Introduction

Monitors Redis replication status and lag for replica instances. Fails the experiment if replication lag exceeds the configured threshold or if the master link goes down (when Require Link Up is enabled).

This check is designed to be combined with attacks that affect replication, such as Stop Sentinel or network-level fault injections. For latency monitoring on the Redis command level, use Latency Check instead.

Use Cases

• Monitor replication health during a Stop Sentinel attack to verify that replicas continue to sync with the master.
• Verify that replicas catch up within acceptable lag after a network partition is healed.
• Fail the experiment if the master link goes down during a fault injection, ensuring replication integrity.

Parameters

Parameter	Description	Default
Duration	How long the check runs. Replication status is polled continuously for this duration.	60s
Max Lag Seconds	Maximum allowed replication lag in seconds. If the replica falls further behind than this threshold, the experiment fails.	10
Require Link Up	Whether to fail the experiment if the master link is reported as down. When enabled, a broken master-replica link causes immediate failure.	true