{
  "id": "metrics_stream_engine",
  "title": "Monitoring v2",
  "url": "https://redis.io/docs/latest/operate/rs/8.0/monitoring/metrics_stream_engine/",
  "summary": "The new metrics engine for monitoring Redis Software.",
  "content": "\nThe new metrics stream engine is generally available as of [Redis Software version 8.0](https://redis.io/docs/latest/operate/rs/release-notes/rs-8-0-releases).\n\nThe new metrics stream engine:\n\n- Exposes the v2 Prometheus scraping endpoint at `https://\u003cIP\u003e:8070/v2`.\n\n- Exports all time-series metrics to external monitoring tools such as Grafana, DataDog, NewRelic, and Dynatrace using Prometheus.\n\n- Enables real-time monitoring, including full monitoring during maintenance operations, which provides full visibility into performance during events such as shards' failovers and scaling operations.\n\n## Integrate with external monitoring tools\n\nTo integrate Redis Software metrics into your monitoring environment, see the following integration guides:\n\n- [Grafana](https://redis.io/docs/latest/integrate/prometheus-with-redis-enterprise)\n\n- [Datadog](https://redis.io/docs/latest/integrate/datadog-with-redis-enterprise)\n\n- [Dynatrace](https://redis.io/docs/latest/integrate/dynatrace-with-redis-enterprise)\n\n- [New Relic](https://redis.io/docs/latest/integrate/new-relic-with-redis-enterprise)\n\nFor a detailed tutorial to deploy a complete monitoring stack with Prometheus and Grafana, see [Redis Software Observability with Prometheus and Grafana](https://redis.io/learn/operate/observability/redis-software-prometheus-and-grafana).\n\nFilter [Libraries and tools](https://redis.io/docs/latest/integrate) by \"observability\" for additional tools and guides.\n\n## Prometheus metrics v2\n\nFor a list of all available v2 metrics, see [Prometheus metrics v2](https://redis.io/docs/latest/operate/rs/monitoring/metrics_stream_engine/prometheus-metrics-v2).\n\nThe v2 scraping endpoint also exposes metrics for `node_exporter` version 1.8.1. For more information, see the [Prometheus node_exporter GitHub repository](https://github.com/prometheus/node_exporter).\n\n## Transition from Prometheus v1 to Prometheus v2\n\nIf you are already using the existing scraping endpoint for integration, do the following to transition from v1 metrics to v2 metrics:\n\n1. Change the `metrics_path` in your Prometheus configuration file from `/` to `/v2` to use the new scraping endpoint.\n\n    Here's an example of the updated scraping configuration in `prometheus.yml`:\n\n    ```yaml\n    scrape_configs:\n      # Scrape Redis Software\n      - job_name: redis-enterprise\n        scrape_interval: 30s\n        scrape_timeout: 30s\n        metrics_path: /v2\n        scheme: https\n        tls_config:\n          insecure_skip_verify: true\n        static_configs:\n          - targets: [\"\u003ccluster_name\u003e:8070\"]\n    ```\n\n1. Use the metrics tables in [this guide](https://redis.io/docs/latest/operate/rs/monitoring/metrics_stream_engine/prometheus-metrics-v1-to-v2) to transition from v1 metrics to equivalent v2 PromQL.\n\nIt is possible to scrape both existing and new endpoints simultaneously, allowing advanced dashboard preparation and a smooth transition.\n\n## Best practices for monitoring\n\nFollow these best practices when monitoring your Redis Software cluster using the metrics stream engine.\n\n### Monitor host-level metrics\n\nFor cluster health, resources, and node stability, monitor these metrics:\n\n| Group | Metric | Why monitor | Unit |\n|-------|--------|-------------|------|\n| CPU utilization | `node_cpu_user`,\u003cbr /\u003e`node_cpu_system` | Detect CPU saturation from Redis or the OS that results in higher latency and queueing. | Seconds (counter) |\n| Memory (freeable) | \u003cspan class=\"break-all\"\u003e`node_memory_MemTotal_bytes`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`node_memory_MemFree_bytes`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`node_memory_Buffers_bytes`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`node_memory_Cached_bytes`\u003c/span\u003e | Detect memory pressure early. Low free memory or cache can precede swapping or out-of-memory errors. | Bytes (gauge) |\n| Swap usage | \u003cspan class=\"break-all\"\u003e`node_ephemeral_storage_free`\u003c/span\u003e | Monitor memory and disk pressure in your setup. Sustained pressure leads to latency spikes. | Bytes (gauge) |\n| Network traffic | \u003cspan class=\"break-all\"\u003e`node_ingress_bytes`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`node_egress_bytes`\u003c/span\u003e | Ensure the network interface is not saturated. Protects replication and client responsiveness. | Bytes (counter) |\n| Disk space | \u003cspan class=\"break-all\"\u003e`node_filesystem_avail_bytes`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`node_filesystem_size_bytes`\u003c/span\u003e | Prevent persistence and logging outages from low disk space. | Bytes (gauge) |\n| Cluster state | `has_quorum{…}` | Monitor whether quorum is maintained (1) or lost (0). | Boolean |\n| | `node_metrics_up` | Monitor whether the node is connected and reporting to the cluster. | Gauge |\n| Licensing | `license_shards_limit` | Track shard capacity limits by type (RAM or flash). | Count |\n| Certificates | \u003cspan class=\"break-all\"\u003e`node_cert_expires_in_seconds`\u003c/span\u003e | Avoid downtime from expired node certificates. | Seconds (gauge) |\n| Services – CPU | \u003cspan class=\"break-all\"\u003e`namedprocess_namegroup_cpu_seconds_total`\u003c/span\u003e | Identify abnormal CPU usage by platform services that can starve Redis, such as `alert_mgr`, `redis_mgr`, `dmc_proxy`. | Seconds (counter) |\n| Services – memory | \u003cspan class=\"break-all\"\u003e`namedprocess_namegroup_memory_bytes`\u003c/span\u003e | Detect memory leaks or outliers in platform services, such as `alert_mgr`, `redis_mgr`, `dmc_proxy`. | Bytes (gauge) |\n\n### Monitor database-level metrics\n\nFor database performance, availability, and efficiency, monitor the following metrics:\n\n| Group | Metric | Why monitor | Unit |\n|-------|--------|-------------|------|\n| Memory | \u003cspan class=\"break-all\"\u003e`redis_server_used_memory`\u003c/span\u003e | Track actual data memory to prevent out-of-memory errors and evictions. | Bytes |\n| Memory | \u003cspan class=\"break-all\"\u003e`redis_server_allocator_allocated`\u003c/span\u003e | Monitor bytes allocated by allocator (includes internal fragmentation). | Bytes |\n| Memory | \u003cspan class=\"break-all\"\u003e`redis_server_allocator_active`\u003c/span\u003e | Monitor bytes in active pages (includes external fragmentation). Use delta/ratio versus allocated to infer defraggable memory. | Bytes |\n| Memory | \u003cspan class=\"break-all\"\u003e`redis_server_active_defrag_running`\u003c/span\u003e | Monitor if defragmentation is active and the intended CPU %. High values can affect performance. | % (gauge) |\n| Latency | \u003cspan class=\"break-all\"\u003e`endpoint_read_requests_latency_histogram`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_write_requests_latency_histogram`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_other_requests_latency_histogram`\u003c/span\u003e | Monitor server-side command latency. | Microseconds |\n| High availability | \u003cspan class=\"break-all\"\u003e`redis_server_master_repl_offset`\u003c/span\u003e | Compute replica throughput and lag using deltas over time. | Bytes (counter) |\n| High availability | \u003cspan class=\"break-all\"\u003e`redis_server_master_link_status`\u003c/span\u003e | Monitor replica link status (up or down) for early warning of high availability risk. | Status |\n| Active-Active | \u003cspan class=\"break-all\"\u003e`database_syncer_dst_lag`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`database_syncer_lag_ms`\u003c/span\u003e | Detect cross-region synchronization delays that impact consistency and SLAs. | Milliseconds (gauge) |\n| Active-Active | \u003cspan class=\"break-all\"\u003e`database_syncer_state`\u003c/span\u003e | Monitor operational state for troubleshooting synchronization issues. | Gauge |\n| Traffic – requests | \u003cspan class=\"break-all\"\u003e`endpoint_read_requests`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_write_requests`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_other_requests`\u003c/span\u003e | Monitor workload mix and spikes that drive capacity and latency. Total equals the sum of all three. | Counter |\n| Traffic – responses | \u003cspan class=\"break-all\"\u003e`endpoint_read_responses`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_write_responses`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_other_responses`\u003c/span\u003e | Validate service responsiveness and symmetry with requests. | Counter |\n| Traffic – bytes | \u003cspan class=\"break-all\"\u003e`endpoint_ingress`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_egress`\u003c/span\u003e | Monitor size trends and watch for sudden growth that impacts egress costs or bandwidth. | Bytes (counter) |\n| Egress queue | \u003cspan class=\"break-all\"\u003e`endpoint_egress_pending`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_egress_pending_discarded`\u003c/span\u003e | Monitor back-pressure and drops that indicate network or client issues. | Bytes (counter) |\n| Connections | \u003cspan class=\"break-all\"\u003e`endpoint_client_connection`\u003c/span\u003e | Monitor accepted connections over time and match against client rollouts or spikes. | Counter |\n| Connections | \u003cspan class=\"break-all\"\u003e`endpoint_client_connection_expired`\u003c/span\u003e | Monitor connections closed due to TTL expiry, which can indicate idle policy or client issues. | Counter |\n| Connections | \u003cspan class=\"break-all\"\u003e`endpoint_longest_pipeline_histogram`\u003c/span\u003e | Monitor long pipelines that can amplify latency bursts and detect misbehaving clients. | Histogram (count) |\n| Connections | \u003cspan class=\"break-all\"\u003e`endpoint_client_connections`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_client_disconnections`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_proxy_disconnections`\u003c/span\u003e | Monitor connection churn and identify who closed the socket (client versus proxy). Current connections ≈ connections − disconnections. | Counter |\n| Cache efficiency | \u003cspan class=\"break-all\"\u003e`redis_server_db_keys`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`redis_server_db_avg_ttl`\u003c/span\u003e | Monitor key inventory and TTL coverage to inform eviction strategy. | Counter |\n| Cache efficiency | \u003cspan class=\"break-all\"\u003e`redis_server_evicted_keys\t`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`redis_server_expired_keys`\u003c/span\u003e | Monitor eviction and expiry rates. Frequent evictions indicate memory pressure or poor sizing. | Counter |\n| Cache efficiency | `cache_hits`,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`cache_hit_rate`\u003c/span\u003e | Monitor hit rate, which drives read latency and cost. Cache hit rate equals \u003cspan class=\"break-all\"\u003ecache_hits/(cache_hits+cache_misses)\u003c/span\u003e. | Count / Ratio (%) |\n| Cache efficiency | \u003cspan class=\"break-all\"\u003e`endpoint_client_tracking_on_requests`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_client_tracking_off_requests`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_disposed_commands_after_client_caching`\u003c/span\u003e | Track client-side caching usage and misuse. | Counter |\n| Big / complex keys | \u003cspan class=\"break-all\"\u003e`redis_server_\u003cdata_type\u003e_\u003csize_or_items\u003e_\u003cbucket\u003e`\u003c/span\u003e | Monitor oversized keys and cardinality that cause fragmentation, slow replication, and CPU spikes. Track to prevent incidents. Examples:\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`strings_sizes_over_512M`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`zsets_items_over_8M`\u003c/span\u003e | Gauge |\n| Security – clients | \u003cspan class=\"break-all\"\u003e`endpoint_client_expiration_refresh`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_client_establishment_failures`\u003c/span\u003e | Monitor unstable clients or problems with authentication or setup. | Counter |\n| Security – LDAP | \u003cspan class=\"break-all\"\u003e`endpoint_successful_ldap_authentication`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_failed_ldap_authentication`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_disconnected_ldap_client`\u003c/span\u003e | Monitor authentication health and detect brute-force attacks or misconfigurations. | Counter |\n| Security – cert-based | \u003cspan class=\"break-all\"\u003e`endpoint_successful_cba_authentication`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_failed_cba_authentication`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`endpoint_disconnected_cba_client`\u003c/span\u003e | Monitor certificate authentication status and failures. | Counter |\n| Security – password | \u003cspan class=\"break-all\"\u003e`endpoint_disconnected_user_password_client`\u003c/span\u003e | Monitor password-authentication client disconnects and correlate with policy changes. | Counter |\n| Security – ACL | \u003cspan class=\"break-all\"\u003e`redis_server_acl_access_denied_auth`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`redis_server_acl_access_denied_cmd`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`redis_server_acl_access_denied_key`\u003c/span\u003e,\u003cbr /\u003e\u003cspan class=\"break-all\"\u003e`redis_server_acl_access_denied_channel`\u003c/span\u003e | Monitor unauthorized access attempts and incorrectly scoped ACLs. | Counter |\n| Configuration | \u003cspan class=\"break-all\"\u003e`db_config`\u003c/span\u003e| This is an information metric that holds database configuration within labels such as: db_name, db_version, db_port, tls_mode. | counter |\n\n",
  "tags": ["docs","operate","rs","kubernetes"],
  "last_updated": "2026-07-23T12:00:25-05:00",
  "children": [{"id":"prometheus-metrics-v1-to-v2","summary":"Transition from v1 metrics to v2 PromQL equivalents.","title":"Transition from Prometheus v1 to Prometheus v2","url":"https://redis.io/docs/latest/operate/rs/8.0/monitoring/metrics_stream_engine/prometheus-metrics-v1-to-v2/"},{"id":"prometheus-metrics-v2","summary":"V2 metrics available to Prometheus as of Redis Software version 7.8.2.","title":"Prometheus metrics v2","url":"https://redis.io/docs/latest/operate/rs/8.0/monitoring/metrics_stream_engine/prometheus-metrics-v2/"},{"id":"alerts-v1-to-v2","summary":"Transition from internal cluster manager alerts to external monitoring alerts using Prometheus.","title":"Transition cluster manager alerts to Prometheus alerts","url":"https://redis.io/docs/latest/operate/rs/8.0/monitoring/metrics_stream_engine/alerts-v1-to-v2/"}]
}