openvibers/openvibe.community Commits

Community: request telemetry in its own table, 30-day retention (#84)

OP OpenVibers committed 29 days ago · 1ff43592dd · 4 files, +86 −0

Plan T1 step 10 (universal telemetry): this service now records its request telemetry using openvibe-sdk 0.46's process-wide collector, the same shape Network uses.

- **Migration** `NNNN_telemetry.sql` (expand): the `telemetry_samples` table from `telemetrySchema()`.
- **`server/app.js`** mounts `telemetryMiddleware` before any route. It does nothing until `init()`, so tests and tools that build the app are unaffected.
- **`server/index.js`** (production start):
  - `telemetry.init({ service, sink: createPgTelemetrySink(...) })` aggregates requests per route template, method and status class. Every 15 s (the singleton's own timer) it writes one `http.request` sample per key plus the active-request, p95 and event-loop-lag gauges. `/shared`, health, readiness and `/metrics` produce no samples.
  - `startTelemetryRetention` deletes rows older than 30 days, a minute after start and then daily.
  - On stop, the retention stops and the collector flushes once before the database closes.
- **`test/telemetry.test.js`**:
  - checks that wiring in the source;
  - checks that five hits on one route flush as one sample (count 5) and that `/metrics` and `/shared` produce none;
  - checks that the migration's table works with the retention (only rows from the last 30 days are kept).

