Make ReshardingRecipientService wait for replication lag across all nodes to be some threshold before building indexes

XMLWordPrintableJSON

    • Cluster Scalability
    • Fully Compatible
    • ClusterScalability Apr14-Apr28
    • None
    • 3
    • None
    • None
    • None
    • None
    • None
    • None

      The index builds in resharding use the default commit quorum which is "votingMembers". Making each recipient wait for the cloning to have replicated to all nodes before building indexes (which are resource-intensive) can help reduce the chance of the nodes not being able to catch up later on. For why the resharding uses the default commit quorum is used instead of "majority" commit quorum, please see this comment in HELP-69920.

            Assignee:
            Cheahuychou Mao
            Reporter:
            Cheahuychou Mao
            Votes:
            0 Vote for this issue
            Watchers:
            4 Start watching this issue

              Created:
              Updated:
              Resolved: