Uploaded image for project: 'Hadoop HDFS'
  1. Hadoop HDFS
  2. HDFS-4257

The ReplaceDatanodeOnFailure policies could have a forgiving option



    • New Feature
    • Status: Closed
    • Minor
    • Resolution: Fixed
    • 2.0.2-alpha
    • 2.6.0
    • hdfs-client
    • None


      Similar question has previously come over HDFS-3091 and friends, but the essential problem is: "Why can't I write to my cluster of 3 nodes, when I just have 1 node available at a point in time.".

      The policies cover the 4 options, with Default being default:

      Disable -> Disables the whole replacement concept by throwing out an error (at the server) or acts as Never at the client.
      Never -> Never replaces a DN upon pipeline failures (not too desirable in many cases).
      Default -> Replace based on a few conditions, but whose minimum never touches 1. We always fail if only one DN remains and none others can be added.
      Always -> Replace no matter what. Fail if can't replace.

      Would it not make sense to have an option similar to Always/Default, where despite trying, if it isn't possible to have > 1 DN in the pipeline, do not fail. I think that is what the former write behavior was, and what fit with the minimum replication factor allowed value.

      Why is it grossly wrong to pass a write from a client for a block with just 1 remaining replica in the pipeline (the minimum of 1 grows with the replication factor demanded from the write), when replication is taken care of immediately afterwards? How often have we seen missing blocks arise out of allowing this + facing a big rack(s) failure or so?


        1. h4257_20140325.patch
          14 kB
          Tsz-wo Sze
        2. h4257_20140325b.patch
          14 kB
          Tsz-wo Sze
        3. h4257_20140326.patch
          14 kB
          Tsz-wo Sze
        4. h4257_20140819.patch
          14 kB
          Tsz-wo Sze
        5. h4257_20140831.patch
          14 kB
          Tsz-wo Sze

        Issue Links



              szetszwo Tsz-wo Sze
              qwertymaniac Harsh J
              0 Vote for this issue
              20 Start watching this issue