 and 44.44.44.44/32, there is no other path except through 13.13.13.3. Check the route-map above and you'll find the answer. There is no succedding line after line 10, which means, it will block out the subnets and not advertised it to R1.

Let's configure the 20th sequence of the route-map.

R2(config)#route-map LOCALPREF500 permit 20

Let's see now if there are changes.

R1#sh ip bgp
BGP table version is 16, local router ID is 1.1.1.1
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
             r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete

  Network          Next Hop            Metric LocPrf Weight Path
* i4.4.4.4/32       12.12.12.2               0    100      0 4 i
*>i                 13.13.13.3               0    300      0 4 i
* i44.44.44.44/32   12.12.12.2               0    100      0 4 i
*>i                 13.13.13.3               0    300      0 4 i
*>i144.144.144.144/32
                   12.12.12.2               0    500      0 4 i
* i                 13.13.13.3               0    300      0 4 i

That's it for local preference. Hopefully I can finish the BGP topics soon. :)

BGP Best Path Selection

Studying the BSCI for CCNP can be hard for the reason that you need to memorize some stuff like BGP Best Path Selection. Few months ago I found this mnemonic helpful to memorize the BGP Path Selection Process. It's easy as eating pancakes and it is We Love Oranges AS Oranges Mean Pure Refreshment. By taking note of the letters in caps, one will immediately remember how BGP selects the best path by the list below.

“We Love Oranges AS Oranges Mean Pure Refreshment”
W	Weight (Highest)
L	LOCAL_PREF (Highest)
O	Originate (local) routes that are advertise through the "network" command or redistributed from an IGP.
AS	AS_PATH (shortest)
O	ORIGIN Code (IGP > EGP > Incomplete)
M	MED (lowest)
P	Paths (External > Internal)
R	RID (lowest)

That was easy! So from now on, every time you see an orange, BGP will always come in mind! Cheers!

Finally A CCNP!

It has been almost 2 years since I became a CCNA. I took and passed the exam May 2007 and waited for more than 1 year to begin my CCNP quest. My reason of not being able to take the exams was not having enough money to pay them. My CCNA paved a way for me to get a job in a company that encourages its employees to go for certifications and paying for them. The opportunities of being a CCNA are huge, how much more now that I got the CCNP.

Now the question is, what's next? Is everything over now I got this certification? Certainly not! I have been readings posts to find out what is the best field to study in preparation for CCIE and I found out that getting a CCDP and other professional Cisco certs are good ways to prepare for CCIE R&S. My level of excitement did not go down when I got my CCNP but I am rather excited to pursue more and learn more. The more I learn, the more I know how much I don't know. Makes sense? :)

Now as promised, I will be posting some of the labs I have tried using dynamips and practical tips to be used in the enterprise environment. Even if nobody reads this blog, this will serve as my personal reference once I forget about something. Cheers!