The only IP failover I trust is carp (http://www.openbsd.org/faq/pf/carp.html) on OpenBSD/FreeBSD. Once set up properly with syncing, you lose no state on a failover (all connection and NAT state is gossiped between cluster nodes sharing an IP address).
The only downside is most services aren't well tested under OpenBSD/FreeBSD these days so you may end up hitting a few edge cases in software designed and tested only under Linux.
For most apps and servers, migrating the IP state is not enough. The app server's connection state cannot be easily migrated. E.g. The MySQL connection of a failed MySQL server will be gone. Migrating the IP connection to a new MySQL server won't do anything.
Carp is good for firewall server redundancy because IP state is all that firewall is maintaining.
Comments
The only IP failover I trust is carp (http://www.openbsd.org/faq/pf/carp.html) on OpenBSD/FreeBSD. Once set up properly with syncing, you lose no state on a failover (all connection and NAT state is gossiped between cluster nodes sharing an IP address).
The only downside is most services aren't well tested under OpenBSD/FreeBSD these days so you may end up hitting a few edge cases in software designed and tested only under Linux.
For most apps and servers, migrating the IP state is not enough. The app server's connection state cannot be easily migrated. E.g. The MySQL connection of a failed MySQL server will be gone. Migrating the IP connection to a new MySQL server won't do anything.
Carp is good for firewall server redundancy because IP state is all that firewall is maintaining.