Brocade FastIron Ethernet Switch Layer 3 Routing Configuration Guide - Manual (Page 417)

Changing the default MED (Metric) used for route redistribution
The Brocade device can redistribute directly connected routes, static IP routes, RIP routes, and OSPF
routes into BGP4. The MED (metric) is a global parameter that specifies the cost that will be applied to
all routes by default when they are redistributed into BGP4. When routes are selected, lower metric
values are preferred over higher metric values. The default BGP4 MED value is 0 and can be assigned
a value from 0 through 4294967295.
NOTE
RIP and OSPF also have default metric parameters. The parameters are set independently for each
protocol and have different ranges.
To change the default metric to 40, enter the following command.
device(config-bgp-router)#default-metric 40
Syntax: default-metric num
The num indicates the metric and can be a value from 0 through 4294967295.
Enabling next-hop recursion
For each BGP4 route learned, the device performs a route lookup to obtain the IP address of the next-
hop for the route. A BGP4 route is eligible for addition in the IP route table only if the following
conditions are true:
• The lookup succeeds in obtaining a valid next-hop IP address for the route.
• The path to the next-hop IP address is an IGP path or a static route path.
By default, the software performs only one lookup for the next-hop IP address for the BGP4 route. If the
next-hop lookup does not result in a valid next-hop IP address, or the path to the next-hop IP address is
a BGP4 path, the software considers the BGP4 route destination to be unreachable. The route is not
eligible to be added to the IP route table.
The BGP4 route table can contain a route with a next-hop IP address that is not reachable through an
IGP route, even though the device can reach a hop farther away through an IGP route. This can occur
when the IGPs do not learn a complete set of IGP routes, so the device learns about an internal route
through IBGP instead of through an IGP. In this case, the IP route table will not contain a route that can
be used to reach the BGP4 route destination.
To enable the device to find the IGP route to the next-hop gateway for a BGP4 route, enable recursive
next-hop lookups. With this feature enabled, if the first lookup for a BGP4 route results in an IBGP path
that originated within the same AS, rather than an IGP path or static route path, the device performs a
lookup on the next-hop IP address for the next-hop gateway. If this second lookup results in an IGP
path, the software considers the BGP4 route to be valid and adds it to the IP route table. Otherwise, the
device performs another lookup on the next-hop IP address of the next-hop for the next-hop gateway,
and so on, until one of the lookups results in an IGP route.
NOTE
You must configure a static route or use an IGP to learn the route to the EBGP multihop peer.
Changing the default MED (Metric) used for route redistribution
FastIron Ethernet Switch Layer 3 Routing Configuration Guide 417
53-1003087-04