In case of disaster or other anomaly recovery there may be cases where new events from a channel have id that are previously seen. Matching only id and channel will make listener to skip those events.
We should match events by:
source_id, channel, timestamp and name
in order to detect duplicate events. If any of the above differs then we have a new event that we should add to store.
In case of disaster or other anomaly recovery there may be cases where new events from a channel have id that are previously seen. Matching only id and channel will make listener to skip those events.
We should match events by: source_id, channel, timestamp and name in order to detect duplicate events. If any of the above differs then we have a new event that we should add to store.