failover.xml revision 7de51a3df987db4a42b3a74a07ef2b1d276ca72e
342N/A<refsect1 id='failover'>
470N/A <title>FAILOVER</title>
342N/A <para>
342N/A The failover feature allows back ends to automatically switch to
342N/A a different server if the primary server fails.
342N/A </para>
342N/A <refsect2 id='failover_syntax'>
342N/A <title>Failover Syntax</title>
342N/A <para>
342N/A The list of servers is given as a comma-separated list; any
342N/A number of spaces is allowed around the comma. The servers are
342N/A listed in order of preference. The list can contain any number
342N/A of servers.
342N/A </para>
342N/A </refsect2>
342N/A <refsect2 id='failover_mechanism'>
342N/A <title>The Failover Mechanism</title>
342N/A <para>
342N/A The failover mechanism distinguishes between a machine and a
342N/A service. The back end first tries to resolve the hostname of a
342N/A given machine; if this resolution attempt fails, the machine is
342N/A considered offline. No further attempts are made to connect
342N/A to this machine for any other service. If the resolution
342N/A attempt succeeds, the back end tries to connect to a service
342N/A on this machine. If the service connection attempt fails,
342N/A then only this particular service is considered offline and
342N/A the back end automatically switches over to the next service.
342N/A The machine is still considered online and might still be tried
342N/A for another service.
342N/A </para>
342N/A <para>
342N/A Further connection attempts are made to machines or services
342N/A marked as offline after a specified period of time; this is
342N/A currently hard coded to 30 seconds.
342N/A </para>
342N/A <para>
342N/A If there are no more machines to try, the back end as a whole
342N/A switches to offline mode, and then attempts to reconnect
342N/A every 30 seconds.
342N/A </para>
342N/A </refsect2>
342N/A</refsect1>
342N/A