Files
traefikoidc/vendor/github.com/redis/go-redis/v9/maintnotifications
lukaszraczyloandGitHub a750c4f5b9 Size computation for allocation may overflow (#99)
* Size computation for allocation may overflow

Performing calculations involving the size of potentially large strings or slices can result in an overflow (for signed integer types) or a wraparound (for unsigned types). An overflow causes the result of the calculation to become negative, while a wraparound results in a small (positive) number.
2025-12-08 11:22:28 +00:00
..

Maintenance Notifications

Seamless Redis connection handoffs during cluster maintenance operations without dropping connections.

⚠️ Important Note

Maintenance notifications are currently supported only in standalone Redis clients. Cluster clients (ClusterClient, FailoverClient, etc.) do not yet support this functionality.

Quick Start

client := redis.NewClient(&redis.Options{
    Addr:     "localhost:6379",
    Protocol: 3, // RESP3 required
	MaintNotificationsConfig: &maintnotifications.Config{
        Mode: maintnotifications.ModeEnabled,
    },
})

Modes

  • ModeDisabled - Maintenance notifications disabled
  • ModeEnabled - Forcefully enabled (fails if server doesn't support)
  • ModeAuto - Auto-detect server support (default)

Configuration

&maintnotifications.Config{
    Mode:                       maintnotifications.ModeAuto,
    EndpointType:               maintnotifications.EndpointTypeAuto,
    RelaxedTimeout:             10 * time.Second,
    HandoffTimeout:             15 * time.Second,
    MaxHandoffRetries:          3,
    MaxWorkers:                 0,    // Auto-calculated
    HandoffQueueSize:           0,    // Auto-calculated
    PostHandoffRelaxedDuration: 0,    // 2 * RelaxedTimeout
}

Endpoint Types

  • EndpointTypeAuto - Auto-detect based on connection (default)
  • EndpointTypeInternalIP - Internal IP address
  • EndpointTypeInternalFQDN - Internal FQDN
  • EndpointTypeExternalIP - External IP address
  • EndpointTypeExternalFQDN - External FQDN
  • EndpointTypeNone - No endpoint (reconnect with current config)

Auto-Scaling

Workers: min(PoolSize/2, max(10, PoolSize/3)) when auto-calculated Queue: max(20×Workers, PoolSize) capped by MaxActiveConns+1 or 5×PoolSize

Examples:

  • Pool 100: 33 workers, 660 queue (capped at 500)
  • Pool 100 + MaxActiveConns 150: 33 workers, 151 queue

How It Works

  1. Redis sends push notifications about cluster maintenance operations
  2. Client creates new connections to updated endpoints
  3. Active operations transfer to new connections
  4. Old connections close gracefully

For more information, see FEATURES