Uploaded image for project: 'HBase'
  1. HBase
  2. HBASE-18036

HBase 1.x : Data locality is not maintained after cluster restart or SSH

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Critical
    • Resolution: Fixed
    • 1.4.0, 1.3.1, 1.2.5, 1.1.10
    • 1.4.0, 1.3.2, 1.1.11, 1.2.7
    • Region Assignment
    • None
    • Reviewed

    Description

      After HBASE-2896 / HBASE-4402, we think data locality is maintained after cluster restart. However, we have seem some complains about data locality loss when cluster restart (eg. HBASE-17963).

      Examining the AssignmentManager#processDeadServersAndRegionsInTransition() code, for cluster start, I expected to hit the following code path:

          if (!failover) {
      // Fresh cluster startup.
      LOG.info("Clean cluster startup. Assigning user regions");
      assignAllUserRegions(allRegions);
    }

      where assignAllUserRegions would use retainAssignment() call in LoadBalancer; however, from master log, we usually hit the failover code path:

          // If we found user regions out on cluster, its a failover.
    if (failover) {
      LOG.info("Found regions out on cluster or in RIT; presuming failover");
      // Process list of dead servers and regions in RIT.
      // See HBASE-4580 for more information.
      processDeadServersAndRecoverLostRegions(deadServers);
    }

      where processDeadServersAndRecoverLostRegions() would put dead servers in SSH and SSH uses roundRobinAssignment() in LoadBalancer. That is why we would see loss locality more often than retaining locality during cluster restart.

      Note: the code I was looking at is close to branch-1 and branch-1.1.

      Attachments

        1. HBASE-18036.v0-branch-1.1.patch
          9 kB
          Stephen Yuan Jiang
        2. HBASE-18036.v0-branch-1.patch
          3 kB
          Stephen Yuan Jiang
        3. HBASE-18036.v1-branch-1.1.patch
          9 kB
          Stephen Yuan Jiang
        4. HBASE-18036.v2-branch-1.1.patch
          9 kB
          Stephen Yuan Jiang

        Issue Links

          is duplicated by

          Bug - A problem which impairs or prevents the functions of the product. HBASE-21006 Balancer - data locality drops 30-40% across all nodes after every cluster-wide rolling restart, not migrating regions back to original RegionServers?

          • Major - Major loss of function.
          • Resolved
          is related to

          Bug - A problem which impairs or prevents the functions of the product. HBASE-18246 Proc-V2 AM: Maintain Data locality in ServerCrashProcedure

          • Critical - Crashes, loss of data, severe memory leak.
          • Resolved
          relates to

          Bug - A problem which impairs or prevents the functions of the product. HBASE-15251 During a cluster restart, Hmaster thinks it is a failover by mistake

          • Major - Major loss of function.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. HBASE-17963 RegionServers lose file locality on unplanned restart

          • Major - Major loss of function.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. HBASE-14129 If any regionserver gets shutdown uncleanly during full cluster restart, locality looks to be lost

          • Major - Major loss of function.
          • Closed

          Improvement - An improvement or enhancement to an existing feature or task. HBASE-17791 Locality should not be affected for non-faulty region servers at startup

          • Blocker - Blocks development and/or testing work, production could not run
          • Open

          Activity

            People

              syuanjiang Stephen Yuan Jiang
              syuanjiang Stephen Yuan Jiang
              Votes:
              4 Vote for this issue
              Watchers:
              11 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: