Operations

Operational reference for the Gordon v7 platform on srv-apps.

Dev loop

• Dev stack — first-run setup, overmind stack, meta-workspace patches, sccache/nextest
• E2E testing — full Docker stack validation, targeted-test loop, failure decision tree
• Make targets — grouped command reference for the workspace root Makefile
• Data backfill — one-shot bootstrap and ongoing backfill via gordon-data CLI

Testnet validation

• Testnet validation — Binance testnet setup, soak criteria, promotion checklist

Live operations

• Live trading — pre-live checklist, testnet-to-live progression, safety rules
• Monitoring — Prometheus, Grafana, Loki on srv-apps; alerts-coverage gate (DP-04)
• Bot warmup fallback — strict/degraded boot semantics for gordon-bot

Incident response

• Incident response — halt-latch, emergency flatten, bot quarantine, operator tokens
• Troubleshooting — service won't start, NATS lag, Postgres LISTEN, BuildKit OOM

Maintenance

• Schema restore — 7-step disaster recovery from restic snapshots on srv-core
• Rotate bot password — gordon_bot DB role password rotation procedure