bugsnag-exporter 1.2.0

Configure API token

You’ll need to provide an API token to access the Bugsnag Data Access API. See the Bugsnag documentation for details.

Start HTTP service

Start the HTTP server like this:

$ BUGSNAG_APITOKEN=MYTOKEN bugsnag_exporter --host=127.0.0.1 --port=9642

Pass --ttl=SECONDS to cache Bugsnag API results for the given time or -1 to disable (default is 600). Prometheus considers metrics stale after 300s, so that’s the highest scrape_interval one should use. However it’s usually unnecessary to hit the API that often, since the vulnerability alert information does not change that rapidly.

You can pass –buckets with a comma-separated list to define the upper bucket bounds that are used to generate the histogram metric.

Configure Prometheus

scrape_configs:
  - job_name: 'bugsnag'
    scrape_interval: 300s
    static_configs:
      - targets: ['localhost:9642']

We export one metric, a histogram called bugsnag_events, with labels {project="MyProject", release_stage="production"}.

A typical alerting PromQL query might be bugsnag_events_bucket{le="+Inf",release_stage="production"} - ignoring (le) bugsnag_events_bucket{le="1000.0",release_stage="production"} > 0

Additionally, a bugsnag_scrape_duration_seconds gauge is exported.

1.2.0 (2021-10-15)

• Add sort=unsorted parameter to handle potentially large result sets

1.1.1 (2021-05-07)

• Throttle paginated requests to avoid hitting the rate limit.

1.1.0 (2019-10-02)

• Use default prometheus histograms with “le” buckets

1.0.0 (2019-09-09)

• First release.