Uploaded image for project: 'Couchbase Java Client'
  1. Couchbase Java Client
  2. JCBC-1118

N1QL with request_plus throughput got lowered after all nodes restart one after another

    XMLWordPrintable

Details

    • Bug
    • Resolution: User Error
    • Critical
    • 2.7.1
    • None
    • None
    • None

    Description

      cluster: 1kv, 2n1ql, 1 index

      n1ql/index_engine gsi
      n1ql/index_type secondary
      n1ql/preload true
      n1ql/prepared true
      n1ql/scan_consistency request_plus

      with above settings, against SPOCK.

      Test was done by restarting all nodes one by one and check throughput.

      Expected: n1ql throughput stays same after restart completed

      Actual : n1ql throughput got lowered (latency got higher) by half

      This is exiting issue even against watson, but there was no ticket to track.

      Here is example output of Watson : http://sdk-testresults.couchbase.com.s3.amazonaws.com/SDK-SDK/CB-4.5.1-xxxx/SvcRestartQuery-N1QL/07-11-17/083073/02aff67b16432c45a28f0bd24e23f067-N1QL.html

      Here is an example output of Spock : http://sdk-testresults.couchbase.com.s3.amazonaws.com/SDK-SDK/CB-5.0.0-3497/SvcRestartQuery-N1QL/08-18-17/078896/580fdde5cef4ec27569c0176df375e0a-N1QL.html

      Attachments

        No reviews matched the request. Check your Options in the drop-down menu of this sections header.

        Activity

          People

            jaekwon.park Jae Park [X] (Inactive)
            jaekwon.park Jae Park [X] (Inactive)
            Votes:
            0 Vote for this issue
            Watchers:
            4 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved:

              Gerrit Reviews

                There are no open Gerrit changes

                PagerDuty