Uploaded image for project: 'Couchbase Server'
  1. Couchbase Server
  2. MB-30748

[Plasma] Increase in index data size after recovery

    XMLWordPrintable

Details

    • Bug
    • Resolution: Duplicate
    • Major
    • 5.5.1
    • 5.5.0
    • storage-engine
    • None

    Description

      Build : 5.5.0-2958 (GA build)

      Steps:
      1. Create a 2 node cluster - kv+index+query , kv+index
      2. Create a default bucket
      3. Use cbworkloadgen to populate data as below in the default bucket.
      /opt/couchbase/bin/cbworkloadgen -r 1.0 -n localhost:8091 -s 10240 -b default -u Administrator -p password -t 4 -j -i 150000 -v
      4. Create index with 1 replica.
      CREATE INDEX `idx1` ON `default`(`name`,`body`) WITH

      { "num_replica":1 }

      5. Let the index build completely.
      6. Note the data_size stat for the index
      7. On node2, kill indexer - pkill -f indexer
      8. After a few mins, note the data_size stat for the index.

      Issue : Data size for the index keeps on increasing about 20-35% each time the index recovery takes place.

      In my test, here is the data size captured after recovery
      Initial : "default:idx1:data_size": 3179952511
      After 1st recovery : "default:idx1:data_size": 4279274374
      After 2nd recovery : "default:idx1:data_size": 5378591480

      Attachments

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

        Activity

          People

            mihir.kamdar Mihir Kamdar (Inactive)
            mihir.kamdar Mihir Kamdar (Inactive)
            Votes:
            0 Vote for this issue
            Watchers:
            7 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved:

              Gerrit Reviews

                There are no open Gerrit changes

                PagerDuty