Traducido con IA. Aún no revisado por la comunidad. Leer en inglés

Across the APIs: Resilience in Integration,Platform Events and Use Of Replay Id Explained (in English)

Updated 2026-11-07
Published 2024-01-10 · update checked 2026-11-07 · Watch on YouTube

What's changed since the video

Each change is dated and checked against Salesforce's own documentation.

  • Rare isn't never. Salesforce now says it plainly: "In rare cases, the event message might not be persisted in the distributed system during the initial or subsequent attempts. This means that the events aren’t delivered to subscribers, and they aren’t recoverable" (Considerations for Defining and Publishing Platform Events, Platform Events Developer Guide, checked 2026-11-07). Both courses close the gap with a return receipt.
  • High volume is now every platform event. New platform events are high volume by default, and standard-volume ones retire in Summer '27 (Standard-Volume Platform Event Allocations, Platform Events Developer Guide, checked 2026-11-07).

Still right

  • A recovery design: the subscriber keeps the last replay ID it handled somewhere that survives, reconnects with it, and catches up in order, as long as it's within the 72 hours.
  • Past the window, or with an ID that isn't valid, the reconnect fails, and the subscriber has to be ready for that.
  • The video calls this "near guaranteed" delivery, not guaranteed, and that's still the honest word.

All videos in this series: Arquitectura de integración

Un video se carga desde YouTube solo cuando haces clic en él.