- DigiCert product docs
- DigiCert DNS
- Route DNS traffic
- Simple failover
Simple failover
Simple failover helps maintain service availability by selecting a destination IP address according to its priority and health. Configure multiple destinations for a record so that an available destination can be selected when another becomes unhealthy.
What simple failover is
Simple failover is a record policy that uses an ordered list of destination IP addresses and health checks to determine which destination to return in DNS responses.
The highest-priority enabled, healthy destination serves the record. Other enabled destinations provide alternatives when higher-priority destinations are unavailable.
Health checks determine destination availability using the monitoring configuration associated with the record.
Explore simple failover on the Records page
To view records configured with simple failover, go to DNS > Zones, select a zone from Primary zones, and open the Records tab.
The page displays the following information:
Information | Description |
|---|---|
Record name | The fully qualified name of the record |
IP addresses | The configured destinations. The selected destination is marked Serving |
Policy | A Failover indicator showing the overall health condition. Hover over the indicator for more information |
TTL | How long the DNS response can be cached, in seconds |
The Failover indicator appears green when destinations are healthy, amber when one or more destinations are unhealthy, and red when all configured destinations are unhealthy.
Open the record to review individual destination statuses, change priorities, or update the monitoring configuration.
Configure simple failover
The following steps describe configuring simple failover for an A record. For information about the record’s name and IPv4 addresses, see A records.
Go to DNS > Zones and select a zone from Primary zones.
On the Records tab, select Add record.
Select A (Address - IPv4) as the record type.
Enter the record Name, or use
@for the zone apex, and enter the TTL.Under Policy, select Simple failover.
Enter the destination IPv4 addresses. Select Add another to add destinations within your plan’s limit.
Arrange the destinations in priority order using the drag handles, with the preferred destination first.
Enable at least two destinations to activate simple failover.
Complete the Monitoring configuration.
Select Save and close, or Save and add another to create another record.
For an existing record, open its edit form, select Simple failover, configure the destinations and monitoring settings, and select Update.
Choose a monitoring protocol appropriate for the service:
Protocol | Available settings |
|---|---|
HTTP or HTTPS | Region, Port, Failure sensitivity, URL, Resource path (optional), Search string (optional) |
TCP | Region, Port, Failure sensitivity, Timeout (optional) |
ICMP (PING) | Region, Failure sensitivity |
The fields displayed depend on the selected protocol.
Destination priority and selection
Destination priority determines the order in which enabled, healthy destinations are selected. Priority 1 is the highest priority.
When the highest-priority destination is healthy and enabled, it serves the record. If it becomes unhealthy, simple failover selects the next enabled, healthy destination in the list.
For example, consider three enabled destinations:
Priority | Destination | Health | Selection |
|---|---|---|---|
1 |
| Down | Unavailable |
2 |
| Up | Serving |
3 |
| Up | Available as the next alternative |
To change the order, drag a destination to its new position and save the record.
Destination statuses and states
The destination table separates health status, serving state, and whether a destination is enabled.
Field | Value | Meaning |
|---|---|---|
Status | Up | The destination is healthy according to its configured health checks |
Down | The destination is unhealthy according to its configured health checks | |
State | Serving | The destination is currently selected for the record |
Enabled | On | The destination participates in failover selection |
Off | The destination is excluded from failover selection but remains in the configuration |
A destination can be healthy without being selected. For example, a lower-priority destination can show Up while a higher-priority destination is Serving.
To temporarily exclude a destination, turn off its Enabled toggle and save the record. To remove it from the configuration, select its delete icon and save.
When destinations are unhealthy
When some destinations are unhealthy, the record displays a warning. Simple failover continues selecting an enabled, healthy destination according to the configured priority order.
A failure does not necessarily change the serving destination. If only a lower-priority destination becomes unhealthy, the higher-priority healthy destination can continue serving.
When all configured destinations are unhealthy, the record displays an error stating that all destination IP addresses are down. Review the destination services and monitoring configuration to investigate the failures.
Destination limits and inactive failover
The number of failover destinations you can configure depends on your plan. When the limit is reached, Add another becomes unavailable. Its tooltip explains that your plan’s failover target limit has been reached.
Simple failover requires at least two enabled destination IP addresses. If fewer than two are enabled, the form displays Enable at least two IP addresses to activate simple failover.
Keeping several addresses in the configuration does not activate failover if only one is enabled. Enable another destination and save the record to restore the required configuration.
Simple failover and TTL
The record’s TTL determines how long a DNS response can remain cached. When simple failover selects a different destination, clients using a cached response may continue using the previous address until that cached response expires.
Failover timing therefore includes both the time needed to detect an unhealthy destination and the remaining cache lifetime. A shorter TTL can reduce how long an earlier response remains cached, but it does not make failure detection immediate.
Changing the TTL does not shorten the lifetime of responses already cached under the previous TTL.