Uploaded image for project: 'Couchbase Server'
  1. Couchbase Server
  2. MB-4710

view immediately returns ""reason": "conn_failed" if one node is down and not failed over yet. it should return partial results instead

    Details

    • Type: Bug
    • Status: Resolved
    • Priority: Blocker
    • Resolution: Fixed
    • Affects Version/s: 2.0-developer-preview-3
    • Fix Version/s: 2.0-developer-preview-4
    • Component/s: ns_server, view-engine
    • Security Level: Public
    • Labels:
      None
    • Environment:
      build 599

      Description

      10.114.167.27 was stopped using //etc/init.d/couchbase-server command

      view defintion : function (doc)

      { emit(doc._id, doc); }

      Bucket Name Nodes Item Count Ops/sec Disk Fetches/sec RAM Usage/Quota Disk Usage
      default 6 108002 1504 0 61.9MB / 12.9GB 123MB
      Views
      Documents

      Subset of nodes failed with the following error:
      [

      { "from": "http://10.114.167.27:8092/_view_merge/?stale=false", "reason": "conn_failed" }

      ]

      http://10.212.167.16:8092/default/_design/dev_hi/_view/hi?full_set=true&stale=false&connection_timeout=60000&limit=10&skip=0

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

        Activity

        Hide
        farshid Farshid Ghods (Inactive) added a comment -

        Filipe,

        im not sure if replica index is enabled or disabled in this build
        the error goes away right after i "failover" the node

        Show
        farshid Farshid Ghods (Inactive) added a comment - Filipe, im not sure if replica index is enabled or disabled in this build the error goes away right after i "failover" the node
        Hide
        filipe manana filipe manana added a comment - - edited

        Farshid, can't reproduce.
        Created a 2 nodes cluster with 160 000 docs, reduce function is "_count".

        1) Queried the view, returned a row with value 160 000 (correct);

        2) kill -9 node 2, node 1's UI tells me node 2 is not alive;

        3) Without doing a failover, queried the view and got:

        fdmanana 14:55:30 ~ > curl 'http://localhost:9500/default/_design/test/_view/view1?stale=false'
        {"rows":[

        {"key":null,"value":80000}

        ],
        "errors":[

        {"from":"http://192.168.1.80:9501/_view_merge/?stale=false","reason":"conn_failed"}

        ]}

        This is correct.
        I tested this with the latest build of everything (clean repo checkout of branch-2.0.xml).

        Show
        filipe manana filipe manana added a comment - - edited Farshid, can't reproduce. Created a 2 nodes cluster with 160 000 docs, reduce function is "_count". 1) Queried the view, returned a row with value 160 000 (correct); 2) kill -9 node 2, node 1's UI tells me node 2 is not alive; 3) Without doing a failover, queried the view and got: fdmanana 14:55:30 ~ > curl 'http://localhost:9500/default/_design/test/_view/view1?stale=false' {"rows":[ {"key":null,"value":80000} ], "errors":[ {"from":"http://192.168.1.80:9501/_view_merge/?stale=false","reason":"conn_failed"} ]} This is correct. I tested this with the latest build of everything (clean repo checkout of branch-2.0.xml).
        Hide
        farshid Farshid Ghods (Inactive) added a comment -

        expected behavior

        Show
        farshid Farshid Ghods (Inactive) added a comment - expected behavior

          People

          • Assignee:
            FilipeManana Filipe Manana (Inactive)
            Reporter:
            farshid Farshid Ghods (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