[Backend] Implement Redis Pub/Sub for Distributed Multi-Instance SSE Broadcasting

[blurbeast]

Overview & Background

FlowFi delivers real-time stream status and balance updates to connected web clients via Server-Sent Events (SSE) implemented in backend/src/services/sse.service.ts.

Currently, SSEService stores client connections in an in-memory Set<Response>. In a production deployment with multiple backend instances behind a load balancer (e.g. Render, AWS ECS, or Kubernetes), an event processed by the indexer on Instance A is only broadcast to clients connected to Instance A. Clients connected to Instance B miss the live update.

---

Detailed Problem Statement

• Multi-container scaling breaks real-time updates for frontend users.
• Without a shared pub/sub message bus, horizontal scaling of backend API instances is impossible.

---

Technical Specification & Architecture

1. Redis Pub/Sub Architecture

[Soroban Indexer] 
       │
       ▼
[Publish: flowfi:stream-events] ──► (Redis Cluster)
                                         │
                 ┌───────────────────────┴───────────────────────┐
                 ▼                                               ▼
         [Backend Node A]                                [Backend Node B]
                 │                                               │
         [Local SSE Clients]                             [Local SSE Clients]

2. Implementation in backend/src/services/sse.service.ts

1. Import Redis publisher and subscriber clients from backend/src/lib/redis.ts.
2. When sseService.broadcast(event) is called:
   • Serialize event payload to JSON.
   • Publish to channel flowfi:events:broadcast.
3. In subscriber initialization:
   • Subscribe to flowfi:events:broadcast.
   • On incoming message, parse payload and dispatch to local clients Set.
4. Resilience & Fallback:
   • If Redis is disconnected or unconfigured in development (REDIS_ENABLED=false), gracefully fall back to local in-memory broadcasting.

---

Target Files

• backend/src/services/sse.service.ts
• backend/src/lib/redis.ts
• backend/src/controllers/sse.controller.ts
• backend/tests/unit/sse.service.test.ts

---

Acceptance Criteria

• Events published on Node A are received and sent to SSE clients connected to Node B via Redis Pub/Sub.
• Graceful fallback to local in-memory broadcast if Redis is offline.
• Heartbeat / keep-alive pings continue to maintain active connections every 15s.
• Integration tests verify multi-subscriber message delivery.

[badmusdare4]

@badmusdare4 has applied to work on this issue as part of the Stellar Wave Program's 8th wave.

Hello maintainer can I work on this issue

ℹ️ Repo Maintainers: To accept this application, review their application or assign @badmusdare4 to this issue.

[Tee-Gee57]

@Tee-Gee57 has applied to work on this issue as part of the Stellar Wave Program's 8th wave.

Hello Maintainer, I would like to work on this issue.

ℹ️ Repo Maintainers: To accept this application, review their application or assign @Tee-Gee57 to this issue.

[0darkside0rbit-star]

@0darkside0rbit-star has applied to work on this issue as part of the Stellar Wave Program's 8th wave.

Please let me work on this

ℹ️ Repo Maintainers: To accept this application, review their application or assign @0darkside0rbit-star to this issue.

[TheShnider]

@TheShnider has applied to work on this issue as part of the Stellar Wave Program's 8th wave.

Hi. I have reviewed this issue and would like to work on it.

ℹ️ Repo Maintainers: To accept this application, review their application or assign @TheShnider to this issue.

[felixumeh]

@felixumeh has applied to work on this issue as part of the Stellar Wave Program's 8th wave.

Hello I would like to work on this issue as it's att of my current area of study. I will deliver on PR as soon as possible

ℹ️ Repo Maintainers: To accept this application, review their application or assign @felixumeh to this issue.

[itzabdoull]

@itzabdoull has applied to work on this issue as part of the Stellar Wave Program's 8th wave.

Hello, I carefully analyzed the issue and identified the relevant sections of the codebase. I'd be happy to work on a clean, reliable implementation and submit a quality pull request.

ℹ️ Repo Maintainers: To accept this application, review their application or assign @itzabdoull to this issue.

[OmniZlatoon]

@OmniZlatoon has applied to work on this issue as part of the Stellar Wave Program's 8th wave.

Sir please assign me your issues. I am desperate to work but issues are scarce. my Eta is 5 hours max.

ℹ️ Repo Maintainers: To accept this application, review their application or assign @OmniZlatoon to this issue.

[benzy018]

@benzy018 has applied to work on this issue as part of the Stellar Wave Program's 8th wave.

Hi maintainer, i would love to work one this issue, i already reviewed the description and will submit a clean PR, please kindly assign

ℹ️ Repo Maintainers: To accept this application, review their application or assign @benzy018 to this issue.