Feature/ Grit-metre

[nanda-kshr]

No description provided.

[Copilot]

Pull request overview

This PR modernizes the scheduler service by migrating the legacy scheduler/* module into a modules/jobs structure, adding a new “Grit Meter” feature (scheduled + manual trigger), and introducing production-focused deployment/runtime configuration (MySQL, Docker, CI deploy, API-key auth).

Changes:

• Replace the old Scheduler module with a Jobs module (controller/service/entity/runner) and add periodic DB sync to the runner.
• Add Grit Meter module (cron-driven daily processing + manual endpoint) and enforce global API-key authentication.
• Add production deployment assets (Dockerfile, docker-compose, GitHub Actions deploy) and update TypeScript config (CommonJS + path aliases).

Reviewed changes

Copilot reviewed 19 out of 23 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
tsconfig.json	Switches TS output to CommonJS, lowers target, adds path aliases.
src/scheduler/scheduler.module.ts	Removes legacy Scheduler module.
src/scheduler/scheduler.controller.spec.ts	Removes legacy e2e tests for Scheduler controller.
src/modules/jobs/jobs.service.ts	Renames service and keeps job persistence + runner scheduling.
src/modules/jobs/jobs.module.ts	Introduces Jobs module wiring (controller/service/runner/entity).
src/modules/jobs/jobs.controller.ts	Renames controller and routes jobs CRUD endpoints.
src/modules/jobs/job-runner.service.ts	Adds DB sync loop + scheduling adjustments + retry/backoff logic changes.
src/modules/jobs/entities/job.entity.ts	Adds MySQL-oriented jobs entity schema.
src/modules/jobs/dto/create-job.dto.ts	Simplifies DTO validation and removes scheduling subtypes.
src/modules/grit-meter/grit-meter.service.ts	Adds daily grit-meter processing cron + DB updates + Discord webhook.
src/modules/grit-meter/grit-meter.module.ts	Registers grit-meter controller/service.
src/modules/grit-meter/grit-meter.controller.ts	Adds manual trigger endpoint for grit-meter processing.
src/common/guards/api-key.guard.ts	Adds API key guard for Bearer token auth.
src/app.module.ts	Switches DB config to MySQL via ConfigModule and applies global guard.
README.md	Expands API documentation for job endpoints.
package.json	Adds @nestjs/config and mysql2 dependencies.
package-lock.json	Locks new dependency tree for config + mysql driver.
Dockerfile	Adds multi-stage Docker build for production image.
docker-compose.yml	Adds compose service for running the scheduler.
.github/workflows/main.yml	Adds production deploy workflow via SSH + docker-compose.
.env.example	Adds example env configuration for MySQL + API key + Discord webhook.
.dockerignore	Adds docker ignore rules for build context.

Suppressed comments (4)

src/modules/jobs/job-runner.service.ts:201

• When a job fails and a retry is scheduled, job.status is never reset from 'running'. The next retry will call execute(), hit if (job.status === 'running') return;, and silently stop all further retries (job remains stuck in running). Set the status back to 'scheduled' before persisting/scheduling retries so retries can run.
  src/modules/jobs/job-runner.service.ts:23
• JOB_SYNC_INTERVAL_MS is parsed with parseInt(...) and used directly in setInterval. If the env var is unset/empty/invalid, parseInt can return NaN, and Node will treat the interval as 0, potentially causing a tight loop and hammering the DB/logs. Validate the parsed value and fall back to a safe default when it’s not a positive finite number.
  src/modules/jobs/job-runner.service.ts:266
• syncWithDb() will reschedule one-off jobs whenever !this.timers.has(job.id). Retry backoff timers are stored under keys like ${id}:retry:${attempt}, so a job that’s waiting on a retry will be treated as “not scheduled” and may get re-scheduled immediately (bypassing backoff) or run twice. Consider also checking for pending retry timers before calling scheduleOneOff().
  src/modules/jobs/jobs.service.ts:36
• Jobs are persisted with status: 'scheduled' even when scheduling is missing required fields (e.g., execute_at for one-off or cron for recurring). The runner then skips scheduling and the job stays permanently "scheduled" but never runs. Consider validating the scheduling payload and returning a 400 before saving/scheduling.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.