Commit Graph

8 Commits

Author SHA1 Message Date
Fusion
42087ddfeb feat(FN-4015): classify and recover from stale module-resolution incidents
Merges FN-4015 to add stale module-resolution incident detection and self-healing in the engine (`transient-error-detector.ts`, `self-healing.ts`) with mobile regression coverage for the shared `AgentErrorDetailsModal` layout.

Fusion-Task-Id: FN-4015
2026-05-11 12:04:20 -07:00
Fusion
631b384837 feat(FN-4005): harden durable transient recovery auto-restart flow
- Add bounded auto-restart handling in in-process runtime for durable agents after transient failures
- Extend self-healing recovery logic and transient error detection to classify and recover retryable runtime interruptions
- Add regression coverage across heartbeat executor, self-healing, and transient detector test suites
- Document durable agent transient recovery behavior in docs/agents.md

Fusion-Task-Id: FN-4005
2026-05-11 09:17:10 -07:00
gsxdsm
56597edc4f fix(engine): retry Codex WebSocket transport drops as transient
pi-ai's openai-codex-responses provider surfaces ChatGPT-plan WebSocket
drops as bare "WebSocket error" / "WebSocket closed <code>". The
underlying ErrorEvent.error is dropped by extractWebSocketError (it only
inspects event.message), so the cause is gone by the time we see it.
The engine then propagated these as fatal failures instead of retrying
a transient network blip.

- Add WebSocket patterns to TRANSIENT_ERROR_PATTERNS so the executor's
  existing transient retry path picks them up.
- Tag the model id onto the thrown error and emit a structured warn in
  promptSessionAndCheck so future drops can be triaged by transport.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-04 23:59:52 -07:00
gsxdsm
a59d058ad6 fix(FN-LOCAL): refresh docs and retry codex server errors 2026-04-22 00:43:08 -07:00
gsxdsm
4bd3dc5c26 Cleanup and fix transient error detection in review 2026-04-18 23:47:05 -07:00
gsxdsm
e5391a05a5 feat(FN-957): filter silent transient errors from executor and triage logs
- Add isSilentTransientError() function to detect noisy transient errors (ETIMEDOUT, ENOTFOUND, ECONNRESET, etc.)
- Filter silent transient errors from executor error logging to reduce noise
- Filter silent transient errors from triage error logging to reduce noise
- Add unit tests for isSilentTransientError covering all error patterns
2026-04-04 22:03:42 -07:00
gsxdsm
e8e77ec234 feat(FN-955): add 'request was aborted' to transient error patterns
- Add /request was aborted/i pattern to TRANSIENT_ERROR_PATTERNS for AI provider abort errors
- Cover case-insensitive matching (e.g., Anthropic streaming aborts)
- Add tests for positive matches, negative cases (bare 'abort'), and classifyError behavior
- Ensure abort errors classify as 'transient', not 'usage-limit'
2026-04-04 21:36:07 -07:00
gsxdsm
bff0c13d85 feat(KB-159): complete Step 4 — update frontend for streaming display 2026-03-30 08:55:07 -07:00