redis.uptime

Measures the elapsed time in seconds since the Redis server process started. This gauge resets to zero on every restart, making it essential for change detection and incident correlation. Unexpected resets indicate crashes, manual restarts, or deployment activity. Long uptimes combined with high memory usage may signal memory fragmentation accumulation that could benefit from a planned restart. Used in availability calculations and as a baseline timestamp for understanding when metrics like cumulative counters began accumulating.