On Tue, Dec 27, 2005 at 12:27:52PM +0100, Sylvain Coutant wrote: > Hi, > > > I think this "depend on" is a nice feature - but I would not > > use for "100% fail save" connections. > > Why not ? It has been coded for this purpose ... >
No. It is a hack to make it possible to use two routers at an IX where you only get one IP. If your master router goes belly up all sessions are dropped and opened on the backup box. This results in a route flap. > > > You must take into account, > > that the session will go down if you trigger a failover. > > Of course, this is the basic of a failover between two routers. > By using two IPs you can do failover without losing all routes on failover. This is in my opinion the prefered way of doing it. > > This > > might be acceptable for some kind of sessions (peerings, backup links) > > but may be undesirable for "main" (transit) links. > > This is *highly* desirable in any situation where one router goes down for > any reason. > If you are using bgpd you can add redundancy to your network in a more flexible way than via carp. Terminate your upstreams on multiple boxes run an IBGP mesh and you get failover too. > Look, we're in a *failover* case. Session going down for 10 seconds is > better than session going down until someone brings it back up ... > The session probably down for more than 10 seconds. Even two bgpd take a bit more than 10 seconds to sync the RIB. -- :wq Claudio
