Hadoop HDFS
  1. Hadoop HDFS
  2. HDFS-1553

Hftp file read should retry a different datanode if the chosen best datanode fails to connect to NameNode

    Details

    • Type: Bug Bug
    • Status: Open
    • Priority: Major Major
    • Resolution: Unresolved
    • Affects Version/s: None
    • Fix Version/s: 0.24.0
    • Component/s: namenode
    • Labels:
      None

      Description

      Currently when reading a file through HftpFileSystem interface, namenode deterministically selects the "best" datanode from which the file is read. But this can cause the read to fail if the best datanode fails to connect to namenode because it never tries another datanode.

      The proposed solution is to send a list of datanode candidates when namenode redirects the read request to the chosen best datanode. The the datanode could redirect the request to the next good datanode when it fails to connect to namenode.

        Activity

        No work has yet been logged on this issue.

          People

          • Assignee:
            Unassigned
            Reporter:
            Hairong Kuang
          • Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

            Dates

            • Created:
              Updated:

              Development