Transition cluster manager alerts to Prometheus alerts
Transition from internal cluster manager alerts to external monitoring alerts using Prometheus.
| Redis Enterprise Software |
|---|
As Redis Enterprise Software transitions from the deprecated monitoring system to the new metrics stream engine, some internal cluster manager alerts were deprecated in favor of external monitoring solutions.
You can use the following table to transition from the deprecated alerts and set up equivalent alerts in Prometheus with PromQL (Prometheus Query Language):
| Cluster manager alert | Equivalent PromQL | Description |
|---|---|---|
| BdbSizeAlert | sum by(db, cluster) (redis_server_used_memory) / sum by(db, cluster) (redis_server_maxmemory) > 0.8 |
Redis server memory usage exceeds 80% |
| NodeMemoryAlert | (node_memory_MemTotal_bytes - node_memory_MemAvailable_bytes) / node_memory_MemTotal_bytes > 0.7 |
Node memory usage exceeds 70% |
| NodeFreeFlashAlert | (node_available_flash_bytes - node_bigstore_free_bytes) / node_available_flash_bytes > 0.7 |
Node flash storage usage exceeds 70% |
| NodeEphemeralStorageAlert | (node_ephemeral_storage_avail_bytes - node_ephemeral_storage_free_bytes) / node_ephemeral_storage_avail_bytes > 0.7 |
Node ephemeral storage usage exceeds 70% |
| NodePersistentStorageAlert | (node_persistent_storage_avail_bytes - node_persistent_storage_free_bytes) / node_persistent_storage_avail_bytes > 0.7 |
Node persistent storage usage exceeds 70% |