You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
**Stacked on #4131** — targets `feat/queue-metrics-and-health`, not
`main`. GitHub will auto-retarget this to `main` once #4131 merges.
(Targeting `main` now would wrongly include all of #4131's own commits.)
UI polish for the Queue metrics & health pages (TRI-12068). 49 commits,
all under `apps/webapp/app/` — no engine/ClickHouse/package changes.
**Highlights**
- **Layout** — consolidated queue-detail header into one filter bar;
shared `MetricsLayout` spacing (2.5 scale); paused-queue banner
mirroring the environment-paused banner.
- **Charts** — inline legends under titles; threshold gradient coloring
(Env saturation orange only >100%; Concurrency orange only at/over the
limit); fullscreen legend spacing; result caching so tab switches don't
reload; brighter hover cursor + full x-axis labels.
- **Tables** — whole-row click via stretched link; custom Queues icon;
removed column sorting; Limit-cell spacing + hover-bright; "Limited by"
tooltip sizing.
- **Controls** — Pause moved into the filter bar (orange label +
accurate tooltip copy); restyled override-concurrency modal.
- **Copy** — plain-language pass; inline colour swatches; Throttled
reads "% of current period".
Also consolidated the three queue-metrics `.server-changes` notes into
one.
**Known follow-up (not in this PR):** the Concurrency-keys table is
capped at 50 keys (union of live Redis backlog + ClickHouse history, no
single cursor across both) and truncates silently — fine for most
queues, a real limit for high-cardinality per-tenant keys.
Verified locally against the `queue-metrics-demo` seed; `pnpm typecheck
--filter webapp` passes.
The concurrency keys table on a queue's page is now paginated, so queues with thousands of keys can page through all of them instead of only showing the top 50.
Copy file name to clipboardExpand all lines: .server-changes/queue-metrics-dashboard.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -3,4 +3,4 @@ area: webapp
3
3
type: feature
4
4
---
5
5
6
-
Queue metrics and health on the Queues page: per-queue depth, throughput, concurrency, throttling, and scheduling-delay charts, plus a per-queue detail view, live queue stats and a backlog chart on the task detail page, and a waiting-in-queue explainer in the run inspector for runs that have not started yet. Off by default; enabled per organization.
6
+
New Queue metrics & health dashboard (per-org opt-in): per-queue depth, throughput, concurrency, throttling and scheduling-delay charts, a per-queue detail view, and live queue stats. Queue concurrency limits set above the environment limit are now rejected instead of being silently capped.
0 commit comments