Skip to content

Add app and runner metrics and operational observability - #20

Merged
j0sh merged 3 commits into
mainfrom
ja-codex/app-runner-observability
Oct 28, 2026
Merged

j0sh merged 3 commits into
mainfrom
ja-codex/app-runner-observability

Conversation

@j0sh

@j0sh j0sh commented Oct 28, 2026

Copy link
Copy Markdown
Collaborator

Operators need application and runner views of calls, failures, streaming traffic, payments, and redemption progress. This adds per-service Prometheus registries with registered app/runner ownership, live transport observations, chain/discovery metrics, and structured events with redacted dependency errors.

Paid orchestrator readiness now observes usable payment storage, fresh chain state, recipient activity, pricing, and redemption-worker completion. Migration 003 persists ticket enqueue times and backfills existing rows at migration time, so historical queue ages are lower bounds. The operational guide documents scrape endpoints, the metric/label contract, payment and traffic queries, and starting alert conditions.

Implements livepeer/naap-ops#52.

Validation:

  • go test -count=1 ./... — all 15 test-bearing packages passed.
  • go test -race ./... — all 15 packages passed, with no race reports.
  • go vet ./..., go build ./..., formatting, and diff checks passed.
  • Behavioral regressions cover paid/unpaid capacity and aliases, HTTP/SSE/WebSocket/Trickle forwarding and teardown, live stream progress and TCP half-closes, payment persistence/replay, durable redemption recovery/migration, discovery lifecycle, and first-pass worker readiness with both empty and failing queues.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant