Configuring Scrape Jobs

1. Defining Static Targets

Property	Description	Example
job_name	Unique identifier for the scrape job	`node_exporter`
static_configs	List of static targets to scrape	`[ { "targets": ["localhost:9100", "server2:9100"] } ]`
scrape_interval	How often to scrape targets (default: 15s)	`30s`
scrape_timeout	Timeout for individual scrape requests (default: 10s)	`5s`

Example: Basic Static Target Configuration

scrape_configs:
  - job_name: "node_exporter"
    static_configs:
      - targets: ["localhost:9100", "server2:9100"]
        labels:
          env: "production"

2. Setting Job-Specific Scrape Interval

Property	Values	Description
scrape_interval	`15s`	How frequently to scrape targets. Default is 15 seconds.
scrape_timeout	`10s`	Timeout for scraping a single target. Default is 10 seconds.

Example: Configuring scrape interval for a job

scrape_configs:
  - job_name: "prometheus"
    scrape_interval: 30s
    static_configs:
      - targets: [localhost:9090]

Note: The scrape_interval setting in a job overrides the global default scrape interval.

3. Configuring Metrics Path

Field	Description	Default	Example
metrics_path	Path to scrape metrics from	`/metrics`	`/app/metrics`
scheme	Protocol to use for scraping (http or https)	`http`	`https`
port	Port to scrape metrics from	`9090`	`8080`

Example: Configuring Metrics Path

scrape_configs:
  - job_name: "myapp"
    metrics_path: /app/metrics
    scheme: https
    static_configs:
      - targets: ["localhost:8080"]

4. Setting Scrape Scheme

Property	Values	Description
scheme	`http`, `https`	Specifies the URL scheme to use when scraping targets. Defaults to `http`.

Example: Configuring HTTP Scrape Scheme

scrape_configs:
  - job_name: "prometheus"
    scheme: http
    static_configs:
      - targets: ['localhost:9090']

Example: Configuring HTTPS Scrape Scheme

scrape_configs:
  - job_name: "secure-metrics"
    scheme: https
    static_configs:
      - targets: ['secure.example.com:8443']

5. Adding Job Labels

Label Name	Value	Description
job_name	`prometheus`	Unique identifier for the scrape job
scrape_interval	`15s`	How frequently to scrape targets
scrape_timeout	`10s`	Timeout for individual scrape requests
metrics_path	`/metrics`	Path to scrape metrics from
honor_labels	`false`	Whether to honor labels from scraped targets
relabel_configs	`[]`	Relabeling rules to apply before scraping

Example: Adding Job Labels

scrape_configs:
  - job_name: "prometheus"
    static_configs:
      - targets: ["localhost:9090"]
    metrics_path: /metrics
    scrape_interval: 15s
    scrape_timeout: 10s
    honor_labels: false
    relabel_configs:
      - source_labels: [__address__]
        target_label: job

6. Setting Scrape Timeout

Property	Values	Description
scrape_timeout	`10s`	Timeout for scraping targets. Default is 10 seconds.
scrape_timeout	`30s`	NEW in Prometheus v2.30+

Example: Configuring Scrape Timeout

scrape_configs:
  - job_name: "prometheus"
    scrape_timeout: 15s
    static_configs:
      - targets: ['localhost:9090']

7. Enabling Sample Limit

Option	Values	Notes
scrape.sample_limit	`0`	Maximum number of samples to scrape per job. 0 = no limit
scrape.max_samples_per_scrape	`100000`	Maximum number of samples to scrape per job. Default is 100k

Example: Configuring Sample Limit in Prometheus Configuration

scrape_configs:
  - job_name: "prometheus"
    static_configs:
      - targets: [localhost:9090]
    scrape_interval: 15s
    sample_limit: 10000

Note: Setting sample_limit helps prevent resource exhaustion from scraping too many samples. It's particularly useful when scraping metrics from services that may generate a large number of time series.

8. Configuring Label Limits

Option	Values	Notes
scrape_config.sample_limit	`1000`	Maximum number of samples per scrape
scrape_config.label_limit	`100`	Maximum number of labels per sample
scrape_config.label_name_length_limit	`1024`	Maximum length of label names in bytes
scrape_config.label_value_length_limit	`2048`	Maximum length of label values in bytes

Example: Setting Label Limits in Configuration

scrape_configs:
  - job_name: "prometheus"
    static_configs:
      - targets: ["localhost:9090"]
    label_limit: 150
    label_name_length_limit: 1200
    label_value_length_limit: 2500

Note: These limits help prevent resource exhaustion from overly large metrics. Adjust based on your system's capacity and typical metric sizes.

9. Using Params for Scraping

Parameter	Description	Example
params	Query parameters to append to the scrape URL	`params: format: prometheus extra: value`

Example: Configuring params in scrape job

scrape_configs:
  - job_name: "prometheus"
    static_configs:
      - targets: ["localhost:9090"]
    params:
      format: prometheus
      extra: value

10. Configuring Honor Labels

Option	Values	Description
honor_labels	`true` \| `false`	If true, Prometheus will use the label values from the scraped target instead of the ones provided by the scrape job. Default is `false`.

Example: Configuring honor_labels

scrape_configs:
  - job_name: "prometheus"
    honor_labels: true
    static_configs:
      - targets: ["localhost:9090"]

Note: When honor_labels is set to true, any labels present in the scraped metrics will override the ones defined in the scrape configuration.

11. Configuring Honor Timestamps

Option	Values	Description
honor_timestamps	`true` \| `false`	When set to `false`, Prometheus will ignore the timestamps provided by the target and use the scrape time instead. When set to `true`, Prometheus will honor the timestamps provided by the target.

Example: Setting honor_timestamps to false

scrape_configs:
  - job_name: "example"
    honor_timestamps: false
    static_configs:
      - targets: ["localhost:9090"]

Note: Setting honor_timestamps to false can be useful when scraping metrics from a source that provides inaccurate timestamps or when you want to align all metrics with the scrape time.

12. Setting Body Size Limit

Configuration Option	Description	Default Value	Notes
scrape.body-size-limit	Maximum size of the response body that Prometheus will accept from a scrape target.	`0`	Set to `0` to disable the limit. A value of `10MB` is recommended for large payloads.

Example: Configuring Body Size Limit

scrape_configs:
  - job_name: "example"
    scrape_interval: 15s
    static_configs:
      - targets: [ "localhost:9090" ]
    body_size_limit: 10MiB