Feature Request: Global Configuration for Scrape Configs

[agolomoodysaada]

What did you do?
Tried to add azure_sd_configs in global block to reuse in multiple scrape jobs.

What did you expect to see?
It works! I don't have to rewrite all my configurations in every job block.

What did you see instead? Under which circumstances?
Unsupported feature

Environment

• System information:

  insert output of uname -srm here

• Prometheus version: 1.7

• Prometheus configuration file:

Current configuration file

global:
  scrape_interval: 5s

scrape_configs:
  - job_name: <service1>
    azure_sd_configs:
      - subscription_id: <...>
        client_id: <...>
        client_secret: <...>
        tenant_id: <...>
        refresh_interval: <...>
    relabel_configs:
      # ...common relabeling configs
      - source_labels: [__meta_azure_machine_name]
        target_label: instance
      - source_labels: [__meta_azure_machine_resource_group]
        target_label: resource_group
      - source_labels: [__meta_azure_machine_location]
        target_label: location
      # ...specific relabeling configs
      # ...
  - job_name: <service2>
    azure_sd_configs:
      - subscription_id: <...>
        client_id: <...>
        client_secret: <...>
        tenant_id: <...>
        refresh_interval: <...>
    relabel_configs:
      # ...common relabeling configs
      - source_labels: [__meta_azure_machine_name]
        target_label: instance
      - source_labels: [__meta_azure_machine_resource_group]
        target_label: resource_group
      - source_labels: [__meta_azure_machine_location]
        target_label: location
      # ...specific relabeling configs
      # ...
  - job_name: <service3>
    azure_sd_configs:
      - subscription_id: <...>
        client_id: <...>
        client_secret: <...>
        tenant_id: <...>
        refresh_interval: <...>
    relabel_configs:
      # ...common relabeling configs
      - source_labels: [__meta_azure_machine_name]
        target_label: instance
      - source_labels: [__meta_azure_machine_resource_group]
        target_label: resource_group
      - source_labels: [__meta_azure_machine_location]
        target_label: location
      # ...specific relabeling configs
      # ...

Desired configuration file

global:
  scrape_interval: 5s
  scrape_configs:
    - azure_sd_configs:
      - subscription_id: <...>
        client_id: <...>
        client_secret: <...>
        tenant_id: <...>
        refresh_interval: <...>
    - relabel_configs:
    # ...common relabeling configs

scrape_configs:
  - job_name: <service1>
    azure_sd_configs:
      - global.scrape_configs.azure_sd_configs
        # ...specific overrides
        refresh_interval: <...>
    relabel_configs:
      - global.scrape_configs.relabel_configs
        # ...specific overrides
      - source_labels: [__meta_azure_machine_name]
        regex:  'service1.*'
        action: keep
  - job_name: <service2>
    azure_sd_configs:
      - global.scrape_configs.azure_sd_configs
        # ...specific overrides
        refresh_interval: <...>
    relabel_configs:
      - global.scrape_configs.relabel_configs
        # ...specific overrides
      - source_labels: [__meta_azure_machine_name]
        regex:  'service2.*'
        action: keep
  - job_name: <service3>
    azure_sd_configs:
      - global.scrape_configs.azure_sd_configs
        # ...specific overrides
        refresh_interval: <...>
    relabel_configs:
      - global.scrape_configs.relabel_configs
        # ...specific overrides
      - source_labels: [__meta_azure_machine_name]
        regex:  'service3.*'
        action: keep

[brian-brazil]

This is something you should handle with configuration management, we do not wish to duplicate such functionality in Prometheus. The feature as-proposed would also be usable only in very very rare circumstances, even for Azure users. It's unusual not to have a mix of SDs.

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.