Changes

ddiff --git a/migrations/0009_telemetry.sql b/migrations/0009_telemetry.sql
nnew file mode 100644
iindex 0000000..14260fa
---- /dev/null
++++ b/migrations/0009_telemetry.sql
@@@ -0,0 +1,28 @@
++-- phase: expand
++-- Request telemetry (openvibe-sdk/telemetry telemetrySchema()): one row per aggregated sample, pruned after 30 days (server/index.js).
++CREATE TABLE IF NOT EXISTS telemetry_samples (
++ id bigint GENERATED ALWAYS AS IDENTITY PRIMARY KEY,
++ at timestamptz NOT NULL,
++ service text NOT NULL,
++ project text,
++ subject text,
++ resource text,
++ provider text,
++ node text,
++ cell text,
++ region text,
++ operation text NOT NULL,
++ latency_ms double precision,
++ queue_delay_ms double precision,
++ ttfb_ms double precision,
++ throughput_per_second double precision,
++ bytes bigint,
++ status text,
++ cache_status text,
++ cost_estimate double precision,
++ route_epoch bigint,
++ trace_id text,
++ extra jsonb NOT NULL DEFAULT '{}'::jsonb
++);
++CREATE INDEX IF NOT EXISTS telemetry_samples_at_idx ON telemetry_samples (at);
++CREATE INDEX IF NOT EXISTS telemetry_samples_service_idx ON telemetry_samples (service, operation, at);
ddiff --git a/server/app.js b/server/app.js
iindex f6d0dc1..7e926cf 100644
---- a/server/app.js
++++ b/server/app.js
@@@ -84,6 +84,7 @@ async function createApp(opts = {}) {
// HTTP golden signals by route template, process metrics, release_info; GET /metrics answers
// direct loopback callers only (Track O). Request metrics only: no content counts.
const metrics = require('openvibe-shared/metrics').instrument(app, { service: 'community', release: release.release });
++ app.use(require('openvibe-sdk/service').telemetryMiddleware); // request telemetry: a no-op until server/index.js calls telemetry.init()
app.locals.metrics = metrics.registry;
app.use(helmet({
ddiff --git a/server/index.js b/server/index.js
iindex 6afc55b..29671b6 100644
---- a/server/index.js
++++ b/server/index.js
@@@ -6,6 +6,8 @@
*/
const config = require('./config');
const { createApp } = require('./app');
++const { telemetry } = require('openvibe-sdk/service');
++const { createPgTelemetrySink, startTelemetryRetention } = require('openvibe-sdk/telemetry');
(async () => {
// PostgreSQL first (migrations run as the owner), then the app: its revocation cutoffs load before it serves.
@@@ -28,6 +30,10 @@ const { createApp } = require('./app');
console.log(`[Community] pastes: this site is the authority (PostgreSQL), identity via ${config.networkUrl}`);
});
server.keepAliveTimeout = 65_000;
++ // Request telemetry (openvibe-sdk/service, plan T1): one aggregated sample per route template, method and status
++ // class each 15 s, stored in this service's telemetry_samples table; rows older than 30 days go daily.
++ telemetry.init({ service: 'community', sink: createPgTelemetrySink({ db: require('./db').getDb(), log: console }), skipPrefixes: ['/shared'] });
++ const telemetryRetention = startTelemetryRetention({ db: require('./db').getDb(), log: console });
// Subscribe Pulse to public activity at Events (idempotent; off without EVENTS_URL / COMMUNITY_EVENTS_SECRET).
let subscriptions = null;
try {
@@@ -62,6 +68,7 @@ const { createApp } = require('./app');
() => within(1000, profilesDone),
() => within(1000, app.locals.relay && app.locals.relay.stop()),
() => within(1500, require('./events').stop()),
++ async () => { if (telemetryRetention) await telemetryRetention.stop(); await telemetry.stop(); },
async () => {
const ev = await require('./events').backlog().catch(() => require('./events').status());
let relay = 'off';
ddiff --git a/test/telemetry.test.js b/test/telemetry.test.js
nnew file mode 100644
iindex 0000000..bea09ce
---- /dev/null
++++ b/test/telemetry.test.js
@@@ -0,0 +1,50 @@
++'use strict';
++/**
++ * Request telemetry (openvibe-sdk/service, plan T1 step 10): app.js mounts the process-wide telemetryMiddleware (a no-op
++ * until init), server/index.js calls telemetry.init() for service 'community' with a PostgreSQL sink, starts the 30-day
++ * retention and flushes on stop. Five hits on one route flush as ONE aggregated http.request sample; /metrics and
++ * /shared make none; the retention deletes only rows older than 30 days.
++ */
++const assert = require('assert');
++const fs = require('fs');
++const path = require('path');
++const { telemetry } = require('openvibe-sdk/service');
++const { pruneTelemetry } = require('openvibe-sdk/telemetry');
++const { boot } = require('./helpers/app');
++(async () => {
++ const read = (p) => fs.readFileSync(path.join(__dirname, '..', p), 'utf8');
++ assert.match(read('server/index.js'), /telemetry\.init\(\{ service: 'community', sink: createPgTelemetrySink\(/, 'index.js starts the collector');
++ assert.match(read('server/index.js'), /startTelemetryRetention\(/, 'index.js starts the retention');
++ assert.match(read('server/index.js'), /[Rr]etention\.stop\(\); await telemetry\.stop\(\);/, 'the stop flushes');
++
++ const t = await boot();
++ const samples = [];
++ telemetry.init({ service: 'community', sink: async (batch) => { samples.push(...batch); }, intervalMs: 3600e3 });
++ const requests = () => samples.filter((s) => s.operation === 'http.request');
++ try {
++ for (let i = 0; i < 5; i++) assert.strictEqual((await t.get('/')).status, 200);
++ await telemetry.flush();
++ const hit = requests().filter((s) => s.resource === '/' && s.status === '2xx');
++ assert.strictEqual(hit.length, 1, JSON.stringify(requests()));
++ assert.strictEqual(hit[0].service, 'community');
++ assert.strictEqual(hit[0].extra.count, 5);
++
++ samples.length = 0;
++ await await t.get('/metrics');
++ await await t.get('/shared/navbar.js');
++ await telemetry.flush();
++ assert.deepStrictEqual(requests(), [], '/metrics and /shared make no sample');
++
++ const db = t.db;
++ await db.query('DELETE FROM telemetry_samples');
++ const insert = (days) => db.query("INSERT INTO telemetry_samples (at, service, operation) VALUES (now() - make_interval(days => $1), 'community', 'http.request')", [days]);
++ await insert(45);
++ await insert(1);
++ assert.strictEqual(await pruneTelemetry({ db }), 1);
++ assert.strictEqual((await db.query('SELECT count(*)::int AS n FROM telemetry_samples')).rows[0].n, 1);
++ console.log('telemetry ok');
++ } finally {
++ await telemetry.stop();
++ await t.close();
++ }
++})().catch((err) => { console.error(err); process.exit(1); });