04-15-2009 02:51 AM
What is the use of
"no bgp default ipv4-unicast" command ?
As per my info by default router exchange IPv4 prefixes.To change this default behavour we have to option to exchange vpnv4 prefies=>
1.Use of "no bgp default ipv4-unicast" command.
2.configure BGP neighbour under BGP and activate the same for in VPNv4 address fanily.
Is this correct ?
04-15-2009 04:44 AM
Hello Ganpat,
the meaning of the command is to prepare the router to be multiprotocol and not only to advertise ipv4 unicast address family that is the default.
So no bgp default ipv4-unicast
means = be multi address-families oriented and check the neighbor activate command in each AF context to see on what address families each neighbor has to be contacted/consulted.
During BGP session setup the capabilities exchange tells on what "channels/arguments" the local node would like to talk with the peer.
For this reason when giving the command all existing BGP sessions are reset to perform capabilities negotiation.
May be in newer IOS images you can reach the AF configuration without using the commanf but in the past it was mandatory to achieve multi-protocol BGP operation.
see
http://www.cisco.com/en/US/docs/ios/iproute/command/reference/irp_bgp1.html#wp1012800
Hope to help
Giuseppe
04-15-2009 05:56 AM
thnaks......Giuseppe
04-15-2009 06:57 AM
Hi
there is another way for a router configured with NLRI:
bgp upgrade-cli
Converts to AF style
04-30-2011 02:33 AM
Hi Giuseppe
I dont entirely agree , the command just disables ipv4 unicast address family peering capability , the router is multiprotocol whether you enable or disable this command . you may have a scenario where you dont want to receive full internet or ipv4 bgp routes e.g MPLS PE node
Wr
05-04-2011 02:29 AM
Hello Petersrule2003,
the behaviour may have changed over time in newer IOS images
>> the command just disables ipv4 unicast address family peering capability
However,
the command reference
http://www.cisco.com/en/US/docs/ios/iproute_bgp/command/reference/irg_bgp1.html#wp1113664
still provides
IPv4 address family routing information is advertised by default for each BGP routing session configured with the neighbor remote-as command, unless you first configure the no bgp default ipv4-unicast command before configuring the neighbor remote-as command.
we need to go a step further sometimes then command reference
hope to help
Giuseppe
01-20-2014 05:08 AM
Hi Ganpat,
I understand this is a really old post, but I see that the question is still now answered. This post also happens to be on the top of google searces so I thought I can probably add to the existing discussion.
I found an explanation for this command, have a look.
http://blog.networkrise.com/2013/02/06/bgp-default-ipv4-unicast/
06-22-2022 10:09 AM
Link is dead.
10-30-2016 06:02 AM
By default, the IPv4 address family is advertised to every peer of every address family.
Lets say you are running IPv4 and IPv6. Lets say you have native IPv6 transit. Do you really want to advertise IPv4 prefixes over a native IPv6 link? No.
So by saying "no bgp default ipv4-unicast" you are disabling this behaviour. After doing this command the router will only advertise IPv4 prefixes to IPv4 neighbours, and will not attempt to advertise IPv4 prefixes to IPv6 neighbours.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide