Step 1 Capture packets on the central SoftCo and gateway. It is found that all calls are released by the SoftCo by sending the BYE message.
Step 2 Capture call logs on the central SoftCo. No useful information is obtained because the call traffic is high onsite.
Step 3 Check the heartbeat interval between the central SoftCo and gateway. It is found that the heartbeat interval from the central SoftCo to the gateway is set to 40s. After the heartbeat interval is changed to 180s, the fault is rectified.
The office has more than 200 local nodes. All the local nodes are connected to the central node through SIP trunks. Recently, dozens of local nodes are added for capacity expansion. As a result, the total number of SIP trunks exceeds 220. The heartbeat interval is set to 40s so that services on a local node can be quickly switched to the central node if the local node cannot receive heartbeat messages.
The root causes of the fault are as follows:
The SoftCo polls all the trunks that have been configured and internal components and modules to which heartbeat messages must be sent, and sends heartbeat messages. In 1s, the SoftCo can send heartbeat messages to 32 modules or trunks. As the number of trunks increases, the SoftCo takes more than 40s to poll all the trunks, internal components, and internal modules. As a result, the fault occurs.
The central SoftCo and local SoftCos send heartbeat messages independently. The number of trunks is large on the central SoftCo, and the heartbeat interval must be changed to 180s. On local SoftCos, the number of trunks is small. Therefore, the heartbeat interval can be retained as 40s so that services on a local node can be quickly switched to the central node if the local node cannot receive heartbeat messages.