In order to finalize the statslib migration new dashboards using Prometheus as the data source need to be created. See steps from T350592: EPIC: migrate in use metrics and dashboards to statslib, most of the code has been migrated and merged, remaining steps are (6), (7) and (8):
- Identify the metric (or group of metrics) that will be to be converted.
- Create/assign a Phabricator subtask linked to this task (with granularity of individual metric or group of metrics) and update task description to reflect which task(s) have been created for which metric(s).
- Follow the migration process as outlined below.
- Secure/Conduct code review(s).
- Deploy the changes to production via the train (https://wikitech.wikimedia.org/wiki/Deployments/Train).
- Verify that the changes have been successfully implemented.
- Place the metrics subtask in a 2-3 week waiting period to allow prometheus time to establish 2-3 weeks of metric history
- After 2-3 week waiting period is complete, update the dashboard:
- Save a copy of the dashboard using legacy metrics as-is into the Legacy grafana dashboard folder
- Replace the old Graphite metric(s) with the new Prometheus metric(s) and save/update the live dashboard
Dashboards and Sections to migrate:
- Special:Homepage and Suggested Edits
- Page rendering (server-side)
- Page rendering (client-side) (except for that one transfer-size panel => T383563)
- Impact module
- Loading suggested task (client-side)
- Image recommendation service
- Errors
- Task pool
- Task suggester
- UserImpact API
- [WIP] Time to editor read for suggested edits tasks
- Growth team product KPI
- Clicks/Save/Reverts
- Questions posted to mentors or help desk
- Mentorship
- Account registration