Uploaded image for project: 'Couchbase Server'
  1. Couchbase Server
  2. MB-5895

when adding 1.8.1 node to 1.8.0/1.7.2 cluster orchestrator takes over but does not use the vbucket map history so add/remove one node won't execute swap rebalance

    XMLWordPrintable

Details

    • Bug
    • Resolution: Won't Fix
    • Blocker
    • None
    • 1.8.1
    • ns_server
    • Security Level: Public
    • Ubuntu 32 bit.
      181-937-rel

    Description

      We wait till, the following message apperars on the UI, before issuing rebalance on this cluster.

      Current master is older and I'll try to takeover (repeated 1 times).

      4 node 180 cluster, added on 181 node and swap rebalanced it with one of the existing 180 node.

      On doing so, the vbuckets movement do not just happen between the node that is being removed and the 181 node that is being added on.

      Please see the logs attached from all the nodes in the cluster.

      Also, attached the screenshot of one such movement.

      Attachments

        1. 10.3.2.153-8091-diag.txt.gz
          1.28 MB
        2. 10.3.2.154-8091-diag.txt.gz
          558 kB
        3. 10.3.2.155-8091-diag.txt.gz
          2.25 MB
        4. 10.3.2.156-8091-diag.txt.gz
          268 kB
        5. 10.3.2.157-8091-diag.txt.gz
          480 kB
        6. 10.3.2.158-8091-diag.txt.gz
          478 kB
        7. Screen Shot 2012-07-12 at 6.03.02 PM.png
          Screen Shot 2012-07-12 at 6.03.02 PM.png
          457 kB
        No reviews matched the request. Check your Options in the drop-down menu of this sections header.

        Activity

          People

            alkondratenko Aleksey Kondratenko (Inactive)
            karan Karan Kumar (Inactive)
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved:

              Gerrit Reviews

                There are no open Gerrit changes

                PagerDuty