Uploaded image for project: 'Hadoop HDFS'
  1. Hadoop HDFS
  2. HDFS-15283

Cache pool MAXTTL is not persisted and restored on cluster restart

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • 3.4.0
    • 3.3.0, 3.1.4, 3.2.2, 2.10.1
    • namenode
    • None
    • Reviewed

    Description

      The cache pool "getMaxRelativeExpiryMs" is never persisted to or read from the FSImage. This means that if a MAXTTL is set on a pool, it will not persist beyond a cluster restart.

      From the protobuf definition, there is an existing field to store it:

      message CachePoolInfoProto {
        optional string poolName = 1;
        optional string ownerName = 2;
        optional string groupName = 3;
        optional int32 mode = 4;
        optional int64 limit = 5;
        optional int64 maxRelativeExpiry = 6; <-- NEVER SET
        optional uint32 defaultReplication = 7 [default=1];
      }
      

      But this is never set in the CacheManager.saveState() or read in CacheManager.loadState().

      Attachments

        1. HDFS-15283.001.patch
          3 kB
          Stephen O'Donnell

        Activity

          People

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

            Dates

              Created:
              Updated:
              Resolved: