Prometheus metrics endpoint

[filthyrake]

Summary

Add additional Prometheus metrics not covered by the initial implementation in PR #424.

Status

Partially implemented in PR #424 (Issue #414). Core metrics are in place at /metrics endpoint on Admin API (port 9001) and Worker API (port 9002).

✅ Already Implemented

• vlog_http_requests_total{method, endpoint, status_code} - Request count
• vlog_http_request_duration_seconds{method, endpoint} - Request latency histogram
• vlog_videos_total{status} - Total videos by status
• vlog_video_views_total - Total view count
• vlog_transcoding_jobs_total{status} - Jobs by status
• vlog_transcoding_job_duration_seconds - Transcoding duration histogram
• vlog_transcoding_queue_size - Pending jobs count
• vlog_workers_total{status} - Workers by status
• vlog_db_* - Database metrics
• vlog_redis_* - Redis metrics
• vlog_reencode_* - Re-encode queue metrics

✅ Configuration Options - COMPLETED

• VLOG_METRICS_ENABLED - Toggle metrics on/off (PR Security: Add optional authentication for metrics endpoint #436)
• VLOG_METRICS_AUTH_REQUIRED - Require authentication for /metrics (PR Security: Add optional authentication for metrics endpoint #436/Security: Add optional authentication for metrics endpoint (#436) #495)

---

Remaining Items

Metrics Not Yet Implemented

• vlog_http_requests_in_progress{endpoint} - Current active requests gauge
• vlog_videos_watch_time_seconds_total - Total watch time counter
• vlog_worker_jobs_completed_total{worker_id} - Jobs completed per worker
• vlog_worker_heartbeat_age_seconds{worker_id} - Time since last heartbeat
• vlog_storage_videos_bytes - Total video storage used gauge

Grafana Dashboard

• Create example Grafana dashboard JSON with panels for:
  • API performance (request rate, latency percentiles, error rate)
  • Transcoding status (queue depth, active jobs, duration)
  • Worker health (online count, heartbeat status)
  • Storage usage trends

Priority

Low - Core metrics are functional. These are nice-to-have enhancements.

---

Updated: Configuration options completed in PR #436/#495. Core implementation completed in PR #424 as part of Issue #414.

[filthyrake]

Core Prometheus metrics were implemented in PR #424 as part of Issue #414.

Updated this issue to track only the remaining nice-to-have items:

• Additional metrics (in-progress requests, watch time, per-worker stats)
• Configuration toggles
• Grafana dashboard

Moved to low-priority since the essential observability is now in place.

[filthyrake]

Updated issue to reflect current implementation status:

Configuration Options are now complete:

• VLOG_METRICS_ENABLED and VLOG_METRICS_AUTH_REQUIRED were implemented in PR Security: Add optional authentication for metrics endpoint #436 and PR Security: Add optional authentication for metrics endpoint (#436) #495

Remaining work:

• 5 additional metrics (nice-to-have)
• Example Grafana dashboard (nice-to-have)