On Tue, Dec 06, 2011 at 08:34:44AM +0100, Hans J. Albertsson wrote: > Actually, automatic failover will not be necessary, it is after all > only a mail server in an environment with fairly resilient users. > They'd be quite happy to have to call a sysadm up for manual failover.
How you design this depends on what components you expect to fail. I have created an SMF service that does manual failover on a two-node cluster. It expects a zpool from an external storage device that can be imported by either node. In this case its a SAS array that's connected to both nodes. The SMF service imports or exports the zpool and creates or deletes a logical IP interface. It also adds dependancies to other services that require these resources. It is in production on Solaris 10 for core services like DNS and NIS. For an e-mail server only, you should consider Cyrus IMAP. It has built-in facilities to synchronize the message store and databases between back-end servers at the application level. In this case, each node has its own separate storage devices. -- -Gary Mills- -refurb- -Winnipeg, Manitoba, Canada- _______________________________________________ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss