Uploaded image for project: 'Flink'
  1. Flink
  2. FLINK-30204 Table Store support separated compact jobs
  3. FLINK-30207

Move split initialization and discovery logic fully into SnapshotEnumerator in Table Store

    XMLWordPrintableJSON

Details

    Description

      It is possible that the separated compact job is started long after the write jobs. so compact job sources need a special split initialization logic: we will find the latest COMPACT snapshot, and start compacting right after this snapshot.

      However, split initialization logic are currently coded into FileStoreSource. We should extract these logic into SnapshotEnumerator so that we can create our special SnapshotEnumerator for compact sources.

      Attachments

        Issue Links

          Activity

            People

              TsReaper Caizhi Weng
              TsReaper Caizhi Weng
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: