Files
David Gallay 95f6a48b9d [REF] monitoring_prometheus: gather metrics on the fly in /metrics
The cron and the prometheus.metric table only existed to pass values from
the cron process to the HTTP workers. Querying the database directly when
/metrics is scraped removes the cron, the table and the staleness.
2026-10-01 11:55:24 +02:00
..

.. image:: https://img.shields.io/badge/licence-AGPL--3-blue.svg
    :alt: License

=========================================
Monitoring: Prometheus metrics queue jobs
=========================================

Expose the number of queue jobs to Prometheus through the */metrics* endpoint
of ``monitoring_prometheus``.

The gauge ``odoo_queue_job_count`` is published with the labels *state* and
*channel*, for the states ``pending``, ``enqueued``, ``started``, ``failed``
and ``wait_dependencies``. The terminal states ``done`` and ``cancelled`` are
not reported.

Values are counted at each scrape of the */metrics* endpoint.

This module is automatically installed when both ``monitoring_prometheus`` and
``queue_job`` are installed.