-
Bug
-
Resolution: Unresolved
-
Minor
-
None
-
CL170 - RHOSO 18.X NEXT
-
False
-
-
False
-
ROLE
-
-
-
en-US (English)
Please fill in the following information:
URL: | https://rol.redhat.com/rol/app/courses/cl170ea-18.0/pages/ch03s03 |
Reporter RHNID: | yuvaraj-rhls |
Section Title: | h2. Configuring OpenStack Services |
User Feedback Description:
Description: Maybe we should mention the new ceilomenter/Prometheus integration?
-
- Ceilometer and Prometheus Integration
1. Ceilometer uses TCP publisher to expose data for Prometheus:
- Ceilometer can now use the TCP publisher to publish metric data to sg-core, which exposes them for scraping by Prometheus[1].
2. Prometheus replaces Gnocchi:
- In RHOSO 18.0, Prometheus replaces Gnocchi for metrics storage and metrics-based autoscaling[1].
3. Telemetry Operator support:
- The Telemetry Operator now supports deploying and operating Prometheus using the `cluster-observability-operator` through a MonitoringStack resource[1].
-
- Configuration and Deployment
- Ceilometer is enabled by default in RHOSO deployments[2].
- A Prometheus metrics exporter is created and exposed from inside the cluster at http://ceilometer-internal.openstack.svc:3000/metrics[2].
- Administrators can configure observability settings in the `openstack_control_plane.yaml` CR file, including:
- Scrape interval for metrics collection
- Storage retention period
- Storage strategy (persistent or ephemeral)
- PVC storage request size for persistent storage[2]
-
- Additional Features
- RHOSO provides telemetry dashboards accessible through the RHOCP console under "Observe" and then "Dashboards"[2].
- The OpenStack CLI now supports `openstack metric` and `openstack alarm` commands for interacting with metrics and alarms, which can be useful for troubleshooting[1].
This integration allows RHOSO to leverage Prometheus for more efficient metrics storage and querying while still utilizing Ceilometer for metric collection, providing a robust telemetry solution for OpenStack deployments on OpenShift.
Citations:
[1] https://docs.redhat.com/en/documentation/red_hat_openstack_services_on_openshift/18.0/html/release_notes/chap-release-info_release-notes
[2] https://docs.redhat.com/en/documentation/red_hat_openstack_services_on_openshift/18.0/html/customizing_the_red_hat_openstack_services_on_openshift_deployment/rhoso-observability_custom_dataplane
[3] https://opendev.org/openstack/kolla-ansible/commit/6cf03122ee6caf51546f673c2899a1fff6fe9f12
[4] https://docs.openstack.org/ceilometer/latest/admin/telemetry-data-pipelines.html
[5] https://community.grafana.com/t/whats-the-point-of-grafana-cloud-prometheus/42061
[6] https://stackoverflow.com/questions/52146798/collect-metrics-from-the-openstack-environment-and-show-it-in-grafana
[7] https://www.reddit.com/r/kubernetes/comments/p2in04/beginner_using_prometheus_what_are_the_proscons/
[8] https://www.reddit.com/r/kubernetes/comments/1351kk7/any_reasons_to_use_prometheus_over_victoriametrics/
Expected result: