streaming: track ingested timestamp in ingestion job #57400
Labels
A-disaster-recovery
C-enhancement
Solution expected to add code/behavior + preserve backward-compat (pg compat issues are exception)
T-disaster-recovery
The cluster stream ingestion job (#57399) needs to be able to track the ingested timestamp across all of it's partition ingestion processors.
Each partition will need to report back to the job the latest timestamp at which:
a) it has received a resolved timestamp event for
b) it has flushed
The job will then also need to keep track of the earliest of these timestamps across all of its stream ingestion processors.
The text was updated successfully, but these errors were encountered: