ReconnectRate

ReconnectRate is a metric used in networked and distributed systems to quantify how often a previously broken connection is re-established. It is typically defined as the number of reconnection events within a specified time window, divided by the duration of that window. In some contexts it is expressed as reconnections per unit time (for example, per second or per minute) or as a fraction of sessions that experience at least one reconnection during the window.

Measurement and scope may vary. ReconnectRate can be computed for individual clients, for groups of clients, or for an entire service. Some implementations distinguish between successful reconnections and failed attempts, and may track reconnection attempts that occur within a backoff period separately from immediate retries. Rolling or sliding time windows are common to reflect recent system behavior, and reconnections may be aggregated across multiple endpoints or services.

Interpretation and usage follow. A high reconnect rate typically indicates network instability, server load issues, or aggressive client retry logic, whereas a very low rate suggests stable connectivity. ReconnectRate is often used alongside related metrics such as ReconnectTime (time to re-establish a connection), Availability, and Session Durations to assess overall reliability and user experience. It can trigger alerts when thresholds are exceeded or inform capacity planning and backoff strategy adjustments, such as the use of jitter or exponential backoff to reduce churn.

Limitations should be noted. ReconnectRate alone does not reveal the duration of downtime, the impact on user experience, or whether reconnections resume normal operation quickly. It should be interpreted in the context of complementary metrics.