Failover resource record sets only: To configure failover, you add the Failover
element to
two resource record sets. For one resource record set, you specify PRIMARY
as the value for
Failover
; for the other resource record set, you specify SECONDARY
. In addition, you
include the HealthCheckId
element and specify the health check that you want Amazon Route 53 to
perform for each resource record set.
Except where noted, the following failover behaviors assume that you have included the HealthCheckId
element in both resource record sets:
-
When the primary resource record set is healthy, Route 53 responds to DNS queries with the applicable value from
the primary resource record set regardless of the health of the secondary resource record set.
-
When the primary resource record set is unhealthy and the secondary resource record set is healthy, Route 53
responds to DNS queries with the applicable value from the secondary resource record set.
-
When the secondary resource record set is unhealthy, Route 53 responds to DNS queries with the applicable value
from the primary resource record set regardless of the health of the primary resource record set.
-
If you omit the HealthCheckId
element for the secondary resource record set, and if the primary
resource record set is unhealthy, Route 53 always responds to DNS queries with the applicable value from the
secondary resource record set. This is true regardless of the health of the associated endpoint.
You can't create non-failover resource record sets that have the same values for the Name
and
Type
elements as failover resource record sets.
For failover alias resource record sets, you must also include the EvaluateTargetHealth
element and
set the value to true.
For more information about configuring failover for Route 53, see the following topics in the Amazon Route 53
Developer Guide: