Two-Node Cluster Mechanism

Service Recovery Mechanism

Impact on Services

Active/Standby redundancy

ϒ⁄Two unified gateways are deployed at the headquarters as the active and standby nodes. The active and standby nodes use a heartbeat mechanism to constantly check each other's status.

NOTE:

The active and standby nodes must use the unified gateways of the same model. Their board layouts and software versions must be the same.

ϒ⁄When the active and standby nodes are running correctly, all users register with the active node. The active node processes all user requests and synchronizes user registration and configuration data to the standby node in real time.

ϒ⁄When the active node fails, the standby node takes over all services from the active node and processes all user requests.

ϒ⁄If the trunk gateway is not deployed, the active and standby nodes both connect to the PSTN. When the active and standby nodes are running correctly, the active node routes calls to the PSTN through trunks. When the active node fails, configurations are required on the PSTN to set up trunk channels with the standby node, and the standby node then routes calls to the PSTN through trunks.

ϒ⁄All service servers support active/standby switching of the gateways. This means these service servers automatically connect to the standby node for service processing after a switching from the active node to the standby node.

After switching from the active node to the standby node due to an active node failure, the standby node processes all services. When detecting that the active node recovers, the standby node automatically switches services to the active node.

ϒ⁄The switching lasts about 2 minutes. During the switching, new calls cannot be initiated. Considering that terminal registration updates to the standby node, the maximum service interruption time is 2 minutes plus the duration of terminal registration updates. The calls that have been set up are not affected.

NOTICE:

The standby node is allowed to take over services for a maximum of 30 days. To avoid service unavailability, repair or replace the faulty node promptly.

ϒ⁄When the standby node takes over the services, data is not synchronized to the active node. After services are switched back to the active node, the active node synchronizes data to the standby node, and the data on the standby node will be overwritten. Before manual switchover, export data from the standby node and import the exported data to the active node so that the data that is generated on the standby node during the fault period will not be lost.

ϒ⁄On the active node of the unified gateway, built-in VMS voice file, recording file, custom voice file, CDR file, and unified gateway voice and language settings do not support synchronization to the standby node. The built-in VMS voice settings can be synchronized to the standby node.