HBase
  1. HBase
  2. HBASE-4235

Attempts to reconnect to expired ZooKeeper sessions

    Details

    • Type: Task Task
    • Status: Resolved
    • Priority: Major Major
    • Resolution: Invalid
    • Affects Version/s: 0.90.5, 0.92.0
    • Fix Version/s: None
    • Component/s: None
    • Labels:
      None

      Description

      In a couple of instances of short network outages, we have observed afterward zombie HBase processes attempting over and over to reconnect to expired ZooKeeper sessions. We believe this is due to ZOOKEEPER-1159. Opening this issue as reference to that.

        Issue Links

          Activity

          Hide
          Andrew Purtell added a comment -

          @Ken: Not sure if it is the same issue. ZooKeeper logs say the session expired?

          Pardon, when I spoke with Camille at Hadoop World NYC the verdict was the patch on 1159 was not correct, and this was the basis for 'ClientCnxn does not propagate session expiration indication'. This got jumbled in my head to an issue resolution. I do have the impression that jira will be resolved that way.

          Show
          Andrew Purtell added a comment - @Ken: Not sure if it is the same issue. ZooKeeper logs say the session expired? Pardon, when I spoke with Camille at Hadoop World NYC the verdict was the patch on 1159 was not correct, and this was the basis for 'ClientCnxn does not propagate session expiration indication'. This got jumbled in my head to an issue resolution. I do have the impression that jira will be resolved that way.
          Hide
          Ken Weiner added a comment -

          I am seeing similar behavior after restart of my HBase 0.90.4 cluster. The messages that are printed continuously look like the below. Is it the same issue? Also, ZOOKEEPER-1159 is not yet resolved yet. Which upstream issue was marked as invalid?

          - Socket connection established to localhost/127.0.0.1:2181, initiating session
          - Unable to read additional data from server sessionid 0x1345e2d625d0009, likely server has closed socket, closing socket connection and attempting reconnect
          
          Show
          Ken Weiner added a comment - I am seeing similar behavior after restart of my HBase 0.90.4 cluster. The messages that are printed continuously look like the below. Is it the same issue? Also, ZOOKEEPER-1159 is not yet resolved yet. Which upstream issue was marked as invalid? - Socket connection established to localhost/127.0.0.1:2181, initiating session - Unable to read additional data from server sessionid 0x1345e2d625d0009, likely server has closed socket, closing socket connection and attempting reconnect
          Hide
          Andrew Purtell added a comment -

          ZooKeeper marked upstream issue as invalid.

          Show
          Andrew Purtell added a comment - ZooKeeper marked upstream issue as invalid.

            People

            • Assignee:
              Andrew Purtell
              Reporter:
              Andrew Purtell
            • Votes:
              0 Vote for this issue
              Watchers:
              1 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Development