Monitoring Metrics | Aviator Documentation

Aviator provides some monitoring metrics for Prometheus. You can monitor your queue length and GitHub API usage with this. Endpoint

https://app.aviator.co/api/metrics

serves Prometheus metrics. This is authed with your Aviator API key. In Prometheus config, you can add a scrape config like this: Copy

scrape_configs:
  - job_name: "aviator-mergequeue"
    metrics_path: "/api/metrics"
    scheme: "https"
    params:
      # REQUIRED: repository names to expose
      repos:
        - "octocat/Hello-World"
        - "octocat/Spoon-Knife"
      # OPTIONAL: target branch names of the PRs (the merge destination).
      # If it's not specified, grab all PRs.
      branches:
        - "main"
    authorization:
      type: "Bearer"
      credentials: "mq_live_... YOUR API KEY HERE"
    static_configs:
      - targets:
        - "app.aviator.co"

You need to update the credentials part and the repository list part. Metrics This endpoint exports following metrics. Name Type Labels Description mergequeue_queued_pr_count Gauge repo Number of queued pull requests mergequeue_reset_event_count Counter repo, branch, reset_type Number of reset events over time mergequeue_github_rest_api_quota_remaining_count Gauge Remaining quota for GitHub REST API mergequeue_github_rest_api_quota_limit_count Gauge Max quota limit for GitHub REST API mergequeue_github_graphql_api_quota_remaining_count Gauge Remaining quota for GitHub GraphQL API mergequeue_github_graphql_api_quota_limit_count Gauge Max quota limit for GitHub GraphQL API Note on metric names You will likely see metrics with some suffixes like

_total

and

_created

. This is Prometheus ecosystem's convention that we need to follow. See the OpenMetrics format spec for details. See Also How to Collect Prometheus Metrics in GCP Previous Webhooks Next Access Management Last updated 1 year ago Was this helpful?

The technical content and code provided seem mostly correct, but there are a few areas that could be clarified or improved:

1. Endpoint Authentication: The text mentions that the endpoint is "authed with your Aviator API key," but it doesn't specify how this authentication is handled. In the Prometheus configuration, the authorization section is used, but it's not clear if this is the only method or if additional headers are required when accessing the endpoint directly.

2. Prometheus Configuration:

   • The params section in the scrape_configs is not standard for Prometheus. Prometheus does not support passing query parameters like repos and branches directly in the scrape_configs. These parameters should be handled by the endpoint itself or through a different configuration mechanism.
   • The authorization section is correctly using a Bearer token, but ensure that the API key is securely managed and not hardcoded in configuration files.

3. Metrics Description:

   • The description of the metrics is clear, but it might be helpful to specify the units or expected values for each metric, especially for quotas (e.g., are they in requests per hour, per day, etc.?).

4. Metric Names:

   • The note on metric names mentions suffixes like _total and _created. While this is a good note, it might be helpful to explain that these suffixes are typically used for counters and gauges, respectively, to indicate cumulative totals or creation times.

5. Documentation Structure:

   • The "See Also" section references "How to Collect Prometheus Metrics in GCP," but no link or further information is provided. It would be helpful to include a hyperlink or additional context.
   • The "Last updated" section states "1 year ago," which might not be accurate or helpful without a specific date.

Overall, the content is mostly accurate, but addressing these points would improve clarity and usability.

No product messaging analysis available for this page.

No competitive analysis available for this page.