Prometheus

Introduction

Prometheus is a tool/database that is used for monitoring. Prometheus adopts a pull based model in getting metrics data by querying each target defined in its configuration.

Prometheus scrapes metrics from instrumented jobs, either directly or via an intermediary push gateway for short-lived jobs. It stores all scraped samples locally and runs rules over this data to either aggregate and record new time series from existing data or generate alerts. Grafana or other API consumers can be used to visualize the collected data.

Our light-4j Prometheus metrics handler collects the API runtime information and saves it to the Prometheus Metric data module. The Prometheus server will pull the metric from metrics_path which is configurable on the Prometheus config yml file (for example: /v1/prometheus) ; The pulling interval is second base on the config file (scrape_interval: 30s). A Grafana instance is hooked to Prometheus to output the metrics on dashboard from two different perspectives.

By default, Prometheus metric targets (nodes to monitor) are provided statically in the prometheus config file. Static targets are simple when we know in advance our infrastructure. But for the microservice architecture, the service API will have dynamic targets on the cloud. So our solution is use consul to provide targets for prometheus to monitor dynamically.

Consul is a service discovery tool by hashicorp, it allows services/virtual machines to be registered to it and then provide dns and http interfaces to query on the state of the registered services/virtual machines.

In Prometheus, we need to configure it using consul_sd_targets. Prometheus will then query the Consul http interface for the catalog of targets to monitor.

Configuration

Configuration file (prometheus.yml) for the MiddlewareHandler (com.networknt.metrics.prometheus.PrometheusHandler)


// If metrics handler is enabled or not
enabled: false

 // If the Prometheus hotspot is enabled or not.
// hotspot include thread, memory, classloader,...
enableHotspot: false

And for each service API which you want to use Prometheus, there are several config changes that need to be made on handler.yml:

  1. We need to add Prometheus middleware handler:
   - com.networknt.metrics.prometheus.PrometheusHandler@prometheus
   - com.networknt.metrics.prometheus.PrometheusGetHandler@getprometheus
  1. Add Prometheus middleware handler to the chains:
chains:
 default:
   - exception
   - prometheus
   - traceability
   - correlation
   - specification
   - security
   - body
   - audit
   - sanitizer
   - validator

3.And add endpoint for Prometheus to pull the metrics data:

- path: '/v1/prometheus'
  method: 'get'
  exec:
    - getprometheus

Change the enabled value to true to enable the prometheus metrics handler;

In the light-4j, the following Prometheus Metric are defined in the Prometheus MiddlewareHandler


    public static final String REQUEST_TOTAL = "requests_total";
    public static final String SUCCESS_TOTAL = "success_total";
    public static final String AUTO_ERROR_TOTAL = "auth_error_total";
    public static final String REQUEST_ERROR_TOTAL = "request_error_total";
    public static final String SERVER_ERROR_TOTAL = "server_error_total";
    public static final String RERSPONSE_TIME_SECOND = "response_time_seconds";

Prometheus provides a collector that can scrape and expose the mBeans of a JMX target.

It runs as a JavaAgent, exposing an HTTP server and serving metrics of the local JVM.

If the user wants to have an enabled hotspot monitor for JVM, change the enableHotspot value to true:

enableHotspot: true

The Hotspot monitor metrics include:

CPU and JVM process:

process_cpu_seconds_total
process_open_fds

JVM memory usage:

jvm_memory_pool_allocated_bytes_total
jvm_memory_pool_bytes_init
jvm_memory_pool_bytes_max
jvm_memory_pool_bytes_used

metrics about JVM thread areas

jvm_threads_peak
jvm_threads_daemon
jvm_threads_started_total
jvm_threads_deadlocked
jvm_threads_state

JVM garbage collectors:

jvm_gc_collection_seconds

Buffer:

jvm_buffer_pool_used_bytes
jvm_buffer_pool_capacity_bytes
jvm_buffer_pool_used_buffers

Configuration for Prometheus

Prometheus is configured via a configuration file. The file is written in YAML format, defined by the Prometheus scheme.

Here is the sample config file. It can be found from the link:

# my global config
global:
  scrape_interval:     15s # Set the scrape interval to every 15 seconds. Default is every 1 minute.
  evaluation_interval: 15s # Evaluate rules every 15 seconds. The default is every 1 minute.
  # scrape_timeout is set to the global default (10s).

# Alertmanager configuration
alerting:
  alertmanagers:
  - static_configs:
    - targets:
      # - alertmanager:9093

# Load rules once and periodically evaluate them according to the global 'evaluation_interval'.
rule_files:
  # - "first_rules.yml"
  # - "second_rules.yml"

# A scrape configuration containing exactly one endpoint to scrape:
# Here it's Prometheus itself.
scrape_configs:
  # The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.
  - job_name: 'services'
    scrape_interval: 15s
    consul_sd_configs:
      - server: 'consul:8500'
        services:
          - 'multimodule_tram-todolist-command-service'
          - 'multimodule_tram-todolist-view-service'
          - 'com.networknt.bookstore-service-api-3.0.1'
          - 'com.networknt.computerstore-service-api-3.0.1'
          - 'com.networknt.foodstore-service-api-3.0.1'
          - 'com.networknt.petstore-service-api-3.0.1'
          - 'com.networknt.marketsample-service-3.0.1'

    tls_config:
      # Skip verification until we have resolved why the certificate validation
      # for the kubelet on API server nodes fail.
      insecure_skip_verify: true

    relabel_configs:
      - source_labels: ['__meta_consul_service']
        regex:         '(.*)'
        target_label:  'service'
        replacement:   '$1'
      - source_labels: ['__meta_consul_tags']
        target_label: 'slave_name'
        regex: '.*slave_name=([^,]*).*'
        replacement: '${1}'
      - source_labels: ['__meta_consul_tags']
        target_label: 'public_hostname'
        regex: '.*public_hostname=([^,]*).*'
        replacement: '${1}'


    metrics_path: /v1/prometheus
    # metrics_path defaults to '/metrics'
    # scheme defaults to 'http'.
    scheme: https

  - job_name: 'prometheus'
    static_configs:
      - targets: ['localhost:9090']

    metrics_path: /metrics

Currently we are using https protocol:

 scheme: https

for local environment, we ignore ssl check by setting in config file:

   insecure_skip_verify: true

Customization

Prometheus Metric names and labels

Every time the series is uniquely identified by its metric name and a set of key-value pairs, also known as labels.

The metric name specifies the general feature of a system that is measured (e.g. http_requests_total - the total number of HTTP requests received). It may contain ASCII letters and digits, as well as underscores and colons. It must match the regex [a-zA-Z_:][a-zA-Z0-9_:]*.

Labels enable Prometheus’s dimensional data model: any given combination of labels for the same metric name identifies a particular dimensional instantiation of that metric (for example: all HTTP requests that used the method POST to the /api/tracks handler). The query language allows filtering and aggregation based on these dimensions. Changing any label value, including adding or removing a label, will create a new time series.

And the service “endpoint” and “clientId” are been added as Labels for Prometheus’s dimensional data model.

The sample result for the Prometheus metric will be like below:

requests_total{clientId="unknown",endpoint="/todos@post",instance="172.18.0.6:8081",job="services",service="multimodule_tram-todolist-command-service"}	1
requests_total{clientId="unknown",endpoint="/todoviews@get",instance="172.18.0.7:8082",job="services",service="multimodule_tram-todolist-view-service"}	1

Integrate Prometheus with Grafana:

The Light framework provided docker-compose file (docker-compose-prometheus.yml) includes the Grafana image: image: grafana/grafana:4.6.3

So the user can open a browser to view the Prometheus from Grafana

http://127.0.0.1:3000/

user: admin
passwork: admin

Please follow the following steps to set the Prometheus up in Grafana:

Adding the data source to Grafana:

  1. Open the side menu by clicking the Grafana icon in the top header.

  2. In the side menu under the Dashboards link you should find a link named Data Sources.

  3. Click the + Add data source button in the top header.

  4. Select Prometheus from the Type dropdown.

After Prometheus works as a backend dataSource for Grafana, the user can customize their own template or dashboard in Grafana for Prometheus metrics.

You can get more details here: http://docs.grafana.org/features/datasources/prometheus/

“Prometheus” was last updated: April 5, 2021: Issue246 (#256) (50b1c10)
Improve this page