cockroachdb / cockroachdb/cockroach
changefeedccl: have each aggregator save its own frontier
Open
A-cdc
C-enhancement
T-cdc
- Dominant language
- Go
- Stars
- 32.5k
- Forks
- 4.1k
- PR merge metrics
- PR metrics pending
Description
Similar to #148124 except we'll have each change aggregator persist its frontier instead of a span-level checkpoint.
This would remove a bottleneck at the change frontier processor.
Note that this currently blocked by the fact that to be able to write to job_info, the node must have the job lease (i.e. the changefeed coordinator): https://cockroachlabs.slack.com/archives/C02DSDS9TM1/p1758294532385269
Jira issue: CRDB-54540
Contributor guide
Assessment
This issue has not been assessed yet.