Photon Realtime and CCU Billing: Connection Lifecycle Timing

Photon Realtime employs a Concurrent User (CCU) licensing model where costs are tied to peak simultaneous connections. In low-traffic environments, maintaining an accurate CCU count is critical to staying within the limits of the Free or low-tier subscriptions and avoiding connection rejections.

There is a technical distinction between a graceful shutdown and an abnormal client disconnection (such as a crash or network timeout). While graceful exits notify the server immediately, abnormal disconnects rely on server-side timeout mechanisms to reclaim the CCU slot.

Given the peak-based billing structure, the latency between a client's actual loss of connectivity and the server's decrementing of the CCU count can impact capacity availability for new users.

• What is the default timeout duration before an abnormal disconnection is registered as a CCU decrement?
• Can this timeout interval be configured to optimize CCU availability for low-traffic workloads?