TFChain WebSocket auto-reconnect and health endpoint #12

New issue

Closed

opened 2026-02-27 16:44:28 +00:00 by mik-tf · 0 comments

mik-tf commented

2026-02-27 16:44:28 +00:00

Member

The backend TFChain WebSocket connection dies permanently on disconnect (connection reset by peer, os error 54). Users see "restart required" error with no recovery until manual restart.

Requirements

Wrap TFChain client in RwLock for safe hot-swap on reconnect
Background health check every 30s, auto-reconnect on failure
GET /api/health endpoint returning chain status + latest block
Enhanced smoke tests for health, balance data, transfer prepare fields

Acceptance Criteria

Backend auto-recovers from WebSocket disconnect within 30s
/api/health returns 200 with latest_block when connected, 503 when disconnected
All 32 smoke tests pass including new data validation tests
All 56 cargo tests + 6 E2E tests still pass

The backend TFChain WebSocket connection dies permanently on disconnect (connection reset by peer, os error 54). Users see "restart required" error with no recovery until manual restart. ## Requirements - [ ] Wrap TFChain client in RwLock for safe hot-swap on reconnect - [ ] Background health check every 30s, auto-reconnect on failure - [ ] GET /api/health endpoint returning chain status + latest block - [ ] Enhanced smoke tests for health, balance data, transfer prepare fields ## Acceptance Criteria - Backend auto-recovers from WebSocket disconnect within 30s - /api/health returns 200 with latest_block when connected, 503 when disconnected - All 32 smoke tests pass including new data validation tests - All 56 cargo tests + 6 E2E tests still pass