H3C Technologies H3C S12500 Series Switches - Manual (Page 457)

441
Configuring IPv6 MBGP soft reset
After you modifying a route selection policy, you must reset IPv6 MBGP connections to make the new one
take effect.
The current IPv6 MBGP implementation supports the route-refresh feature that enables dynamic route
refresh without terminating IPv6 MBGP connections.
If a peer that does not support route-refresh exists in the network, you must configure the peer
keep-all-routes command to save all routes from the peer. When the routing policy is changed, the
system will update the IPv6 MBGP routing table and apply the new policy.
Performing soft reset through route refresh
If the peer is enabled with route refresh, when the IPv6 MBGP route selection policy is modified on a
router, the router advertises a route-refresh message to its IPv6 MBGP peers, which resend their routing
information to the router after they receive the message. Therefore, the local router can perform dynamic
route update and apply the new policy without terminating IPv6 MBGP connections.
To configure IPv6 MBGP soft reset through route refresh:
Step Command Remarks
1. Enter system view. system-view N/A
2. Enter BGP view. bgp as-number N/A
3. Enter IPv6 address family
view.
ipv6-family N/A
4. Enable IPv6 BGP route refresh
for a peer or a peer group.
peer { ipv6-group-name |
ipv6-address } capability-advertise
route-refresh
Optional.
Enabled by default.
Performing soft reset manually
If the peer does not support route refresh, you can use the peer keep-all-routes command to save all the
route updates from the peer, and then use the refresh bgp ipv6 multicast command to soft-reset IPv6
MBGP connections to refresh the IPv6 MBGP routing table and apply the new policy without terminating
IPv6 MBGP connections.
To perform soft reset manually:
Step Command Remarks
1. Enter system view. system-view N/A
2. Enter BGP view. bgp as-number N/A
3. Enter IPv6 address family
view.
ipv6-family N/A
4. Enter IPv6 MBGP address
family view.
ipv6-family multicast N/A
5. Keep all routes from a peer or
a peer group regardless of
whether they pass the
inbound filtering policy.
peer { ipv6-group-name |
ipv6-address } keep-all-routes
Not kept by default.