Uploaded image for project: 'Couchbase Server'
  1. Couchbase Server
  2. MB-7843

[RN 2.0.1]Documentation for new parameter rebalanceMovesBeforeCompaction used for improving rebalance speed

    Details

    • Flagged:
      Release Note

      Description

      MB-6726 - Rebalance is slow when indexing/compaction and query load are going on in parallel

      rebalanceMovesBeforeCompaction is the parameter that customers can use during rebalance to improve rebalance speeds but at the cost of index file size.

      [ALK]
      wget --post-data='rebalanceMovesBeforeCompaction=256' --user=Administrator --password=asdasd http://lh:9000/internalSettings

      So that's usual POST to usual /internalSettings API with field name rebalanceMovesBeforeCompaction.

      Default value is 16. And that's how many in- or out- going vbuckets will be moved per node before pausing all moves, triggering views compaction and awaiting its completion.

      View compaction is forbidden during vbucket moves. So larger values means less frequent view compactions. Which means less time "spent" on it, but it also means greater disk usage by views.

      For additional info ping Alk, Pavel.

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

        Activity

        Hide
        kzeller kzeller added a comment -

        Added to RN:

        <issue type="cb" ref="MB-7843"/>

        <rntext>

        <para>
        There is a new REST-API call which can be used if you
        experience a slow rebalance while indexing and compaction are in
        progress. For more information, see
        <xref linkend="couchbase-admin-restapi-rebalance-before-compaction" />.
        </para>

        Added to Chapter, which is cross referenced:

        <section id="couchbase-admin-restapi-rebalance-before-compaction">
        <title>Adjusting Rebalance during Compaction</title>
        <para>If you perform a rebalance while a node is undergoing index compaction,
        you may experience delays in rebalance. There is REST-API parameter as of
        Couchbase Server 2.0.1 you can use to improve rebalance performance.
        If you do make this selection, you will reduce the performance of index
        compaction which can result in larger index file size.</para>
        <para>To make this request:</para>
        <programlisting>
        wget --post-data='rebalanceMovesBeforeCompaction=256'
        --user=Administrator --password=pass http://lh:9000/internalSettings
        </programlisting>
        <para>This needs to be made as POST request to the <literal>
        /internalSettings</literal> endpoint. By default this setting is 16, which
        specifies the number of vBuckets which will moved per node before
        pausing all moves, which will then trigger index compaction. Index compaction
        will not be performed while vBuckets are being moved, so if you
        specify a larger value, it means that the server will spend less time
        compacting the index, which will result in larger index files that take
        up more disk space.</para>
        </section>

        Show
        kzeller kzeller added a comment - Added to RN: <issue type="cb" ref=" MB-7843 "/> <rntext> <para> There is a new REST-API call which can be used if you experience a slow rebalance while indexing and compaction are in progress. For more information, see <xref linkend="couchbase-admin-restapi-rebalance-before-compaction" />. </para> Added to Chapter, which is cross referenced: <section id="couchbase-admin-restapi-rebalance-before-compaction"> <title>Adjusting Rebalance during Compaction</title> <para>If you perform a rebalance while a node is undergoing index compaction, you may experience delays in rebalance. There is REST-API parameter as of Couchbase Server 2.0.1 you can use to improve rebalance performance. If you do make this selection, you will reduce the performance of index compaction which can result in larger index file size.</para> <para>To make this request:</para> <programlisting> wget --post-data='rebalanceMovesBeforeCompaction=256' --user=Administrator --password=pass http://lh:9000/internalSettings </programlisting> <para>This needs to be made as POST request to the <literal> /internalSettings</literal> endpoint. By default this setting is 16, which specifies the number of vBuckets which will moved per node before pausing all moves, which will then trigger index compaction. Index compaction will not be performed while vBuckets are being moved, so if you specify a larger value, it means that the server will spend less time compacting the index, which will result in larger index files that take up more disk space.</para> </section>
        Hide
        kzeller kzeller added a comment -

        Added to RN:

        <issue type="cb" ref="MB-7843"/>

        <rntext>

        <para>
        There is a new REST-API call which can be used if you
        experience a slow rebalance while indexing and compaction are in
        progress. For more information, see
        <xref linkend="couchbase-admin-restapi-rebalance-before-compaction" />.
        </para>

        Added to Chapter, which is cross referenced:

        <section id="couchbase-admin-restapi-rebalance-before-compaction">
        <title>Adjusting Rebalance during Compaction</title>
        <para>If you perform a rebalance while a node is undergoing index compaction,
        you may experience delays in rebalance. There is REST-API parameter as of
        Couchbase Server 2.0.1 you can use to improve rebalance performance.
        If you do make this selection, you will reduce the performance of index
        compaction which can result in larger index file size.</para>
        <para>To make this request:</para>
        <programlisting>
        wget --post-data='rebalanceMovesBeforeCompaction=256'
        --user=Administrator --password=pass http://lh:9000/internalSettings
        </programlisting>
        <para>This needs to be made as POST request to the <literal>
        /internalSettings</literal> endpoint. By default this setting is 16, which
        specifies the number of vBuckets which will moved per node before
        pausing all moves, which will then trigger index compaction. Index compaction
        will not be performed while vBuckets are being moved, so if you
        specify a larger value, it means that the server will spend less time
        compacting the index, which will result in larger index files that take
        up more disk space.</para>
        </section>

        Show
        kzeller kzeller added a comment - Added to RN: <issue type="cb" ref=" MB-7843 "/> <rntext> <para> There is a new REST-API call which can be used if you experience a slow rebalance while indexing and compaction are in progress. For more information, see <xref linkend="couchbase-admin-restapi-rebalance-before-compaction" />. </para> Added to Chapter, which is cross referenced: <section id="couchbase-admin-restapi-rebalance-before-compaction"> <title>Adjusting Rebalance during Compaction</title> <para>If you perform a rebalance while a node is undergoing index compaction, you may experience delays in rebalance. There is REST-API parameter as of Couchbase Server 2.0.1 you can use to improve rebalance performance. If you do make this selection, you will reduce the performance of index compaction which can result in larger index file size.</para> <para>To make this request:</para> <programlisting> wget --post-data='rebalanceMovesBeforeCompaction=256' --user=Administrator --password=pass http://lh:9000/internalSettings </programlisting> <para>This needs to be made as POST request to the <literal> /internalSettings</literal> endpoint. By default this setting is 16, which specifies the number of vBuckets which will moved per node before pausing all moves, which will then trigger index compaction. Index compaction will not be performed while vBuckets are being moved, so if you specify a larger value, it means that the server will spend less time compacting the index, which will result in larger index files that take up more disk space.</para> </section>
        Hide
        dipti Dipti Borkar added a comment -

        Thanks Karen, is this live now?

        Show
        dipti Dipti Borkar added a comment - Thanks Karen, is this live now?

          People

          • Assignee:
            kzeller kzeller
            Reporter:
            anil Anil Kumar
          • Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved:

              Gerrit Reviews

              There are no open Gerrit changes