a14560c7c9
Long research turns failed mid-task with "Lost connection to the AI provider".
Node's global fetch (undici) defaults BOTH headersTimeout and bodyTimeout to
300_000ms, and the chat provider + the external-MCP dispatcher both ran on it
with no override, so:
- the z.ai chat stream dropped when a late step's huge accumulated context
pushed the model's time-to-first-token past 5 min (the model reasons
server-side with NO streamed reasoning, so the connection is silent until the
first answer token — reproduced: even a trivial glm-5.2 query has a ~4-8s
first-chunk gap; a long run reaches 400k+-token steps), or a reasoning model
paused >5 min between chunks (bodyTimeout);
- the crawl4ai SSE transport, held open across the whole turn, dropped when it
idled >5 min between tool calls.
Fix: a dedicated undici dispatcher whose stream timeouts are raised to a
generous-but-FINITE silence timeout (default 15 min, AI_STREAM_TIMEOUT_MS) on
each path. NOT disabled (0): that would let a genuinely hung provider — with the
client still connected — hang forever, since the turn's abortSignal only fires on
client disconnect. The timeout bounds SILENCE (time-to-first-byte and the gap
BETWEEN chunks), NOT total turn duration, so an arbitrarily long turn that keeps
streaming is never cut; only a stream quiet for >15 min is treated as a hang.
- ai-streaming-fetch.ts: createStreamingFetch() + streamTimeoutMs() /
streamingDispatcherOptions() (the shared, configurable timeout).
- ai.service: the chat provider fetch is createStreamingFetch(), wrapped by the
existing passive ECONNRESET telemetry (createDiagnosticFetch gained an
optional baseFetch) so the telemetry observes the SAME transport.
- mcp-clients: the SSRF-pinned Agent uses streamingDispatcherOptions().
Investigation: reproduced the transport mechanism against the real z.ai endpoint
(a 1ms headersTimeout throws UND_ERR_HEADERS_TIMEOUT — the exact drop) and ran
the actual research agent to a ~428k-token context. Verified the fixed path
streams cleanly live (glm-5.2 turns finish; telemetry confirms the streaming
fetch is in use).
Tests: ai-streaming-fetch.spec (default 15m + env override + invalid fallback +
both-timeouts + streams a delayed response); ai-http-diagnostics + ai/mcp specs
green. server tsc clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A progressive Node.js framework for building efficient and scalable server-side applications.
Description
Nest framework TypeScript starter repository.
Installation
$ npm install
Running the app
# development
$ npm run start
# watch mode
$ npm run start:dev
# production mode
$ npm run start:prod
Migrations
# This creates a new empty migration file named 'init'
$ npm run migration:create --name=init
# Generates 'init' migration file from existing entities to update the database schema
$ npm run migration:generate --name=init
# Runs all pending migrations to update the database schema
$ npm run migration:run
# Reverts the last executed migration
$ npm run migration:revert
# Reverts all migrations
$ npm run migration:revert
# Shows the list of executed and pending migrations
$ npm run migration:show
## Test
```bash
# unit tests
$ npm run test
# e2e tests
$ npm run test:e2e
# test coverage
$ npm run test:cov
Support
Nest is an MIT-licensed open source project. It can grow thanks to the sponsors and support by the amazing backers. If you'd like to join them, please read more here.
Stay in touch
- Author - Kamil Myśliwiec
- Website - https://nestjs.com
- Twitter - @nestframework
License
Nest is MIT licensed.