阅读中文版本

Build a Distributed Job Scheduler

Distributed Job Scheduler — a complete walkthrough of scheduled job execution, covering job queue, time trigger, lock acquirer, and testing.

job_scheduler_distributed/
├── src/index.ts
├── src/job-queue.ts
├── src/time-trigger.ts
├── tests/job_scheduler_distributed.test.ts
└── package.json

// Job queue: scheduled job execution core
export function jobqueue(input: JobqueueInput): JobqueueResult {
  // Build the job queue that anchors scheduled job execution and defines the core data shape.
  return processJobqueue(input);
}

// Time trigger — handles scheduled job execution Add the time trigger so
export class Timetrigger {
  constructor(private opts: TimetriggerOptions) {}
  async run(ctx: Context): Promise<TimetriggerOutput> {
    return this.execute(ctx);
  }
}

// Lock acquirer: scheduled job execution stage
export const lockacquirer = (next: Next) => async (req: Request) => {
  const out = await transform(req);
  return next(out);
};

// Worker pool for scheduled job execution
export async function workerpool(stream: AsyncIterable<Chunk>): Promise<Aggregate> {
  let acc = init();
  for await (const c of stream) acc = merge(acc, c);
  return acc;
}

// Miss detector: scheduled job execution finalization
export function missdetector(state: State): Result {
  validate(state);
  return finalize(state);
}

Project overview

This project builds a scheduled job execution system whose reason for existing is running jobs at their scheduled time across a cluster so each job runs exactly once even during failover. The temptation in scheduled job execution is to bolt the interesting part onto a working prototype and discover the hard cases in production; this build inverts that order and treats a worker that claims a job, crashes, and leaves the lock held so the job never runs until the lock expires as the first thing designed, not the last.

You will implement Job queue, Time trigger, Lock acquirer, Worker pool, and Miss detector, each as a module with one responsibility and a test that proves it in isolation. The scheduled job execution pipeline is then the wiring that composes them, and the integration test is where the composition is shown to survive a worker that claims a job, crashes, and leaves the lock held so the job never runs until the lock expires.

The walkthrough is written for a developer who has used a scheduled job execution system but has not had to make one honest. It favors explicit data shapes over clever abstractions, and it names the failure each design choice prevents rather than presenting the choice as a style preference.

Each phase ends with something observable — a job queue that returns the right shape, a time trigger that rejects bad input, an integration test that goes green — so you can stop at any point and still have a scheduled job execution system that works for the cases covered so far.

What you will build

The finished scheduled job execution system exposes five capabilities, each with a focused test:

• Job queue
• Time trigger
• Lock acquirer
• Worker pool
• Miss detector

Job queue is the foundation that every later stage assumes; Time trigger is where scheduled job execution correctness lives; Lock acquirer handles a worker that claims a job, crashes, and leaves the lock held so the job never runs until the lock expires; Worker pool makes the system observable; and Miss detector is the graceful-degradation layer that keeps overload from becoming corruption.

The build keeps the scheduled job execution logic in ordinary modules you can read in one sitting, and pulls in dependencies only for the parts that are genuinely tedious to hand-roll. That keeps the diff between the tutorial and a real codebase small, and keeps the failure modes visible in code rather than hidden behind a framework.

Architecture and data model

The system is organized around Job queue, Time trigger, and Lock acquirer, with Worker pool tying them together. Data flows as a queue of jobs with a next-run time, claimed by a worker through a distributed lock so a failover does not double-run the job.

Each component is a module with a pure interface where practical: the same input produces the same output, and side effects are pushed to the edges. That discipline is what makes the phases independently testable, because a test for Lock acquirer does not need to stand up Job queue and Time trigger to exercise the interesting branch.

The data model is deliberately explicit. A queue of jobs with a next-run time, claimed by a worker through a distributed lock so a failover does not double-run the job Keeping the shape written down — in a schema, a type, or a fixture — means a change to the shape is a visible diff rather than a silent runtime shift, and it means the integration test can assert against a concrete shape instead of guessing.

Configuration and code are kept apart: the configuration says what scheduled job execution should do, the modules say how. A reviewer can change a threshold or a policy without reading the implementation, which keeps the risky changes small and lets the diff stand as the review.

The failure path through Lock acquirer is designed before the happy path is finished, because a worker that claims a job, crashes, and leaves the lock held so the job never runs until the lock expires is the case that distinguishes a scheduled job execution system from a demo. Every component declares what it does when its inputs are wrong, and the integration test asserts each of those declarations.

