All Integrations
DatabasesTigerOps agent + Helm chart

Valkey Integration

Monitor Redis-compatible Valkey metrics, cluster health, and command latency across your key-value store fleet. Get predictive memory alerts and AI hotspot detection before cache issues impact your application.

Setup

How It Works

01

Deploy TigerOps Valkey Exporter

Add the TigerOps Helm chart to your cluster and point it at your Valkey node addresses. The exporter auto-discovers all cluster nodes, replication topology, and keyspace statistics via the INFO command.

02

Configure Remote Write

Point the Valkey exporter to your TigerOps remote-write endpoint. All memory, command rate, replication lag, and cluster slot metrics flow in within minutes. TLS and AUTH password are both supported.

03

Set Memory & Eviction Alerts

Define maxmemory utilization thresholds and eviction rate alerts per node. TigerOps fires predictive alerts when memory growth rate will breach your maxmemory limit before it causes key evictions.

04

Correlate with Application Cache Hits

TigerOps links Valkey hit/miss ratios and command latency spikes with application-level trace data, so you can see exactly which service endpoints are suffering from cache misses or keyspace hotspots.

Capabilities

What You Get Out of the Box

Command Latency Monitoring

Per-command latency histograms using Valkey LATENCY HISTORY and LATENCY LATEST APIs. Track slow commands, identify latency spikes for SET, GET, ZADD, and pipeline operations per node.

Cluster Health & Slot Coverage

Cluster node state, slot assignment coverage, migration status, and failover event detection. Alert when cluster slot coverage drops below 100% or when nodes enter a failed or pfail state.

Memory & Eviction Tracking

Used vs. maxmemory ratio, RSS memory overhead, eviction policy effectiveness, evicted key counts, and expired key rates per node. Predict memory exhaustion before it causes cache thrashing.

Replication Lag Monitoring

Primary-replica replication offset lag, replication buffer size, replica connection state, and partial resync vs. full resync counts to keep your replication topology healthy.

Keyspace & Hit Rate Analysis

Per-database keyspace statistics, global hit rate and miss rate, key expiration rates, and keyspace notification event counts for complete cache efficiency visibility.

AI Hotspot Detection

TigerOps AI detects keyspace hotspots and command rate anomalies — identifying when a single key or command type is consuming disproportionate CPU time and causing latency for all other commands.

Configuration

Helm Values for Valkey Exporter

Deploy the TigerOps Valkey exporter to your Kubernetes cluster with these Helm values.

valkey-exporter-values.yaml
# TigerOps Valkey Exporter Helm values
# helm repo add tigerops https://charts.atatus.net
# helm install valkey-exporter tigerops/valkey-exporter -f values.yaml

valkeyExporter:
  # Single node or cluster seed addresses
  nodes:
    - valkey-node-0.valkey.svc.cluster.local:6379
    - valkey-node-1.valkey.svc.cluster.local:6379
    - valkey-node-2.valkey.svc.cluster.local:6379

  # Authentication
  auth:
    passwordSecret:
      name: valkey-auth-secret
      key: password

  # TLS configuration
  tls:
    enabled: true
    caFile: /etc/valkey-tls/ca.crt

  # Latency history collection
  latency:
    enabled: true
    historyDepth: 100   # events per command
    eventResetInterval: 1h

  # Cluster topology discovery
  cluster:
    enabled: true
    pollInterval: 5s

remoteWrite:
  endpoint: https://ingest.atatus.net/api/v1/write
  bearerToken: "${TIGEROPS_API_KEY}"
  scrapeInterval: 15s

# Alert thresholds
alerts:
  memoryUsagePct: 80
  evictedKeysPerMinute: 100
  replicationLagBytes: 1048576   # 1 MB
  clusterSlotsOk: 16384
  commandLatencyP99Ms: 10
FAQ

Common Questions

Is TigerOps compatible with Valkey and Redis simultaneously?

Yes. TigerOps uses the standard Redis INFO, LATENCY, and CLUSTER commands, which are fully compatible with Valkey 7.x and Redis 6.x/7.x. You can monitor Valkey and Redis nodes in the same TigerOps workspace with unified dashboards.

How does TigerOps monitor Valkey cluster failover events?

The TigerOps exporter polls CLUSTER INFO and CLUSTER NODES every 5 seconds. When a failover occurs, TigerOps records the event, annotates the metric timeline, and fires an alert with the affected node, slot range, and failover duration.

Can TigerOps track per-key access patterns for hotspot detection?

TigerOps uses the OBJECT FREQ command (LFU policy) and MONITOR sampling to detect hot keys without incurring the overhead of full keyspace monitoring. You can configure the sampling rate and top-N hotspot reporting threshold.

Does TigerOps support Valkey Sentinel deployments?

Yes. TigerOps connects to Valkey Sentinel instances to monitor sentinel health, quorum status, and automatic failover events alongside the primary and replica node metrics. Sentinel topology changes are recorded and alerted automatically.

How does TigerOps predict Valkey memory exhaustion?

TigerOps uses linear regression on the used_memory metric over the last 24 hours to forecast when memory will reach your maxmemory threshold. You receive an early warning with the estimated time to exhaustion and the top key families contributing to growth.

Get Started

Stop Discovering Valkey Memory Exhaustion During Peak Traffic

Predictive memory alerts, command latency tracking, and AI hotspot detection. Deploy in 5 minutes.