From e1475e42d672c5e6ed70bfd4b68c2f6d97f5d5b2 Mon Sep 17 00:00:00 2001 From: "mintlify[bot]" <109931778+mintlify[bot]@users.noreply.github.com> Date: Wed, 19 Aug 2026 22:47:15 +0000 Subject: [PATCH] docs(changelog): add entry for merge queue testing slot metrics --- ...08-19-merge-queue-testing-slot-metrics.mdx | 21 +++++++++++++++++++ changelog/index.mdx | 6 ++++++ 2 files changed, 27 insertions(+) create mode 100644 changelog/2026-08-19-merge-queue-testing-slot-metrics.mdx diff --git a/changelog/2026-08-19-merge-queue-testing-slot-metrics.mdx b/changelog/2026-08-19-merge-queue-testing-slot-metrics.mdx new file mode 100644 index 00000000..c1dfff0a --- /dev/null +++ b/changelog/2026-08-19-merge-queue-testing-slot-metrics.mdx @@ -0,0 +1,21 @@ +--- +title: "Merge Queue: New testing capacity metrics" +description: "The Merge Queue Prometheus endpoint now exposes separate gauges for active CI testing slots and PRs currently in the testing phase, making batching-aware capacity tracking straightforward." +slug: merge-queue-testing-slot-metrics +date: 2026-08-19 +category: Merge Queue +type: update +--- + +**Two new gauges on the Merge Queue Prometheus endpoint make it easy to track CI capacity separately from the number of PRs under test.** + +Previously, testing activity was reported as a single number, which conflated CI slot usage with the number of PRs in flight when batching was on. The endpoint now exposes: + +- **`mq_testing_slots_active`** — number of active CI slots currently in use. A batch counts as one slot, so this maps directly to CI concurrency. +- **`mq_prs_testing`** — number of PRs currently in the `TESTING` state. With batching enabled, this can exceed `mq_testing_slots_active` because several PRs share a slot. + +Use `mq_testing_slots_active` for CI utilization alerts against your configured concurrency, and `mq_prs_testing` when you want to see how many PRs are being validated at once. + +Both metrics carry the standard `repo`, `branch`, and `queue_type` labels and are available at `GET /v1/getMergeQueueMetrics` alongside the existing point-in-time gauges and 1-hour window metrics. + +See [Merge Queue metrics](/merge-queue/administration/metrics) for the full metric reference and example PromQL queries. diff --git a/changelog/index.mdx b/changelog/index.mdx index 62b126d8..e9a3b027 100644 --- a/changelog/index.mdx +++ b/changelog/index.mdx @@ -8,6 +8,12 @@ rss: true ### August 2026 + + **[Merge Queue: New testing capacity metrics](/changelog/2026-08-19-merge-queue-testing-slot-metrics)** + + The Prometheus metrics endpoint now exposes `mq_testing_slots_active` and `mq_prs_testing` as separate gauges, so batching-aware CI capacity tracking is straightforward. + + **[Flaky Tests: Asana ticketing integration](/changelog/2026-08-11-flaky-tests-asana-ticketing)**