OpenCode probes local ports every 30s, spamming Django/FastAPI dev servers with 404s
philmehew · ghdev · 2026-10-22
A GitHub issue reports that OpenCode V2's built-in local runtime discovery probes fixed default endpoints for vLLM (127.0.0.1:8000), LM Studio (1234), and Ollama (11434) roughly every 30 seconds, with no documented way to disable it.
The reporter's Django app runs on port 8000, and logs show OpenCode continuously sending GET /health and GET /v1/models requests. The issues:
- 404 noise injected into app logs every 30 seconds;
- Django's DisallowedHost (400 + traceback) triggers when the loopback host isn't in ALLOWEDHOSTS;
- Probes carry B3/Traceparent headers sent to third-party services.
The proposal asks for a config switch (e.g. providers.vllm.discovery: false or a global localDiscovery: false), plus probing only when the endpoint's /v1/models response shape matches an expected runtime instead of blind fixed-interval polling. Since 8000 is also the default for FastAPI/uvicorn, most developers running local Python web services are affected.
More from coding & agent
- Claude Code Pushes Urgent Security Update Across All Installations — umeshonai · 2026-10-23
- Claude cut a 2.6-hour video export to 20 minutes — but only after being told to rebuild the server — byhewar · 2026-10-23
- 3 flight-booking rules my AI agent checks that I'd never do by hand — Efistoffeles · 2026-10-23
- OpenJet claims Qwen3.8-27B coding runs 1.9x faster than llama.cpp on one RTX 3090 — JeremyCMorgan · 2026-10-23
- Building an autonomous SEO agent in Hermes with three skills: programmatic pages, Keywords Everywhere, DataForSEO — EXM7777 · 2026-10-23
- Building a sports betting AI agent with onchain execution and human approval by default: architecture and lessons — Dry-Comfortable-2678 · 2026-10-23