Skip to content
|
search
Plugin Hub
header icon

Prometheus

Expose metrics related to Kong and proxied upstream services in Prometheus exposition format, which can be scraped by a Prometheus Server.

Configuration Reference

This plugin is compatible with DB-less mode.

In DB-less mode, you configure Kong Gateway declaratively. Therefore, the Admin API is mostly read-only. The only tasks it can perform are all related to handling the declarative config, including:

  • Setting a target's health status in the load balancer
  • Validating configurations against schemas
  • Uploading the declarative configuration using the /config endpoint

The database will always be reported as reachable in Prometheus with DB-less. Additionally, the DB entity count metric (kong_db_entities_total) is not emitted in DB-less mode.

Example plugin configuration

The following examples provide some typical configurations for enabling the prometheus plugin on a service.

Make the following request:

curl -X POST http://localhost:8001/services/SERVICE_NAME|SERVICE_ID/plugins \
    --data "name=prometheus"

Replace SERVICE_NAME|SERVICE_ID with the id or name of the service that this plugin configuration will target.

First, create a KongPlugin resource:

apiVersion: configuration.konghq.com/v1
kind: KongPlugin
metadata:
  name: prometheus-example
config:
  EXAMPLE_PARAMETER: EXAMPLE_VALUE
plugin: prometheus

Next, apply the KongPlugin resource to a service by annotating the service as follows:

apiVersion: v1
kind: Service
metadata:
  name: SERVICE_NAME|SERVICE_ID
  labels:
    app: SERVICE_NAME|SERVICE_ID
  annotations:
    konghq.com/plugins: prometheus-example
spec:
  ports:
  - port: 80
    targetPort: 80
    protocol: TCP
    name: SERVICE_NAME|SERVICE_ID
  selector:
    app: SERVICE_NAME|SERVICE_ID

Replace SERVICE_NAME|SERVICE_ID with the id or name of the service that this plugin configuration will target.

Note: The KongPlugin resource only needs to be defined once and can be applied to any service, consumer, or route in the namespace. If you want the plugin to be available cluster-wide, create the resource as a KongClusterPlugin instead of KongPlugin.

Add this section to your declarative configuration file:

plugins:
- name: prometheus
  service: SERVICE_NAME|SERVICE_ID
  config:
    EXAMPLE_PARAMETER: EXAMPLE_VALUE

Replace SERVICE_NAME|SERVICE_ID with the id or name of the service that this plugin configuration will target.

You can configure this plugin through the Konnect UI.

From the servicehub icon Service Hub, select a service version, then set up the plugin:

  1. In the Plugins section, click Add Plugin.
  2. Find and select the Prometheus plugin.
  3. Click Create.

You can configure this plugin through the Kong Manager UI.

  1. In Kong Manager, select the workspace.
  2. From the Services section, click View for the service row.
  3. From the plugin section, click Add Plugin.
  4. Find and select the Prometheus plugin.
  5. If the option is available, select Scoped.
  6. Add the service name and ID to the Service field if it is not already pre-filled.
  7. Click Create.

A plugin which is not associated to any service, route, or consumer is considered global, and will be run on every request. Read the Plugin Reference and the Plugin Precedence sections for more information.

The following examples provide some typical configurations for enabling the prometheus plugin globally.

Make the following request:

curl -X POST http://localhost:8001/plugins/ \
  --data "name=prometheus"

Create a KongClusterPlugin resource and label it as global:

apiVersion: configuration.konghq.com/v1
kind: KongClusterPlugin
metadata:
  name: <global-prometheus>
  annotations:
    kubernetes.io/ingress.class: kong
  labels:
    global: \"true\"
config:
  <optional_parameter>: <value>
plugin: prometheus

Add a plugins entry in the declarative configuration file:

plugins:
- name: prometheus
  config:
    EXAMPLE_PARAMETER: EXAMPLE_VALUE

You can configure this plugin through the Kong Manager UI.

  1. In Kong Manager, select the workspace.
  2. Open Plugins from the menu, then click New Plugin.
  3. Find and select the Prometheus plugin.
  4. If the option is available, set the plugin scope to Global.
  5. Click Create.

Parameters

Here's a list of all the parameters which can be used in this plugin's configuration:

Form Parameter Description
name
required

Type: string		 The name of the plugin, in this case prometheus.
service.name or service.id

Type: string		 The name or ID of the service the plugin targets.

Set one of these parameters if adding the plugin to a service through the top-level /plugins endpoint.

Not required if using /services/SERVICE_NAME|SERVICE_ID/plugins.
enabled

Type: boolean

Default value: true		 Whether this plugin will be applied.
config.per_consumer
optional

Type: boolean

Default value: false

A boolean value that determines if per-consumer metrics should be collected. If enabled, the kong_http_requests_total and kong_bandwidth_bytes metrics fill in the consumer label when available.
config.status_code_metrics
optional

Type: boolean

Default value: false

A boolean value that determines if status code metrics should be collected. If enabled, http_requests_total, stream_sessions_total metrics will be exported.
config.lantency_metrics
optional

Type: boolean

Default value: false

A boolean value that determines if status code metrics should be collected. If enabled, kong_latency_ms, upstream_latency_ms and request_latency_ms metrics will be exported.
config.bandwidth_metrics
optional

Type: boolean

Default value: false

A boolean value that determines if status code metrics should be collected. If enabled, bandwidth_bytes and stream_sessions_total metrics will be exported.
config.upstream_health_metrics
optional

Type: boolean

Default value: false

A boolean value that determines if status code metrics should be collected. If enabled, upstream_target_health metric will be exported.

Metrics tracked by this plugin are available on both the Admin API and Status API at the http://localhost:<port>/metrics endpoint. Note that the URL to those APIs will be specific to your installation; see Accessing the metrics.

This plugin records and exposes metrics at the node level. Your Prometheus server will need to discover all Kong nodes via a service discovery mechanism, and consume data from each node’s configured /metrics endpoint.

Grafana dashboard

Metrics exported by the plugin can be graphed in Grafana using a drop in dashboard: https://grafana.com/dashboards/7424.

Available metrics

  • DB reachability: A gauge type with a value of 0 or 1, which represents whether DB can be reached by a Kong node.
  • Connections: Various Nginx connection metrics like active, reading, writing, and number of accepted connections.
  • Dataplane Status: The last seen timestamp, config hash, config sync status and certificate expiration timestamp for data plane nodes is exported to control plane.
  • Enterprise License Information: The Kong Gateway license expiration date, features and license signature. Those metrics are only exported on Kong Gateway.
  • DB Entity Count : A gauge metric that measures the current number of database entities.
  • Number of Nginx timers : A gauge metric that measures the total number of Nginx timers, in Running or Pending state.

Following metrics are disabled by default as it may create high cardinality of metrics and may cause performance issues:

When status_code_metrics is set to true:

  • Status codes: HTTP status codes returned by upstream services. These are available per service, across all services, and per route per consumer.

When lantency_metrics is set to true:

  • Latencies Histograms: Latency (in ms), as measured at Kong:
    • Request: Total time taken by Kong and upstream services to serve requests.
    • Kong: Time taken for Kong to route a request and run all configured plugins.
    • Upstream: Time taken by the upstream service to respond to requests.

When bandwidth_metrics is set to true:

  • Bandwidth: Total Bandwidth (egress/ingress) flowing through Kong. This metric is available per service and as a sum across all services.

When upstream_health_metrics is set to true:

  • Target Health: The healthiness status (healthchecks_off, healthy, unhealthy, or dns_error) of targets belonging to a given upstream as well as their subsystem (http or stream).

Here is an example of output you could expect from the /metrics endpoint:

curl -i http://localhost:8001/metrics

Response:

HTTP/1.1 200 OK
Server: openresty/1.15.8.3
Date: Tue, 7 Jun 2020 16:35:40 GMT
Content-Type: text/plain; charset=UTF-8
Transfer-Encoding: chunked
Connection: keep-alive
Access-Control-Allow-Origin: *

# HELP kong_bandwidth_bytes Total bandwidth (ingress/egress) throughput in bytes
# TYPE kong_bandwidth_bytes counter
kong_bandwidth_bytes{service="google",route="google.route-1",direction="egress",consumer=""} 264
kong_bandwidth_bytes{service="google",route="google.route-1",direction="ingress",consumer=""} 93
# HELP kong_datastore_reachable Datastore reachable from Kong, 0 is unreachable
# TYPE kong_datastore_reachable gauge
kong_datastore_reachable 1
# HELP kong_http_requests_total HTTP status codes per consumer/service/route in Kong
# TYPE kong_http_requests_total counter
kong_http_requests_total{service="google",route="google.route-1",code="200",source="service",consumer=""} 1
# HELP kong_node_info Kong Node metadata information
# TYPE kong_node_info gauge
kong_node_info{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",version="3.0.0"} 1
# HELP kong_kong_latency_ms Latency added by Kong and enabled plugins for each service/route in Kong
# TYPE kong_kong_latency_ms histogram
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="5"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="7"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="10"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="15"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="20"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="30"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="50"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="75"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="100"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="200"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="500"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="750"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="1000"} 1
kong_kong_latency_ms_bucket{service="google",route="google.route-1",le="+Inf"} 1
kong_kong_latency_ms_count{service="google",route="google.route-1"} 1
kong_kong_latency_ms_sum{service="google",route="google.route-1"} 4
# HELP kong_memory_lua_shared_dict_bytes Allocated slabs in bytes in a shared_dict
# TYPE kong_memory_lua_shared_dict_bytes gauge
kong_memory_lua_shared_dict_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong",kong_subsystem="http"} 40960
kong_memory_lua_shared_dict_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_cluster_events",kong_subsystem="http"} 40960
kong_memory_lua_shared_dict_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_core_db_cache",kong_subsystem="http"} 823296
kong_memory_lua_shared_dict_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_core_db_cache_miss",kong_subsystem="http"} 90112
kong_memory_lua_shared_dict_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_db_cache",kong_subsystem="http"} 794624
kong_memory_lua_shared_dict_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_db_cache_miss",kong_subsystem="http"} 86016
kong_memory_lua_shared_dict_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_healthchecks",kong_subsystem="http"} 40960
kong_memory_lua_shared_dict_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_locks",kong_subsystem="http"} 61440
kong_memory_lua_shared_dict_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_process_events",kong_subsystem="http"} 40960
kong_memory_lua_shared_dict_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_rate_limiting_counters",kong_subsystem="http"} 86016
kong_memory_lua_shared_dict_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="prometheus_metrics",kong_subsystem="http"} 57344
# HELP kong_memory_lua_shared_dict_total_bytes Total capacity in bytes of a shared_dict
# TYPE kong_memory_lua_shared_dict_total_bytes gauge
kong_memory_lua_shared_dict_total_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong",kong_subsystem="http"} 5242880
kong_memory_lua_shared_dict_total_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_cluster_events",kong_subsystem="http"} 5242880
kong_memory_lua_shared_dict_total_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_core_db_cache",kong_subsystem="http"} 134217728
kong_memory_lua_shared_dict_total_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_core_db_cache_miss",kong_subsystem="http"} 12582912
kong_memory_lua_shared_dict_total_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_db_cache",kong_subsystem="http"} 134217728
kong_memory_lua_shared_dict_total_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_db_cache_miss",kong_subsystem="http"} 12582912
kong_memory_lua_shared_dict_total_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_healthchecks",kong_subsystem="http"} 5242880
kong_memory_lua_shared_dict_total_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_locks",kong_subsystem="http"} 8388608
kong_memory_lua_shared_dict_total_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_process_events",kong_subsystem="http"} 5242880
kong_memory_lua_shared_dict_total_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="kong_rate_limiting_counters",kong_subsystem="http"} 12582912
kong_memory_lua_shared_dict_total_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",shared_dict="prometheus_metrics",kong_subsystem="http"} 5242880
# HELP kong_memory_workers_lua_vms_bytes Allocated bytes in worker Lua VM
# TYPE kong_memory_workers_lua_vms_bytes gauge
kong_memory_workers_lua_vms_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",pid="21173",kong_subsystem="http"} 64329517
kong_memory_workers_lua_vms_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",pid="21174",kong_subsystem="http"} 46314808
kong_memory_workers_lua_vms_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",pid="21175",kong_subsystem="http"} 46681598
kong_memory_workers_lua_vms_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",pid="21176",kong_subsystem="http"} 46637209
kong_memory_workers_lua_vms_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",pid="21177",kong_subsystem="http"} 46234336
kong_memory_workers_lua_vms_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",pid="21178",kong_subsystem="http"} 46180420
kong_memory_workers_lua_vms_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",pid="21179",kong_subsystem="http"} 46161105
kong_memory_workers_lua_vms_bytes{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",pid="21180",kong_subsystem="http"} 46366877
# HELP kong_nginx_connections_total Number of connections by subsystem
# TYPE kong_nginx_connections_total gauge
kong_nginx_connections_total{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",subsystem="http",state="accepted"} 296
kong_nginx_connections_total{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",subsystem="http",state="active"} 9
kong_nginx_connections_total{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",subsystem="http",state="handled"} 296
kong_nginx_connections_total{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",subsystem="http",state="reading"} 0
kong_nginx_connections_total{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",subsystem="http",state="total"} 296
kong_nginx_connections_total{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",subsystem="http",state="waiting"} 0
kong_nginx_connections_total{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",subsystem="http",state="writing"} 9
# HELP kong_nginx_metric_errors_total Number of nginx-lua-prometheus errors
# TYPE kong_nginx_metric_errors_total counter
kong_nginx_metric_errors_total 0
# HELP kong_nginx_requests_total Total number of requests
# TYPE kong_nginx_requests_total gauge
kong_nginx_requests_total{node_id="849373c5-45c1-4c1d-b595-fdeaea6daed8",subsystem="http"} 296
# HELP kong_nginx_timers Number of Nginx timers
# TYPE kong_nginx_timers gauge
kong_nginx_timers{state="pending"} 1
kong_nginx_timers{state="running"} 39
# HELP kong_request_latency_ms Total latency incurred during requests for each service/route in Kong
# TYPE kong_request_latency_ms histogram
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="25"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="50"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="80"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="100"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="250"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="400"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="700"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="1000"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="2000"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="5000"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="10000"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="30000"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="60000"} 1
kong_request_latency_ms_bucket{service="google",route="google.route-1",le="+Inf"} 1
kong_request_latency_ms_count{service="google",route="google.route-1"} 1
kong_request_latency_ms_sum{service="google",route="google.route-1"} 6
# HELP kong_upstream_latency_ms Latency added by upstream response for each service/route in Kong
# TYPE kong_upstream_latency_ms histogram
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="25"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="50"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="80"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="100"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="250"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="400"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="700"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="1000"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="2000"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="5000"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="10000"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="30000"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="60000"} 1
kong_upstream_latency_ms_bucket{service="google",route="google.route-1",le="+Inf"} 1
kong_upstream_latency_ms_count{service="google",route="google.route-1"} 1
kong_upstream_latency_ms_sum{service="google",route="google.route-1"} 2

Note: Upstream targets’ health information is exported once per subsystem. If both stream and HTTP listeners are enabled, targets’ health will appear twice. Health metrics have a subsystem label to indicate which subsystem the metric refers to.

Accessing the metrics

In most configurations, the Kong Admin API will be behind a firewall or would need to be set up to require authentication. Here are a couple of options to allow access to the /metrics endpoint to Prometheus:

  1. If the Status API is enabled, then its /metrics endpoint can be used. This is the preferred method.

  2. The /metrics endpoint is also available on the Admin API, which can be used if the Status API is not enabled. Note that this endpoint is unavailable when RBAC is enabled on the Admin API (Prometheus does not support Key-Auth to pass the token).

Changelog

Kong Gateway 3.0.x

  • High cardinality metrics are now disabled by default.
  • Decreased performance penalty to proxy traffic when collecting metrics.
  • The following metric names were adjusted to add units to standardize where possible:
    • http_status to http_requests_total
    • latency to kong_request_latency_ms/kong_upstream_latency_ms/kong_kong_latency_ms
    • kong_bandwidh to kong_bandwidth_bytes
    • nginx_http_current_connections/nginx_stream_current_connections to nginx_connections_total
    • Removed: http_consumer_status
  • New metric: session_duration_ms for monitoring stream connections
  • New metric: node_info is a single gauge set to 1 that outputs the node’s ID and Kong Gateway version
  • Latency was split into four different metrics: kong_latency_ms, upstream_latency_ms, request_latency_ms (HTTP), and session_duration_ms (stream). Buckets details follow:
    • Kong Latency and Upstream Latency can operate at orders of different magnitudes. Separate these buckets to reduce memory overhead.
  • request_count and consumer_status were merged into http_requests_total. If the per_consumer config is set to false, the consumer label will be empty. If the per_consumer config is true, the consumer label will be filled.
  • http_requests_total has a new label source. It can be set to exit, error, or service.
  • All Memory metrics have a new label, node_id.
  • Plugin version bumped to 3.0.0
  • The node_id label was added to memory metrics.

Kong Gateway 2.8.x

  • Adds a new metric:
    • kong_nginx_timers (gauge): total number of Nginx timers, in Running or Pending state.
  • Add two new metrics:
    • kong_db_entities_total (gauge): total number of entities in the database
    • kong_db_entity_count_errors (counter): measures the number of errors encountered during the measurement of kong_db_entities_total

Kong Gateway 2.5.x

  • New data_plane_cluster_cert_expiry_timestamp metric
  • Added subsystem label to Upstream Target health metrics

Kong Gateway 2.4.x

  • Added the per_consumer configuration parameter to export per-consumer status.

Kong Gateway 2.3.x

  • The plugin can now export Kong Enterprise licensing information.