What are Metrics in Prometheus?

About

This page is about metrics in Prometheus.

Example

Base

api_http_requests_total{method="POST", handler="/messages"}

where:

• the metric name is api_http_requests_total
• the labels are:
  • method=“POST”
  • handler=“/messages”

Metrics from the Prometheus server itself

prometheus_http_requests_total{code="200",handler="/"} 0
prometheus_http_requests_total{code="200",handler="/-/healthy"} 0
prometheus_http_requests_total{code="200",handler="/-/quit"} 0
prometheus_http_requests_total{code="200",handler="/-/ready"} 0
prometheus_http_requests_total{code="200",handler="/-/reload"} 0
prometheus_http_requests_total{code="200",handler="/alerts"} 0
prometheus_http_requests_total{code="200",handler="/api/v1/*path"} 0
prometheus_http_requests_total{code="200",handler="/api/v1/admin/tsdb/clean_tombstones"} 0
prometheus_http_requests_total{code="200",handler="/api/v1/admin/tsdb/delete_series"} 0
prometheus_http_requests_total{code="200",handler="/api/v1/admin/tsdb/snapshot"} 0
prometheus_http_requests_total{code="200",handler="/api/v1/alertmanagers"} 0

Syntax

A metric identifier is composed uniquely of:

• a mandatory name label
• and one or more labels.

<metric name>{<label name>=<label value>, ...}
# equivalent to
{name="metric name", <label name>=<label value>, ...}

This is the same notation that OpenTSDB uses.

The metrics identifier is also known as time serie because when they are read/scraped, they create a time series with the following values:

• metric identifier
• time
• value

Name

The metric name is a built-in label that identifies a metrics uniquely (ie a serie)

Example for the total number of HTTP requests received:

http_requests_total

is equivalent to

{ name="http_requests_total" }

The name may contain:

• ASCII letters
• digits,
• underscores,
• and colons.

It must match the regex: [a-zA-Z_:][a-zA-Z0-9_:]*.

It must comply to the convention. For example:

• an accumulating count (counter) has total as a suffix

Management

Conf

See Prometheus - Collector to configure when the metrics are collected.

Location

By default, the metrics are made available at a URL endpoint known as a target with /metrics as path

Example:

• http://hostname/metrics
• http://localhost:9090/metrics for the metrics of Prometheus with the default conf.

Example of metrics for the prometheus server.

# HELP go_gc_duration_seconds A summary of the pause duration of garbage collection cycles.
# TYPE go_gc_duration_seconds summary
go_gc_duration_seconds{quantile="0"} 0
go_gc_duration_seconds{quantile="0.25"} 0
go_gc_duration_seconds{quantile="0.5"} 0
go_gc_duration_seconds{quantile="0.75"} 0
go_gc_duration_seconds{quantile="1"} 0.0010366
go_gc_duration_seconds_sum 0.0020374
go_gc_duration_seconds_count 15
# HELP go_goroutines Number of goroutines that currently exist.
# TYPE go_goroutines gauge
go_goroutines 41
.......

Built-in attributes: Job / instance

When Prometheus scrapes a target, it attaches some labels automatically to the scraped time series which serve to identify the scraped target:

• job: The configured job name that the target belongs to.
• instance: The <host>:<port> part of the target's URL that was scraped.

More … Doc

How to get all metrics in the Prometheus server?

With the API

http://localhost:9090/api/v1/label/__name__/values