penpot/backend/test/backend_tests/metrics_test.clj
Andrey Antukh d3ca9a0b83
✨ Add unified jobs substrate replacing legacy one-shot task system (#11542)
* 🎉 Move backend jobs to the unified job table substrate

Replace the generic 'task' table with a unified 'job' table as the
durable substrate for all backend jobs, and migrate every producer
and consumer to it. The legacy 'task' table stops receiving writes
but stays dormant (still cleaned by tasks-gc); upload-session
chunking and the other post-branch upstream changes are preserved.

New 0154 migration creates the 'job' table: dispatch and lifecycle
columns (name, queue, priority, scheduled_at, retries, status,
timestamps, props) plus optional user-facing ledger columns
(profile_id, target, progress, error, result, resource_id,
expires_at). No modified_at trigger: the application updates it on
every write for lease and orphan detection. Partial indexes cover
the dispatcher claim query, orphan scanning, and per-profile
listing.

New app.jobs public API: submit!/invoke!, heartbeats and progress
reporting, a precompiled job-def registry (name, schema, handler,
decoder, validator), an ephemeral request/response mode, and a
management API for external workers. The dispatcher claims
new/retry rows (FOR UPDATE SKIP LOCKED) with a plain JSON redis
payload and marks lease-expired running rows as orphaned; the
runner executes per-queue with the same retry conventions.

New jobs GC (daily): deletes expired rows and retains internal
terminal rows per :jobs-retention, touching referenced storage
objects so storage-gc-touched reclaims them. Job resources live in
a dedicated job-resource bucket with a reclaim branch in
storage-gc-touched.

All production call-sites (sendmail, delete-object, demo-purge,
file-gc, offload, webhooks, quotes, nitrate bulk delete,
snapshots, telemetry) and the ten cron tasks now submit through
app.jobs; cron is a pure scheduler and the legacy worker registry,
task-table inserts, and per-namespace handlers are removed.

Includes migrations, dispatcher, runner, jobs, request/response,
jobs GC, management API, and cron test suites.

AI-assisted-by: glm-5.3-flash
AI-assisted-by: muse-spark-1.3-contributor
AI-assisted-by: kimi-k3
AI-assisted-by: mimo-v2.5

* ✨ Address review comments on jobs substrate

Resolves the four open review threads on the jobs PR. Orphaned jobs
are now marked aborted, a system-side terminal state with no retry:
the dispatcher sweep stores the end event with outcome aborted in a
single bulk insert in the same transaction, records the terminal
outcome next to the orphaned counter, and reports an error log;
orphans keep alerting through the log and the counter, with no
user notification.

The aborted state is terminal everywhere: the status CHECK, the
sweep, the lifecycle writers, the jobs GC retention, the backlog
gauges and the user status mapping, with tests covering each
transition. The claim comment, the 2.20 doc version and the legacy
task table notes are corrected, and objects-gc heartbeats per chunk.

