moveChunk

Definition

  1. db.adminCommand( { moveChunk : <namespace> ,
  2.                  find : <query> ,
  3.                  to : <string>,
  4.                  _secondaryThrottle : <boolean>,
  5.                  writeConcern: <document>,
  6.                  _waitForDelete : <boolean> } )

Alternatively:

  1. db.adminCommand( { moveChunk : <namespace> ,
  2.                  bounds : <array> ,
  3.                  to : <string>,
  4.                  _secondaryThrottle : <boolean>,
  5.                  writeConcern: <document>,
  6.                  _waitForDelete : <boolean> } )

The moveChunk command has the following fields:

FieldTypeDescriptionmoveChunkstringThe namespace of the collection where thechunk exists. Specify the collection’s full namespace,including the database name.finddocumentAn equality match on the shard key that specifies the shard-key valueof the chunk to move. Specify either the bounds field or thefind field but not both. Do not use the find field toselect chunks in collections that use a hashed shard key.boundsarrayThe bounds of a specific chunk to move. The array must consist of twodocuments that specify the lower and upper shard key values of a chunkto move. Specify either the bounds field or the find field butnot both. Use bounds to select chunks in collections that use ahashed shard key.tostringThe name of the destination shard for the chunk._secondaryThrottlebooleanOptional. Starting in MongoDB 3.4, for WiredTiger,defaults to false.

  • If true, then by default, each document move during chunkmigration propagates to at least one secondary before the balancerproceeds with the next document. This is equivalent to a writeconcern of { w: 2 }.

Use the writeConcern option to specify a different write concern.

  • If false, the balancer does not wait for replication to asecondary and instead continues with the next document.

For more information, seeSecondary Throttle.writeConcerndocumentOptional. A document that expresses the write concern that the _secondaryThrottle will use towait for secondaries during the chunk migration.

writeConcern requires _secondaryThrottle: true._waitForDeletebooleanOptional. Internal option for testing purposes. The default is false. If setto true, the delete phase of a moveChunk operationblocks.

The value of bounds takes the form:

  1. [ { hashedField : <minValue> } ,
  2.   { hashedField : <maxValue> } ]

The chunk migration sectiondescribes how chunks move between shards on MongoDB.

See also

split, sh.moveChunk(),sh.splitAt(), and sh.splitFind().

Considerations

Only use the moveChunk in special circumstancessuch as preparing your sharded cluster for an initialingestion of data, or a large bulk import operation. In most casesallow the balancer to create and balance chunks in sharded clusters.See Create Chunks in a Sharded Cluster for more information.

Behavior

Indexes

Changed in version 3.0.0: In previous versions, moveChunk would build indexes aspart of the migrations.

moveChunk requires that all indexes exist on thetarget (i.e. to ) shard before migration and returns an error if arequired index does not exist.

Meta Data Error

moveChunk returns the following error message if anothermetadata operation is in progress on the chunks collection:

  1. errmsg: "The collection's metadata lock is already taken."

If another process, such as a balancer process, changes meta datawhile moveChunk is running, you may see thiserror. You may retry the moveChunk operation withoutside effects.