[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: bgpd and two CARPed routers



* Hyb <maillist_(_dot_)_localhost_(_at_)_ntlworld_(_dot_)_com> [2005-08-16 21:12]:
> You've mentioned previously on the list about nexthop mangling for iBGP (ie.
> the preference not to). Is it possible to run the aforemtioned setup without
> manually setting the nexthop to the peer's other /30 address?

well, nexthops are not modified on iBGP, so both routers have to be 
able to reach the nexthop. you might need static routes or some 
internal routing protocol to achieve that. not sure why you talk about 
a /30, in this setup you usually use a /29 to each uplink.

> And on a semi
> related note, what happens when inbound traffic comes from our upstream and
> into core2, which has BACKUP state on the internal facing CARP group -
> presumably it needs to bounce off the exchangepoint network and through
> core1?

it needs some way to reach your internal network, be it through core1 
or the address the ohysical interface for that carp interface.

-- 
BS Web Services, http://www.bsws.de/
OpenBSD-based Webhosting, Mail Services, Managed Servers, ...
Unix is very simple, but it takes a genius to understand the simplicity.
(Dennis Ritchie)