AI-assisted-by: muse-spark-1.3-contributor
2026-10-06 07:33:12 +02:00

139 lines
5.9 KiB
Clojure

;; This Source Code Form is subject to the terms of the Mozilla Public
;; License, v. 2.0. If a copy of the MPL was not distributed with this
;; file, You can obtain one at http://mozilla.org/MPL/2.0/.
;;
;; Copyright (c) KALEIDOS SUBSIDIARY SL
(ns backend-tests.metrics-test
(:require
[app.common.exceptions :as ex]
[app.metrics :as mtx]
[app.metrics.definition :as-alias mdef]
[clojure.test :as t]
[integrant.core :as ig])
(:import
io.prometheus.client.Collector$MetricFamilySamples
io.prometheus.client.Collector$MetricFamilySamples$Sample
io.prometheus.client.CollectorRegistry))
(def ^:private valid-definitions? @#'app.metrics/valid-definitions?)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Helpers
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
(defn- sample-names
[metrics]
(->> (mtx/get-registry metrics)
(.metricFamilySamples)
(enumeration-seq)
(mapcat (fn [^Collector$MetricFamilySamples family]
(map (fn [^Collector$MetricFamilySamples$Sample sample]
(.-name sample))
(.samples family))))
(set)))
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Tests
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
(t/deftest process-metrics-are-exported
;; the process cpu and file descriptor families come from the
;; prometheus client `StandardExports`, registered by `app.metrics`.
;; They are read reflectively from the OS MXBean and depend on the
;; `jdk.management` module at runtime: a pruned jlink JRE turns the
;; MXBean into `sun.management.BaseOperatingSystemImpl`, the reflective
;; getters fail and the families are silently dropped (that is how the
;; production backend lost `process_open_fds`). This test pins the
;; contract the fd alert relies on.
(let [metrics (ig/init-key :app.metrics/metrics {:default {}})
names (sample-names metrics)]
(t/is (contains? names "process_open_fds"))
(t/is (contains? names "process_max_fds"))
(t/is (contains? names "process_cpu_seconds_total"))))
(t/deftest label-coercion
(t/are [value fallback expected]
(= expected (mtx/label value fallback))
"fs" "unknown" "fs"
:s3 "unknown" "s3"
200 "unknown" "200"
nil "unknown" "unknown"
nil "default" "default"))
(t/deftest run-contains-label-arity-bug
;; A wrong label count throws inside the prometheus client; the
;; recording failure must not propagate to the caller.
(let [registry (CollectorRegistry.)
collector (mtx/create-collector {::mdef/name "penpot_test_run_arity"
::mdef/help "test helper"
::mdef/type :counter
::mdef/labels ["a" "b"]
:app.metrics/registry registry})
instance (reify mtx/IMetrics
(get-collector [_ _] collector))]
(t/is (nil? (mtx/run! instance :id :x :inc 1 :labels ["only-one"])))))
(t/deftest run-fails-loudly-without-metrics-instance
;; A missing or invalid instance is a wiring bug: it must fail on every
;; invocation instead of silently dropping the measurement. With
;; asserts enabled this is an AssertionError; with asserts disabled the
;; collector lookup (outside the recording guard) throws instead.
(t/is (thrown? Throwable (mtx/run! nil :id :x :inc 1)))
(t/is (thrown? Throwable (mtx/run! :not-metrics :id :x :inc 1))))
(t/deftest run-records-on-success
(let [registry (CollectorRegistry.)
collector (mtx/create-collector {::mdef/name "penpot_test_run"
::mdef/help "test helper"
::mdef/type :counter
::mdef/labels ["result"]
:app.metrics/registry registry})
instance (reify mtx/IMetrics
(get-collector [_ _] collector))]
(t/is (true? (mtx/run! instance :id :x :inc 1 :labels ["ok"])))))
(t/deftest instance-resolves-the-metrics-from-a-cfg
;; the resolver is a map read and nothing more: it hands back the very
;; instance, not a copy and not a wrapper
(let [metrics (ig/init-key :app.metrics/metrics {:default {}})]
(t/is (identical? metrics (mtx/instance {::mtx/metrics metrics})))))
(t/deftest instance-raises-on-a-cfg-without-metrics
;; A missing instance is a wiring bug, and the resolver is the one place
;; that says so: a module that resolves it at the top of a write sees the
;; failure before anything is persisted.
(let [cause (try
(mtx/instance {})
(catch Throwable e e))]
(t/is (some? cause))
(t/is (= :missing-metrics (:code (ex-data cause))))
(t/is (re-find #"::mtx/metrics" (:hint (ex-data cause))))))
(t/deftest definitions-schema-accepts-all-consumed-keys
(t/is (true? (valid-definitions?
{:test-timing
{::mdef/name "penpot_test_timing"
::mdef/help "test"
::mdef/type :histogram
::mdef/labels ["op"]
::mdef/buckets [5 10 25]}
:test-summary
{::mdef/name "penpot_test_summary"
::mdef/help "test"
::mdef/type :summary
::mdef/quantiles [[0.5 0.01]]
::mdef/max-age 60}}))))
(t/deftest definitions-schema-rejects-unknown-keys
;; A typo such as ::mdef/bucksets must fail at startup instead of
;; silently falling back to the default histogram buckets. The
;; definition map is closed, so unknown keys do not validate.
(t/is (false? (valid-definitions?
{:bad {::mdef/name "penpot_bad_metric"
::mdef/help "typo check"
::mdef/type :histogram
::mdef/bucksets [100]}}))))