Quickstart
Integration Features
Dashboards
Alerts
Documentation
dashboards
Redis (OpenTelemetry) quickstart contains 1 dashboard. These interactive visualizations let you easily explore your data, understand context, and resolve problems faster.
Show MoreShow Less
Redis OpenTelemetry Dashboard
See all
alerts
Redis (OpenTelemetry) observability quickstart contains 8 alerts. These alerts detect changes in key performance metrics. Integrate these alerts with your favorite tools (like Slack, PagerDuty, etc.) and New Relic will let you know when something needs your attention.
Show MoreShow Less
1. Memory utilization is too high
This alert is triggered when the memory utilization exceeds 90%. Please ensure the `redis.maxmemory` metric from the Redis receiver is enabled for this alert to function correctly.
2. High key eviction rate
This alert is triggered when the key eviction rate exceeds 1000 keys per second, indicating significant memory pressure. High eviction rates typically correlate with declining cache hit ratios and increased application latency.
3. Redis Cluster slots degraded
This alert is triggered when the number of healthy slots drops below 16384, indicating partial data unavailability. Some hash slots are in PFAIL or FAIL state — keys mapped to those slots cannot be accessed.
4. Redis Cluster node departed
This alert is triggered when the number of known nodes in the cluster drops, indicating a node has departed or become unreachable. This reduces cluster capacity and fault tolerance — if the departed node was a master with no replica, data in its slots becomes unavailable.
5. Cache hit ratio is too low
This alert is triggered when the keyspace hit ratio drops below 50%, indicating that more than half of key lookups are failing. This typically means keys are being evicted faster than they can be warmed, or TTL configuration needs review.
6. Redis Cluster state is FAIL
This alert is triggered when the Redis Cluster state drops to FAIL (0), indicating one or more hash slots are unavailable. This means data stored in affected slots cannot be read or written — immediate investigation is required.
7. Rejected connections detected
This alert is triggered when Redis rejects client connections due to the maxclients limit being reached. Any rejected connection indicates clients are being turned away and requires immediate action.
8. Anomaly in connected clients
This alert is triggered when the number of connected clients deviates significantly from the baseline, either up or down. A sudden increase may indicate a thundering herd or connection leak. A sudden decrease may indicate upstream failures or network issues.
documentation
Redis (OpenTelemetry) observability quickstart contains 1 documentation reference. This is how you'll get your data into New Relic.
Show MoreShow Less