Prerequisites and dependencies

Have these ready before you install anything, so the scheduled job execution setup is not blocked by a missing tool:

• Node.js 20 LTS or later
• Familiarity with scheduled job execution fundamentals
• A local environment to run the test suite

The runtime dependencies are chosen to keep the scheduled job execution datapath legible:

• redis (^4.7.0) — Queue and locks
• fastify (^5.0.0) — Scheduler API

# setup for job-scheduler-distributed
pnpm init
pnpm add redis fastify
pnpm add -D typescript tsx vitest

Versions are pinned to a major range so the scheduled job execution install is reproducible; commit the lockfile. The job queue reads its operational parameters from the environment and validates them at startup, so a missing value fails the boot rather than the first scheduled job execution request.

Directory structure

Keep each scheduled job execution component in its own file so it can be tested without starting the others:

job_scheduler_distributed/
├── src/index.ts
├── src/job-queue.ts
├── src/time-trigger.ts
├── tests/job_scheduler_distributed.test.ts
└── package.json

Phase 1: Job queue

Build the job queue that anchors scheduled job execution and defines the core data shape.

// Job queue: scheduled job execution core
export function jobqueue(input: JobqueueInput): JobqueueResult {
  // Build the job queue that anchors scheduled job execution and defines the core data shape.
  return processJobqueue(input);
}

After this phase the job queue works in isolation and is covered by a unit test. The job queue defines the data shape that every later stage assumes, so getting it right first means the later phases compile against a stable contract.

Treat the job queue as the single source of truth for the scheduled job execution data shape. Validate inputs at the boundary and reject early, so a malformed value never reaches Time trigger or Lock acquirer where its effect would be harder to localize.

Phase 2: Time trigger

Add the time trigger so scheduled job execution can running jobs at their correctly.

// Time trigger — handles scheduled job execution Add the time trigger so
export class Timetrigger {
  constructor(private opts: TimetriggerOptions) {}
  async run(ctx: Context): Promise<TimetriggerOutput> {
    return this.execute(ctx);
  }
}

After this phase the time trigger works in isolation and is covered by a unit test. This is where scheduled job execution correctness lives: the time trigger is what makes running jobs at their scheduled time across a cluster so each job runs exactly once even during failover true, and its test is the proof.

Keep the time trigger pure where you can. A function that takes an input and returns an output is trivial to test and impossible to mock wrong; a class with hidden state is the opposite. Push side effects to the caller and keep the time trigger about the transformation.

Phase 3: Lock acquirer

Wire the lock acquirer to handle a worker that claims without losing data.

// Lock acquirer: scheduled job execution stage
export const lockacquirer = (next: Next) => async (req: Request) => {
  const out = await transform(req);
  return next(out);
};

After this phase the lock acquirer works in isolation and is covered by a unit test. The lock acquirer is the difference between a scheduled job execution system that works on the happy path and one that survives a worker that claims a job, crashes, and leaves the lock held so the job never runs until the lock expires.

Design the failure before the success. Decide what lock acquirer does when a worker that claims a job, crashes, and leaves the lock held so the job never runs until the lock expires: reject, retry, shed, or degrade — and write that test first. A scheduled job execution system whose failure behavior is unspecified will pick one at random under pressure, and it is usually the wrong one.

Phase 4: Worker pool

Integrate the worker pool to make scheduled job execution observable and testable end to end.

// Worker pool for scheduled job execution
export async function workerpool(stream: AsyncIterable<Chunk>): Promise<Aggregate> {
  let acc = init();
  for await (const c of stream) acc = merge(acc, c);
  return acc;
}

After this phase the worker pool works in isolation and is covered by a unit test. Observability is not a finishing touch; it is how you know the scheduled job execution system is doing what it claims in production.

Emit one structured event per meaningful action, with enough context to localize a failure without a second round-trip. The worker pool is what turns “something is slow” into “the time trigger is slow for inputs matching this shape,” which is the difference between an alert and an investigation.

Phase 5: Miss detector

Finalize the miss detector so the scheduled job execution system degrades gracefully under overload.

// Miss detector: scheduled job execution finalization
export function missdetector(state: State): Result {
  validate(state);
  return finalize(state);
}

After this phase the miss detector works in isolation and is covered by a unit test. The miss detector is what keeps overload from becoming corruption: when the scheduled job execution system is asked for more than it can do, it degrades in a defined way instead of failing in an undefined one.

Pick a degradation strategy for miss detector — shed, queue, or backpressure — and make it explicit in code and in a test. Leaving the behavior to chance is how a scheduled job execution traffic spike becomes a data loss incident.

Testing

Each phase has a focused test: job queue, time trigger, and lock acquirer are exercised in isolation, then an integration test drives the full scheduled job execution pipeline and asserts the a worker that claims case is handled.

Start with unit tests that pin each component’s contract: the job queue returns the documented shape, the time trigger transforms correctly, and the lock acquirer fails in the documented way. These tests are fast and they localize regressions to the component that broke.

Then add an integration test that composes the components and drives a realistic scheduled job execution scenario end to end. Assert the happy path first, then deliberately trigger a worker that claims a job, crashes, and leaves the lock held so the job never runs until the lock expires and assert the system degrades rather than corrupts. Because the components are pure where practical, the integration test stays deterministic rather than flaky.

A scheduled job execution suite must also prove the negative: bad input is rejected at the boundary, overload is shed rather than queued indefinitely, and a partial failure does not leave the system in an inconsistent state. These are the tests that catch the regressions a happy-path suite never would.

The entire scheduled job execution suite runs with no external service; every dependency is either in-process or stubbed, so every case is reproducible in CI. That is what lets the lock acquirer failure-mode assertions live as tests rather than as a manual checklist that drifts out of date.

# run the scheduled job execution test suite
pnpm vitest run

Deployment and operations

Run at least two scheduled job execution instances behind a load balancer and share state in Redis. Set resource limits so scheduled job execution overload surfaces as backpressure rather than failure.

Run at least two stateless instances behind a load balancer, and move any shared state into a shared store so the scheduled job execution behavior is global rather than per instance. Size limits so the system sheds load before it runs out of memory, because a scheduled job execution system that OOMs under load fails in a way that is hard to reason about.

Drain on deploy: stop accepting new work, let in-flight scheduled job execution operations finish, then exit. Because the system holds no state that cannot be reconstructed, a rollout is a matter of cycling instances one at a time while watching the error rate.

When a dependency is unavailable, decide explicitly whether the scheduled job execution system fails closed or degrades. Leaving that choice to chance is how a dependency outage becomes a silent data loss, so make the choice visible in code and in a test.

Security considerations

Validate all scheduled job execution inputs at the boundary, reject oversized payloads before buffering them, and keep secrets in the environment rather than the repository.

The scheduled job execution system is a choke point, so it is the right place to reject malformed input before it reaches time trigger or lock acquirer. Validate the shape and the size at the edge, and let the inner components assume what they receive is well-formed.

Never log secrets or full payloads; log enough to localize a failure without recording data that should not be on disk. Treat any external call as untrusted and validate its response before using it, because a scheduled job execution system that trusts its inputs blindly is one malformed response away from corruption.

Keep dependencies pinned and run an audit in CI. A transitive dependency with a known vulnerability is still a vulnerability in your scheduled job execution system, and the lockfile is what makes the audit meaningful.

Observability

Emit one structured event per scheduled job execution operation with the inputs’ shape, the outcome, and a correlation id. Track error rate and latency per component and alert on a step change rather than a single spike.

The correlation id is the thread that ties a user-visible symptom to the scheduled job execution decision and to the component log entry for the same operation. Without it, an investigation starts from a timestamp and a guess.

Watch the lock acquirer failure rate separately from the overall error rate. A jump in lock acquirer failures usually means a worker that claims a job, crashes, and leaves the lock held so the job never runs until the lock expires, not a bug in time trigger, and the two have different owners and different fixes.

Chart the share of scheduled job execution operations that shed, queue, or succeed. A change in the mix points at the cause faster than an aggregate latency can, because each class has a different root.

Troubleshooting

A worker that claims a job, crashes, and leaves the lock held so the job never runs until the lock expires. Reproduce with a minimal fixture, then add a guard test so the scheduled job execution system rejects the condition loudly.

Performance regresses under scheduled job execution load. Profile with the sampling analyzer, locate the hot path in time trigger, and add a targeted benchmark to lock the regression.

Keep the scheduled job execution dashboards separate from dependency dashboards. When an operation fails, the correlation id lets you decide in seconds whether the fault is job queue, time trigger, or lock acquirer, instead of guessing from a timeout.

Limitations and next steps

• Single-region; multi-region replication is out of scope.
• The job queue is in-process and not horizontally shared by default.
• No auth layer; assume deployment behind an edge that enforces identity.

The scheduled job execution system is single-region by design; extending to multi-region means making job queue shareable across regions, which is a real change to the consistency model rather than a config flag. The lock acquirer is in-process, which is correct for one instance but must move to a shared store before scaling out.

There is no authentication or authorization layer; the system assumes it is deployed behind an edge that enforces identity. Adding one is straightforward but out of scope for this build, and doing it here would distract from the scheduled job execution correctness that is the point.

The natural next step, once the scheduled job execution pipeline is trusted in production, is to add a short-TTL cache for idempotent reads and a richer worker pool that exports histograms rather than counters. Both are additive and neither changes the contracts this build establishes.