Uploaded image for project: 'JGroups'
  1. JGroups
  2. JGRP-1382

VERIFY_SUSPECT can kick a live node soon after it re-joins

    Details

    • Type: Bug
    • Status: Resolved (View Workflow)
    • Priority: Major
    • Resolution: Done
    • Affects Version/s: 2.4.10, 2.6.20
    • Fix Version/s: 2.6.21, 3.0
    • Labels:
      None

      Description

      Due to a timing issue, VERIFY_SUSPECT can kick a good member from the cluster soon after joining.

      A live node currently in the cluster stops responding temporarily (garbage collection, NIC down, etc).
      FD sends up a suspect message to VERIFY_SUSPECT.
      VERIFY_SUSPECT times out and sends up the suspect message.
      Before the node is kicked from the cluster, FD sends up another suspect message to VERIFY_SUSPECT.
      Node is kicked from the cluster.
      Node resumes, is shunned, and rejoins.
      VERIFY_SUSPECT times out and sends up the suspect message.
      Node is kicked from the cluster again.

        Gliffy Diagrams

          Attachments

            Issue Links

              Activity

                People

                • Assignee:
                  dereed Dennis Reed
                  Reporter:
                  dereed Dennis Reed
                • Votes:
                  0 Vote for this issue
                  Watchers:
                  1 Start watching this issue

                  Dates

                  • Created:
                    Updated:
                    Resolved: