feat(clickhouse,webapp): keep 10-second resolution on the env metrics rollup

The env rollup's win comes from dropping the queue dimension, not from
coarser buckets: row count is queue-independent (~8640/day/env), so full
10-second granularity stays cheap at any range. Env header tiles and
saturation charts now resolve short-range detail exactly like the
per-queue charts, and the current-value tiles read the latest 10-second
bucket instead of a minute-wide one.

Author: ericallam

apps/webapp/app/v3/querySchemas.ts

@@ -809,16 +809,17 @@ export const queueMetricsSchema: TableSchema = {
 };
 
 /**
- * Schema definition for the env_metrics table (trigger_dev.env_metrics_1m_v1).
- * Environment-level rollup of queue_metrics into 1-minute buckets with the queue
- * dimension dropped, so header tiles and saturation charts cost the same regardless
- * of how many queues the environment has.
+ * Schema definition for the env_metrics table (trigger_dev.env_metrics_v1).
+ * Environment-level rollup of queue_metrics with the queue dimension dropped, so
+ * header tiles and saturation charts cost the same regardless of how many queues
+ * the environment has. Keeps the full 10-second granularity: row count is
+ * queue-independent, so even 30-day ranges stay small.
  */
 export const envMetricsSchema: TableSchema = {
   name: "env_metrics",
-  clickhouseName: "trigger_dev.env_metrics_1m_v1",
+  clickhouseName: "trigger_dev.env_metrics_v1",
   description:
-    "Environment-level concurrency, saturation, throttling, and scheduling-delay metrics (1-minute buckets)",
+    "Environment-level concurrency, saturation, throttling, and scheduling-delay metrics (10-second buckets)",
   timeConstraint: "bucket_start",
   tenantColumns: {
     organizationId: "organization_id",
@@ -846,7 +847,7 @@ export const envMetricsSchema: TableSchema = {
     bucket_start: {
       name: "bucket_start",
       ...column("DateTime", {
-        description: "The start of the 1-minute aggregation bucket",
+        description: "The start of the 10-second aggregation bucket",
         example: "2024-01-15 09:30:00",
         coreColumn: true,
       }),
@@ -907,6 +908,7 @@ export const envMetricsSchema: TableSchema = {
     },
   },
   timeBucketThresholds: [
+    { maxRangeSeconds: 3 * 60 * 60, interval: { value: 10, unit: "SECOND" } },
     { maxRangeSeconds: 12 * 60 * 60, interval: { value: 1, unit: "MINUTE" } },
     { maxRangeSeconds: 2 * 24 * 60 * 60, interval: { value: 5, unit: "MINUTE" } },
     { maxRangeSeconds: 7 * 24 * 60 * 60, interval: { value: 15, unit: "MINUTE" } },

internal-packages/clickhouse/schema/035_create_queue_metrics_v1.sql

@@ -92,11 +92,12 @@ SELECT
 FROM trigger_dev.queue_metrics_raw_v1
 GROUP BY organization_id, project_id, environment_id, queue_name, bucket_start;
 
--- (4) Env-level 1m rollup (no queue dimension) for header tiles/saturation charts.
+-- (4) Env-level 10s rollup (no queue dimension) for header tiles/saturation charts.
+-- Row count is queue-independent (~8640/day/env), so full granularity stays cheap at any range.
 -- No counter deltas on purpose: cross-queue deltaSumTimestamp state merges mix unrelated
 -- odometers (env totals must GROUP BY queue then sum). TDigest because an env-level
 -- reservoir absorbs every sample in the environment.
-CREATE TABLE IF NOT EXISTS trigger_dev.env_metrics_1m_v1
+CREATE TABLE IF NOT EXISTS trigger_dev.env_metrics_v1
 (
   organization_id  LowCardinality(String),
   project_id       LowCardinality(String),
@@ -119,11 +120,11 @@ TTL bucket_start + INTERVAL 30 DAY
 SETTINGS ttl_only_drop_parts = 1;
 
 -- (5) MV: raw -> env rollup.
-CREATE MATERIALIZED VIEW IF NOT EXISTS trigger_dev.env_metrics_1m_mv_v1
-TO trigger_dev.env_metrics_1m_v1 AS
+CREATE MATERIALIZED VIEW IF NOT EXISTS trigger_dev.env_metrics_mv_v1
+TO trigger_dev.env_metrics_v1 AS
 SELECT
   organization_id, project_id, environment_id,
-  toStartOfInterval(event_time, INTERVAL 1 MINUTE) AS bucket_start,
+  toStartOfInterval(event_time, INTERVAL 10 SECOND) AS bucket_start,
   max(env_queued)         AS max_env_queued,
   max(env_running)        AS max_env_running,
   max(env_limit)          AS max_env_limit,
@@ -197,8 +198,8 @@ GROUP BY organization_id, project_id, environment_id, queue_name, bucket_start;
 -- +goose Down
 DROP VIEW IF EXISTS trigger_dev.queue_metrics_5m_mv_v1;
 DROP TABLE IF EXISTS trigger_dev.queue_metrics_5m_v1;
-DROP VIEW IF EXISTS trigger_dev.env_metrics_1m_mv_v1;
-DROP TABLE IF EXISTS trigger_dev.env_metrics_1m_v1;
+DROP VIEW IF EXISTS trigger_dev.env_metrics_mv_v1;
+DROP TABLE IF EXISTS trigger_dev.env_metrics_v1;
 DROP VIEW IF EXISTS trigger_dev.queue_metrics_mv_v1;
 DROP TABLE IF EXISTS trigger_dev.queue_metrics_v1;
 DROP TABLE IF EXISTS trigger_dev.queue_metrics_raw_v1;

internal-packages/clickhouse/src/queueMetrics.test.ts

@@ -215,7 +215,7 @@ describe("queue_metrics_v1", () => {
   );
 
   clickhouseTest(
-    "5m and env rollups agree with the 10s tier, and cross-queue totals sum per queue",
+    "5m and env rollups agree with the 10s tier, and env buckets are 10s",
     async ({ clickhouseContainer }) => {
       const ch = new ClickHouse({ url: clickhouseContainer.getConnectionUrl(), name: "test" });
 
@@ -226,6 +226,14 @@ describe("queue_metrics_v1", () => {
         ...counter("started", "roll-b", 3, [500, 600, 700]),
         { ...base("gauge", "roll-a"), running: 4, queued: 9, env_running: 30, env_limit: 50 },
         { ...base("gauge", "roll-b"), running: 2, queued: 1, env_running: 45, env_limit: 50 },
+        {
+          ...base("gauge", "roll-a"),
+          event_time: "2026-06-30 12:00:15",
+          running: 1,
+          queued: 2,
+          env_running: 20,
+          env_limit: 50,
+        },
       ].map((row) => ({ ...row, organization_id: rollOrg }));
       const [insertError] = await ch.queueMetrics.insertRaw(rows, SYNC);
       expect(insertError).toBeNull();
@@ -269,19 +277,23 @@ describe("queue_metrics_v1", () => {
         query: `SELECT
             max(max_env_running) AS max_env_running,
             max(max_env_limit) AS max_env_limit,
+            uniqExact(bucket_start) AS buckets,
             round(quantilesTDigestMerge(0.5, 0.9, 0.95, 0.99)(wait_quantiles)[4]) AS wait_p99
-          FROM trigger_dev.env_metrics_1m_v1
+          FROM trigger_dev.env_metrics_v1
           WHERE organization_id = {org: String}`,
         schema: z.object({
           max_env_running: z.coerce.number(),
           max_env_limit: z.coerce.number(),
+          buckets: z.coerce.number(),
           wait_p99: z.coerce.number(),
         }),
         params: z.object({ org: z.string() }),
       })({ org: rollOrg });
       expect(envError).toBeNull();
       expect(envRows![0]!.max_env_running).toBe(45);
       expect(envRows![0]!.max_env_limit).toBe(50);
+      // 12:00:05 and 12:00:15 land in separate 10s env buckets (12:00:00 and 12:00:10).
+      expect(envRows![0]!.buckets).toBe(2);
       expect(envRows![0]!.wait_p99).toBeGreaterThanOrEqual(600);
       expect(envRows![0]!.wait_p99).toBeLessThanOrEqual(1000);