BGP MED attribute

First I would like to make a quick review on BGP Best path selection process:

BGP assigns the first valid path as the current best path. BGP then compares the best path with the next path in the list, until BGP reaches the end of the list of valid paths. This list provides the rules that are used to determine the best path:

1. Prefer the path with the highest WEIGHT.
2. Prefer the path with the highest local preference (default value is 100)
3. Prefer the path that was locally originated via a network or aggregate BGP subcommand or through redistribution from an IGP.

Note:  Local paths that are sourced by the network or redistribute commands are preferred over local aggregates that are sourced by the aggregate-address command.

4. Prefer the path with the shortest AS_PATH.
5. Prefer the path with the lowest origin type.

Note: IGP is lower than EGP, and EGP is lower than INCOMPLETE.

6. Prefer the path with the lowest MED
7. Prefer eBGP over iBGP paths.
8. Prefer the path with the lowest IGP metric to the BGP next hop.
9. Determine if multiple paths require installation in the routing table for BGP Multipath.
10. When both paths are external, prefer the path that was received first (the oldest one).
11. Prefer the route that comes from the BGP router with the lowest router ID.
12. If the originator or router ID is the same for multiple paths, prefer the path with the minimum cluster list length.
13. Prefer the path that comes from the lowest neighbor address.

Reference: http://www.cisco.com/en/US/tech/tk365/technologies_tech_note09186a0080094431.shtml

In this lab i'm going to show how MED affect the path selection. MED which is a non-transitive, optional, attribute can be used by BGP peers to influence the inbound path selection.

This is the network topology i'm going to use:

 

In the following lab, R6 is adverting prefix 192.168.61.0/24, where R1 is learning this prefix from R2, R3 and R5, which are all external BGP, peers (eBGP), and choose R5 as the best path:

R1#sh ip bgp 192.168.61.0/24 BGP routing table entry for 192.168.61.0/24, version 2 Paths: (3 available, best #3, table Default-IP-Routing-Table) Flag: 0x820   Advertised to update-groups:      1           65005 65006     10.1.15.5 from 10.1.15.5 (10.1.56.5)       Origin IGP, localpref 100, valid, external   65003 65004 65006     10.1.13.3 from 10.1.13.3 (10.1.34.3)       Origin IGP, localpref 100, valid, external   65002 65006     10.1.12.2 from 10.1.12.2 (2.2.2.2)       Origin IGP, localpref 100, valid, external, best

 

So who R1 is choose the best path?

R1 is comparing R5 path to R3 path and using the BGP bets path process list he choose the winner – in this case R5 due to rule no. 4.

Then compare R5 path to R2 path – both have the same path characteristic and R2 is chosen as the winner following rule no. 13.

If I will shut down R2 the path from R5 will be the preferred one:

R1# *Mar  4 18:10:55.145: %BGP-5-ADJCHANGE: neighbor 10.1.12.2 Down Peer closed the session   R1#show ip bgp 192.168.61.0/24 BGP routing table entry for 192.168.61.0/24, version 3 Paths: (2 available, best #1, table Default-IP-Routing-Table) Flag: 0x820   Advertised to update-groups:      1           65005 65006     10.1.15.5 from 10.1.15.5 (10.1.56.5)       Origin IGP, localpref 100, valid, external, best   65003 65004 65006     10.1.13.3 from 10.1.13.3 (10.1.34.3)       Origin IGP, localpref 100, valid, external

 

But if I bring back R2, it won’t change that, and R5 will continue to be the bets path:

R1# *Mar  4 18:11:18.365: %BGP-5-ADJCHANGE: neighbor 10.1.12.2 Up   R1#sh ip bgp 192.168.61.0/24 BGP routing table entry for 192.168.61.0/24, version 3 Paths: (3 available, best #2, table Default-IP-Routing-Table) Flag: 0x820   Advertised to update-groups:      1           65002 65006     10.1.12.2 from 10.1.12.2 (2.2.2.2)       Origin IGP, localpref 100, valid, external   65005 65006     10.1.15.5 from 10.1.15.5 (10.1.56.5)       Origin IGP, localpref 100, valid, external, best   65003 65004 65006     10.1.13.3 from 10.1.13.3 (10.1.34.3)       Origin IGP, localpref 100, valid, external

 

The reason will be rule no. 10 – the oldest learned path.

Now let’s configure some metrics and these routers, showing R2 as an example:

R2(config)# ip prefix-list PL_NET61 per 192.168.61.0/24 R2(config)# route-map RM_R1 permit 10 R2(config-route-map)# match ip address prefix-list PL_NET61 R2(config-route-map)# set metric 1000 R2(config-route-map)# exit R2(config-route-map)# route-map RM_R1 permit 20 R2(config-route-map)# router bgp 65002 R2(config-router)# neighbor 10.1.12.1 route-map RM_R1 out

 

 

R3 and R5 are configured with the same beside the metric which is 300 and 500 respectively.

After clearing the BGP peers from R1:

R1#sh ip bgp 192.168.61.0/24 BGP routing table entry for 192.168.61.0/24, version 2 Paths: (3 available, best #3, table Default-IP-Routing-Table)   Advertised to update-groups:      1           65003 65004 65006     10.1.13.3 from 10.1.13.3 (10.1.34.3)       Origin IGP, metric 300, localpref 100, valid, external   65005 65006     10.1.15.5 from 10.1.15.5 (10.1.56.5)       Origin IGP, metric 500, localpref 100, valid, external   65002 65006     10.1.12.2 from 10.1.12.2 (2.2.2.2)       Origin IGP, metric 1000, localpref 100, valid, external, best

 

R2 is still the best path although his metric is the highest among the peers, the reason is that R1 isn’t comparing MED from peers from different AS’s (following rule no. 6). This comparison will occur only if the first (the neighboring) AS is the same in the two paths.

So now let’s configure R1 to compare the MED although it comes from different AS’s:

R1# conf t Enter configuration commands, one per line.  End with CNTL/Z. R1(config)#router bgp 65001 R1(config-router)#bgp always-compare-med

 

Clear the BGP neighbor list and look again on the BGP table:

R1#sh ip bgp 192.168.61.0/24 BGP routing table entry for 192.168.61.0/24, version 2 Paths: (3 available, best #3, table Default-IP-Routing-Table) Flag: 0x820   Advertised to update-groups:      1           65002 65006     10.1.12.2 from 10.1.12.2 (2.2.2.2)       Origin IGP, metric 1000, localpref 100, valid, external   65003 65004 65006     10.1.13.3 from 10.1.13.3 (10.1.34.3)       Origin IGP, metric 300, localpref 100, valid, external   65005 65006     10.1.15.5 from 10.1.15.5 (10.1.56.5)       Origin IGP, metric 500, localpref 100, valid, external, best

 

In MED lower is better but we can see that R5 has chosen as the best path although R3 has much lower metric, the reason is rule no. 4 – AS_PATH, which comes before the rule no. 6 -MED, and we can see that R3 as-path is much longer.

We can configure R1 to ignore the AS_PATH attribute using a hidden command:

R1# conf t Enter configuration commands, one per line.  End with CNTL/Z. R1(config)#router bgp 65001 R1(config-router)# bgp bestpath as-path ignore

 

This command is hidden since IOS 12.4 but it’s still valid and we can use it.

Let’s clear the BGP neighbors and look on R1 BGP table:

R1#sh ip bgp 192.168.61.0/24 BGP routing table entry for 192.168.61.0/24, version 2 Paths: (3 available, best #1, table Default-IP-Routing-Table) Flag: 0x820   Advertised to update-groups:      1           65003 65004 65006     10.1.13.3 from 10.1.13.3 (10.1.34.3)       Origin IGP, metric 300, localpref 100, valid, external, best   65002 65006     10.1.12.2 from 10.1.12.2 (2.2.2.2)       Origin IGP, metric 1000, localpref 100, valid, external   65005 65006     10.1.15.5 from 10.1.15.5 (10.1.56.5)       Origin IGP, metric 500, localpref 100, valid, external

 

And we can see that R3 is the best path to prefix 192.168.61.0/24.

 

Now let’s remove the metric from R5 and see what’s happen:

R1#sh ip bgp 192.168.61.0/24 BGP routing table entry for 192.168.61.0/24, version 2 Paths: (3 available, best #1, table Default-IP-Routing-Table) Flag: 0x820   Advertised to update-groups:      1           65005 65006     10.1.15.5 from 10.1.15.5 (10.1.56.5)       Origin IGP, localpref 100, valid, external, best   65003 65004 65006     10.1.13.3 from 10.1.13.3 (10.1.34.3)       Origin IGP, metric 300, localpref 100, valid, external   65002 65006     10.1.12.2 from 10.1.12.2 (2.2.2.2)       Origin IGP, metric 1000, localpref 100, valid, external

 

We can see that R5 is the best path as paths received with no MED are assigned a MED of 0 which is lower.

Issuing the following command:

 

R1# conf t Enter configuration commands, one per line.  End with CNTL/Z. R1(config)#router bgp 65001 R1(config-router)# bgp bestpath med missing-as-worst

 

This command will cause R1 to treat paths with no MED as worst and give them the worst MED value available:

R1#sh ip bgp 192.168.61.0/24 BGP routing table entry for 192.168.61.0/24, version 2 Paths: (3 available, best #1, table Default-IP-Routing-Table) Flag: 0x820   Advertised to update-groups:      1           65003 65004 65006     10.1.13.3 from 10.1.13.3 (10.1.34.3)       Origin IGP, metric 300, localpref 100, valid, external, best   65002 65006     10.1.12.2 from 10.1.12.2 (2.2.2.2)       Origin IGP, metric 1000, localpref 100, valid, external   65005 65006     10.1.15.5 from 10.1.15.5 (10.1.56.5)       Origin IGP, metric 4294967295, localpref 100, valid, external