Uploaded image for project: 'Apache Ozone'
  1. Apache Ozone
  2. HDDS-8699

Further Replication Manager Improvements

    XMLWordPrintableJSON

Details

    • Epic
    • Status: Open
    • Major
    • Resolution: Unresolved
    • None
    • None
    • None
    • None
    • Further Replication Manager Improvements

    Description

      Splitting out some tasks from HDDS-7759 we don't have time for right now, but that may be good candidates for future work so we would like to keep track of them.

      Attachments

        Issue Links

          1.
          Enable dynamic reconfiguration of some RM properties Sub-task Resolved Attila Doroszlai
          2.
          Datanode Command Count Updated related DEBUG logging is too frequent Sub-task Resolved Siddhant Sangwan
          3.
          ReplicationManager: Unhealthy replicas could block Ratis containers being recovered Sub-task Resolved Stephen O'Donnell
          4.
          Decommissioning blocked because of under replicated EC containers Sub-task Resolved Stephen O'Donnell
          5.
          LegacyReplicationManager: Unhealthy replicas could block under replication handling Sub-task Resolved Siddhant Sangwan
          6.
          LegacyReplicationManager: Delete excess unhealthy with force=true Sub-task Resolved Stephen O'Donnell
          7.
          LegacyReplicationManager: Handle all UNHEALTHY replicas of a CLOSING container Sub-task Resolved Siddhant Sangwan
          8.
          LegacyReplicationManager: Save UNHEALTHY replicas with highest BCSID for a QUASI_CLOSED container Sub-task Resolved Siddhant Sangwan
          9.
          ReplicationManager: Handle all UNHEALTHY replicas of a CLOSING Ratis container Sub-task Resolved Stephen O'Donnell
          10.
          LegacyReplicationManager: Ignore any Datanodes that are not in-service and healthy when finding unique origins Sub-task Resolved Siddhant Sangwan
          11.
          LegacyReplicationManager: Unhealthy replicas of a sufficiently replicated container can block decommissioning Sub-task Resolved Siddhant Sangwan
          12.
          LegacyReplicationManager: Do not count unique origin nodes as over-replicated Sub-task Resolved Ethan Rose
          13.
          Replication Manager could incorrectly use QUASI_CLOSED replicas as replication sources for CLOSED containers Sub-task Resolved Siddhant Sangwan
          14.
          ReplicationManager: Ignore any Datanodes that are not in-service and healthy when finding unique origins Sub-task Resolved Siddhant Sangwan
          15.
          Replication Manager: Do not count unique origin nodes as over-replicated Sub-task Resolved Stephen O'Donnell
          16.
          Legacy Replication Manager should consider that UNHEALTHY replicas might be decommissioning Sub-task Resolved Siddhant Sangwan
          17.
          Incorrect sorting order for all unhealthy replicas in RatisOverReplicationHandler Sub-task Resolved Stephen O'Donnell
          18.
          Replication Manager: Save UNHEALTHY replicas with highest BCSID for a QUASI_CLOSED container Sub-task Resolved Siddhant Sangwan
          19.
          ReplicationManager: Unhealthy replicas of a sufficiently replicated container can block decommissioning Sub-task Resolved Siddhant Sangwan
          20.
          LegacyReplicationManager: Pending deletes on unhealthy replicas can cause calculation errors Sub-task Resolved Unassigned
          21.
          Push replication may fail with StatusRuntimeException: UNAVAILABLE: Channel shutdown invoked Sub-task Open Unassigned
          22.
          Inconsistent command handler stats Sub-task Resolved Unassigned
          23.
          ReplicationManager: Notify when dead nodes or nodes go out of service Sub-task Open Unassigned
          24.
          ReplicationManager: Create RM integration tests Sub-task Open Unassigned
          25.
          Create CLI tool to trigger push replication Sub-task Open Unassigned
          26.
          Investigate QUASI_CLOSED containers with only one UNHEALTHY and empty replica Sub-task Open Unassigned
          27.
          Determine the criteria for calling a container empty in the SCM Sub-task Open Unassigned
          28.
          Provide API to check a container via Replication Manager Sub-task Resolved Stephen O'Donnell
          29.
          Over Replication Check of all UNHEALTHY replicas is broken Sub-task Resolved Siddhant Sangwan
          30.
          Decommission should get the correct ContainerReplicaCount object Sub-task Resolved Unassigned
          31.
          UNHEALTHY replicas of QUASI_CLOSED container with unique origins should be handled during decommission Sub-task Resolved Siddhant Sangwan

          Activity

            People

              Unassigned Unassigned
              sodonnell Stephen O'Donnell
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

                Created:
                Updated: