Skip to main content

Frequently Asked Questions

Common questions about RelayX, answered honestly.

Is RelayX a database?​

No. The KV store is designed for small, ephemeral state—things like session data, feature flags, or temporary coordination between services. It's not a replacement for a real database.

For persistent, queryable data, use a proper database (Postgres, MySQL, MongoDB, etc.). RelayX complements your database—it doesn't replace it.

Is it exactly-once delivery?​

No. RelayX provides at-least-once delivery. This means in rare cases (network issues, retries, crashes), your handler might receive the same message more than once.

What this means for you: Design your handlers to be idempotent. Use unique message IDs to deduplicate if needed. This is a common pattern in distributed systems and keeps RelayX fast and reliable.

What are the message size limits?​

Message size limits depend on your plan. Current limits are displayed in your RelayX Console.

As a general guideline, keep messages small—send references (IDs, URLs) rather than large payloads when possible. If you're hitting limits, that's often a sign to restructure your approach.

Can I use RelayX for multiplayer games?​

It depends on your game.

✅ Good fit:

  • Turn-based games (chess, card games)
  • Strategy games with slower update cycles
  • Real-time collaboration (100-200ms latency is fine)
  • Game lobbies, matchmaking, chat

❌ Probably not ideal:

  • Fast-paced FPS or fighting games needing <20ms latency
  • Physics-based multiplayer requiring tight synchronization

For twitch-reflex games, you'll likely need a dedicated game server with UDP networking. RelayX can still handle the non-realtime parts (lobbies, leaderboards, chat).

What happens if my subscriber crashes?​

It depends on what you're using:

Pub/Sub: Messages are delivered to connected subscribers. If your subscriber is down, it misses those messages. You can use message history to retrieve recent messages when reconnecting.

Queues: Messages are redelivered after a visibility timeout if not acknowledged. Your crashed subscriber won't lose the message—another instance (or the same one after restart) will pick it up.

Do you support webhooks?​

Not as a built-in feature. RelayX is designed for persistent connections, not HTTP callbacks.

However, you can easily bridge to webhooks yourself: set up a small subscriber service that listens to RelayX and forwards messages to your webhook endpoints. This gives you more control over retry logic, batching, and error handling anyway.

What about SOC2/HIPAA compliance?​

Coming soon. We're actively working toward SOC2 compliance.

If you have specific compliance requirements (HIPAA, SOC2, GDPR DPA, etc.), contact us. We're happy to discuss your needs and timeline.

How does pricing work?​

Pricing is based on messages per month. We have a generous free tier for development and small projects.

See relay-x.io/pricing for current plans and limits.

Can I self-host RelayX?​

No. RelayX is a managed service only. There's no self-hosted or on-premise option.

This lets us focus on reliability, performance, and features without the complexity of supporting diverse deployment environments. You get a service that just works.

Still have questions?​



Need Help?

Join our Discord server, post your concern & someone from our team will help you out ✌️