Skip to content

fix: increase non-streaming request timeout 120s → 300s#230

Merged
nikazzio merged 1 commit into
mainfrom
fix/non-streaming-timeout
Jun 4, 2026
Merged

fix: increase non-streaming request timeout 120s → 300s#230
nikazzio merged 1 commit into
mainfrom
fix/non-streaming-timeout

Conversation

@nikazzio
Copy link
Copy Markdown
Owner

@nikazzio nikazzio commented Jun 4, 2026

Summary

  • Aumenta HTTP_REQUEST_TIMEOUT_SECS da 120s a 300s per i client HTTP non-streaming dei cloud provider
  • gpt-5.4 con reasoning attivo o documenti grandi (28K+ token) supera regolarmente i 2 minuti → timeout → retry loop infinito
  • Aggiunto invariante in ARCHITECTURE.md: Ollama usa streaming, cloud provider usano non-streaming — non cambiare senza motivo esplicito

Test plan

  • Ricompilare con npm run tauri:dev
  • Eseguire una pipeline su chunk grande con gpt-5.4
  • Verificare che non scatti il timeout a 2 minuti
  • Verificare che il retry non parta per timeout

🤖 Generated with Claude Code

gpt-5.4 with reasoning or large documents can exceed 120s.
Also document the Ollama streaming / cloud non-streaming invariant.
@nikazzio nikazzio merged commit 9a94057 into main Jun 4, 2026
6 checks passed
@nikazzio nikazzio deleted the fix/non-streaming-timeout branch June 4, 2026 20:21
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