Skip to content

Lobby and Matchmaking

Danil Vilmont edited this page Aug 5, 2026 · 3 revisions

Lobby and Matchmaking

Lobby routing has three modes:

  • velocity: preserve Velocity's native try and forced-host choice;
  • external: target one operator-managed backend already registered with Velocity; or
  • managed: own one persistent SLS-LITE blueprint instance.

SLS-Limbo is the lightweight fallback for players who otherwise have no safe backend. It is not a primary mode and is not the normal queue destination. In velocity mode, existing native or plugin redirects remain authoritative; SLS-LITE changes only a final disconnect into an SLS-Limbo redirect. Failed initial and forced-host connections are retried against their exact selected destination, and players rescued from a backend outage return when that backend is available. Deliberately selecting /server sls-limbo creates no fallback tracking record, so that player remains there until they choose another route. Its ping text, virtual player sample, brand, join message, boss bar, title, header/footer, and supported dimension are operator-configurable under lobby.limbo.presentation; each visual surface can be disabled independently. Text uses bounded MiniMessage and is serialized safely. SLS-LITE still owns the bind address, port, forwarding, protocol, capacity, traffic, and lifecycle settings. Players who request a cold game from a healthy backend stay there while the destination starts, then transfer directly when it becomes ready.

Matchmaking prefers a ready compatible instance with capacity, accounts for queued slots, and creates a new instance only within blueprint and host limits. Queue entries terminate on success, timeout, cancellation, disconnect, startup failure, instance failure, or shutdown.

Canonical references: SLS-Limbo, Configuration, and Operations.

Clone this wiki locally