Routing

Hi, Here is a question. Anyone can help to answer?

Here is a topology R1(ge-1/0) ------ (ge-1/1)R2(ge-1/2) ----- (ge-1/3)R3

R3 advertise bgp routes to R2. The interface ge-1/2 in R2 is routing-instance ABC, while ge-1/1 in R2 is routing-instance xyz. Do you think if it is possible for R2 to advertise the bgp routes to R1 through R2' ge-1/1?  Thank you

Hi,

This can be possible by sharing the routes between ABC and xyz RI. rib-groups can be used for this.

To advertise a route to a BGP neighbor, that route needs to be present in that routing-table.

However, in doing so I don't really see the purpose of having 2 RIs, unless I missed something.

Cheers,

Ashvin

The current configuration in R2' ge-1/1 is routing-instance xyz. If we change it into routing-instance ABC, that would be easy for us to advertise that routes to R1. but we do not like to change so because it already has some traffic going there. 

So how can can we share the routes between ABC and xyz RI? Can you send a link here? Thanks

There are two options to leak routes between VRF's:

1) auto export
2) Rib groups

Easiest one is auto export

Here is the link for the same:
http://www.juniper.net/techpubs/en_US/junos12.3/information-products/topic-collections/nce/auto-export-understanding/auto-export-understanding.pdf

Basically you have to add auto-export knob in both vrf under vrf x routing-options and also have a common route target
between both VPN's in addition to their own/native unique route targets.

You can export/import route targets via policy something like

VPNA {
instance-type vrf;
interface ge-4/0/1.35;
interface ge-2/1/1.70;
route-distinguisher 7777:1111;
vrf-import import-vpn-a
vrf-export export-vpn-a
routing-options {
auto-export
}

VPNB {
instance-type vrf;
interface ge-1/1/1.40;
interface ge-3/2/2.50;
route-distinguisher 7777:2222;
vrf-import import-vpn-b
vrf-import export-vpn-b
routing-options {
auto-export

}

VPN A route-target policy :

vrf import:

policy-options{
policy-statement import-vpn-a {
term 1 {
from {
protocol bgp ;
community vpn-a;
}
then {
accept;
}
}

}
community vpn-a member target:7777:1111;>>>>vpn-a native rt
community vpn-a member target:4:4;>>common rt for both vpn
 
 vrf-export:
 

 
 policy-options{
policy-statement export-vpn-a {
term 1 {
from {
protocol [bgp direct];
}
then 
{
community add vpn-a;

}
}

}

VPN B route-target policy :

vrf import:

policy-options{
policy-statement import-vpn-b {
term 1 {
from {
protocol bgp ;
community vpn-b;
}
then {
accept;
}
}

}
community vpn-a member target:8888:2222;>>>>vpn-B native rt
community vpn-a member target:4:4;>>common rt for both vpn
 
vrf-export:
 

 
 policy-options{
policy-statement export-vpn-b {
term 1 {
from {
protocol [bgp direct];
}
then 
{
community add vpn-b;

}
}

}

Hi, 

Both auto-export and rib-groups are possible if the instance-type is vrf.

If instance-type is virtual-router, I belive rib-group would be the only option.

Can you advise what is the instance-type of the RIs?

Cheers,

Ashvin

Another option is to use logical tunnel interfaces to connect the two virtual routers internally and setup peering or static routes between the two virtual routers.

MX

https://www.juniper.net/documentation/en_US/junos15.1/topics/usage-guidelines/services-configuring-logical-tunnel-interfaces.html

SRX

https://kb.juniper.net/InfoCenter/index?page=content&id=KB21260

We already changed topology. So the issue has been solved. Thank you all