Skip to content

Latest commit

 

History

History
90 lines (64 loc) · 2.77 KB

metrics.md

File metadata and controls

90 lines (64 loc) · 2.77 KB
Raw

Metrics

The metrics system must be enabled via metrics.enabled. If enabled, metrics are exposed at http://localhost:2112/metrics.

You can check that the metrics system is working without starting a Prometheus server:

$ curl 127.0.0.1:2112/metrics

Available metrics

A few metrics are currently exposed (just for demonstration purposes):

  • sedge_completed_total: number of completed invocations (Counter, per function)
  • sedge_exectime: execution time for each function (Histogram, per function)

Prometheus Integration

Various Prometheus configurations can be considered to scrape Serverledge metrics:

  • A centralized Prometheus server in the Cloud (likely not scalable...)
  • A Prometheus server in each Edge zone
  • A Prometheus server in the Cloud with a Prometheus Agent on each Serverledge node (details below)

Example: Prometheus Agent + Cloud

As regards the last option, it requires Prometheus instances to use the following (minimal) configuration.

In the Serverledge node, Prometheus must be started with --enable-feature=agent and the following lines in the configuration:

remote_write:
  - url: "http://<prometheus_cloud_host>:9091/api/v1/write"

Example configuration:

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).

# A scrape configuration containing exactly one endpoint to scrape:
scrape_configs:
  - job_name: "serverledge"
    # metrics_path defaults to '/metrics'
    # scheme defaults to 'http'.
    static_configs:
      - targets: ["<serverledge_host>:2112"]

remote_write:
  - url: "http://<prometheus_cloud_host>:9091/api/v1/write"

In the Cloud, Prometheus must be started with --web.enable-remote-write-receiver. Example configuration:

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.

Example script to launch both Prometheus instances on the same host (for testing):

docker run \
    --name prom \
    -d \
    -p 9090:9090 \
    -v $(pwd)/prometheus.yml:/etc/prometheus/prometheus.yml \
    prom/prometheus --enable-feature=agent \
	--config.file=/etc/prometheus/prometheus.yml

docker run \
    --name promRemote \
    -d\
    \
    -p 9091:9090 \
    -v $(pwd)/prometheus_remote.yml:/etc/prometheus/prometheus.yml \
    prom/prometheus --web.enable-remote-write-receiver \
	--config.file=/etc/prometheus/prometheus.yml

References