Loading
Read the Value Streams Dashboard counts from GLQL
What does this MR do and why?
Moves the Merge requests and Pipelines count panels on the Value Streams Dashboard in Explore to GLQL. Both have an analytics source with a totalCount metric, so they're the first VSD panels that can move. The other panels wait on GLQL data gaps.
- The panels query
type = MergeRequestandtype = PipelinewithtotalCount. They have no date field, because the legacy counts aren't tied to a date range. - The dashboard now declares
data: [clickhouse], because the GLQL engines read ClickHouse. Without ClickHouse the whole dashboard is hidden, including the legacy panels. - The usage row is 3 grid rows taller, so the GLQL stat tiles fit their subtitle without scrolling. The panels below move down to match.
References
- GLQL panel conversion: https://gitlab.com/gitlab-org/gitlab/-/work_items/638158
- GLQL data gaps: https://gitlab.com/groups/gitlab-org/-/work_items/24216
- Epic: https://gitlab.com/groups/gitlab-org/-/work_items/24057
Screenshots or screen recordings
| Before | After |
|---|---|
![]() |
![]() |
Both read 380 merge requests and 1,275 pipelines for the same group.
How to set up and validate locally
ClickHouse and Siphon need to be running in GDK.
-
In the Rails console, enable the flag:
Feature.enable(:gi_migration_value_streams_dashboard) -
Open Explore > Analytics dashboards > Value Streams Dashboard and select a group.
-
The Merge requests and Pipelines counts match the legacy group dashboard's counts for the same group.
MR acceptance checklist
Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.
Edited by Ezekiel Kigbo



