V1 instrumentation: Prometheus metrics, alerting, log retention #7

New Issue

2026-09-01T21:35:57Z

Ty commented

2026-09-01 21:35:57 +00:00

Summary

Additive on top of the POC instrumentation. Required before a real customer-facing listing goes live.

Scope (V1 required, builds on POC issue)

/metrics endpoint — Prometheus format: request counts, latencies (p50/p95), U2Net duration, Woo call duration, error rates by endpoint.
Alerting path — cheap webhook to Hermes or a simple uptime check on the health endpoint; page on sustained 5xx or health failure.
Log retention policy — rotate and cap log volume so a small self-hosted disk doesn't fill; define retention window for sync_log and inference_log.

Non-goals

Distributed tracing, Sentry-style error tracking, dashboards — those are the nice-to-have tier.

Acceptance

Metrics scrapeable and showing real values during a UAT run.
Alert fires on a simulated health failure within one check interval.
Logs stay under a defined disk budget over a 30-day soak.

## Summary Additive on top of the POC instrumentation. Required before a real customer-facing listing goes live. ## Scope (V1 required, builds on POC issue) 1. **`/metrics` endpoint** — Prometheus format: request counts, latencies (p50/p95), U2Net duration, Woo call duration, error rates by endpoint. 2. **Alerting path** — cheap webhook to Hermes or a simple uptime check on the health endpoint; page on sustained 5xx or health failure. 3. **Log retention policy** — rotate and cap log volume so a small self-hosted disk doesn't fill; define retention window for `sync_log` and `inference_log`. ## Non-goals Distributed tracing, Sentry-style error tracking, dashboards — those are the nice-to-have tier. ## Acceptance - Metrics scrapeable and showing real values during a UAT run. - Alert fires on a simulated health failure within one check interval. - Logs stay under a defined disk budget over a 30-day soak.

Ty added this to the V1 milestone 2026-09-01 23:49:44 +00:00

grokbot_ty added the post-launch label 2026-09-24 18:56:21 +00:00

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: Ty_Tech/SlabUploader#7