High availability#
After an initial Replicator instance is deployed, it can be paired with a secondary headless deployment for high availability.
Replicator supports two types of high availability configurations as described below.
Note
Only unpaired headless deployments with no profiles saved can be selected as the secondary instance in a high availability pair. All profiles saved to the primary instance will automatically be synced to the secondary at regular intervals.
Secondary instances will still be listed in the Admin > Resources > Replicators, but only their name can be changed in the configuration tray.
All configuration (profiles, collectors, etc.) for an HA pair is done through the primary instance. The secondary instance cannot be associated with new profiles or collectors.
Single-network configuration#
Single-network high availability uses a virtual IP address as a packet destination and can be enabled when both the primary and the secondary Replicator instances are deployed on the same subnet.
If the primary Replicator instance becomes unavailable, streams are automatically picked up by the secondary instance, which will continue packet replication/forwarding to the same collectors. When the primary instance becomes available again, it will immediately resume its previous functions.
Setup
Single-network high availability for a Replicator instance is enabled as follows:
Register and deploy the headless Replicator appliance that will be used as the secondary instance.
In the Admin > Resources > Replicators view, click on the primary instance to open the configuration tray.
Toggle on High Availability and select the secondary instance from the Secondary IP dropdown.
Toggle on the Virtual IP option, and the enter the following details in the provided fields:
Virtual IP address exporters will be configured to send packets to
Virtual router ID for the high availability pair (must be unique to the pair)
Enter a failover delay (in minutes) before the secondary instance takes over replication functions.
Click the Save button the save the configuration changes.
After a single-network high availability pair has been created, configure all exporters currently sending packets to the primary instance to send the packets to the virtual IP address instead.
Multi-network configuration#
In multi-network high availability (paired instances on different subnets), exporters must be configured to send packets to both the primary and secondary instances.
In this configuration, the secondary instance continuously monitors the primary and will take over replication if it detects that the primary is offline. Once the primary becomes available again, any configuration changes are synced to it, and the secondary instance will revert to its previous passive monitoring mode.
Setup
Register and deploy the headless Replicator appliance that will be used as the secondary instance.
In the Admin > Resources > Replicators view, click on the primary instance to open the configuration tray.
Toggle on High Availability and select the secondary instance from the Secondary IP dropdown.
Enter a failover delay (in minutes) before the secondary instance takes over replication functions.
Click the Save button the save the configuration changes.
After a multi-network high availability pair has been created, configure all exporters currently sending packets to the primary instance to send the same packets to the secondary instance.
Note
If the primary instance goes offline, the secondary instance will first verify that collectors can be reached before starting replication.
Reverting HA pairings#
To revert paired Replicator instances back to single appliances, toggle off High Availability for the primary instance in the Admin > Resources > Replicators configuration tray.
This will unpair the instances and allow them to be used as single deployments again. The primary instance will retain all profiles, collectors, and other settings previously applied, and the secondary instance will be reverted to its unused, post-deployment state.