Slow convergence between F5 Loadbalancer and Juniper router

Can you be more explicit? What's your topology? I assume eBGP between viprion and router? Is there a redundant pairs of routers running iBGP between them? Are the Viprions active/active or active/standby?

Also can you be more explicit re failure conditions - Link down?

Please confirm all.

We are using eBGP between Viprion and Router. Routers are in redundant pairs but we are not running iBGP between routers. Viprions are in active standby mode.

When the link between router and viprion is down failover is happening between viprions but router is receiving routes from viprion after 10-15 seconds.

I would run iBGP between your routers so that when failover occurs the 2nd router already has a routing entry with the necessary next-hop (the viprion floating self IP).

Hi Suresh,

I am assuming you have a full-mesh between the Juniper routers and the Active/Standby Viprions. I have the same setup (no floating self-IPs). Are the routers able to detect which is the active unit and which one is the standby. When we used eBGP, the Junipers would still point to standby unit as the next-hop (when a failover occurred) to reach an internal host. Is there a way in BGP that lets the routers detect the active unit and change their next-hop accordingly in the event of a failover? Any help appreciated. Thanks!