Merge Chunks in a Sharded Cluster

On this page本页内容

Overview概述

The mergeChunks command allows you to combine continuous chunks on the same shard into a single chunk. This tutorial explains how to merge neighboring chunks in a sharded cluster.mergeChunks命令允许您将同一分片上的连续块合并为单个块。本教程解释了如何在分片集群中合并相邻的块。

Procedure过程

Note注意

Examples in this procedure use a members collection in the test database, using the username field as the shard key.本过程中的示例使用test数据库中的members集合，将username字段用作分片键。

Identify Chunk Ranges确定区块范围

In mongosh, identify the chunk ranges with the following operation:在mongosh中，使用以下操作标识区块范围：

sh.status()

In the output, the chunk ranges appear after the chunk counts for each sharded collection, as in the following example:在输出中，区块范围显示在每个分片集合的区块计数之后，如以下示例所示：

--- Sharding Status ---
  sharding version: {
     "_id" : 1,
     "minCompatibleVersion" : 5,
     "currentVersion" : 6,
     "clusterId" : ObjectId("5ebf0bfd3eeb6037ec7cbba9")
  }
  shards:
        {  "_id" : "shardA",  "host" : "shardA/shardA-m1.example.net:27018,shardA-m2.example.net:27018,shardA-m3.example.net:27018",  "state" : 1 }
        {  "_id" : "shardB",  "host" : "shardB/shardB-m1.example.net:27018,shardB-m2.example.net:27018,shardB-m3.example.net:27018",  "state" : 1 }
  active mongoses:
        "4.4.0" : 1
  autosplit:
        Currently enabled: yes
  balancer:
        Currently enabled:  yes
        Currently running:  no
        Failed balancer rounds in last 5 attempts:  0
        Migration Results for the last 24 hours:
                519 : Success
  databases:
        {  "_id" : "config",  "primary" : "config",  "partitioned" : true }
                config.system.sessions
                        shard key: { "_id" : 1 }
                        unique: false
                        balancing: true
                        chunks:
                                shardA   512
                                shardB   512
                        too many chunks to print, use verbose if you want to force print
        {  "_id" : "test",  "primary" : "shardA",  "partitioned" : true,  "version" : {  "uuid" : UUID("22c042fc-7e3d-4c6d-992d-f3d714759781"),  "lastMod" : 1 } }
                test.members
                        shard key: { "username" : 1 }
                        unique: false
                        balancing: true
                        chunks:
                                shardA   7
                                shardB   7
                        { "username" : { "$minKey" : 1 } } -->> { "username" : "user16643" } on : shardB Timestamp(13, 0)
                        { "username" : "user16643" } -->> { "username" : "user2329" } on : shardB Timestamp(16, 0)
                        { "username" : "user2329" } -->> { "username" : "user29937" } on : shardB Timestamp(17, 0)
                        { "username" : "user29937" } -->> { "username" : "user36583" } on : shardB Timestamp(18, 0)
                        { "username" : "user36583" } -->> { "username" : "user43229" } on : shardB Timestamp(19, 0)
                        { "username" : "user43229" } -->> { "username" : "user49877" } on : shardB Timestamp(20, 0)
                        { "username" : "user49877" } -->> { "username" : "user56522" } on : shardB Timestamp(21, 0)
                        { "username" : "user56522" } -->> { "username" : "user63169" } on : shardA Timestamp(21, 1)
                        { "username" : "user63169" } -->> { "username" : "user69816" } on : shardA Timestamp(10, 1)
                        { "username" : "user69816" } -->> { "username" : "user76462" } on : shardA Timestamp(11, 1)
                        { "username" : "user76462" } -->> { "username" : "user83108" } on : shardA Timestamp(12, 1)
                        { "username" : "user83108" } -->> { "username" : "user89756" } on : shardA Timestamp(14, 1)
                        { "username" : "user89756" } -->> { "username" : "user96401" } on : shardA Timestamp(15, 1)
                        { "username" : "user96401" } -->> { "username" : { "$maxKey" : 1 } } on : shardA Timestamp(15, 2)

The chunk ranges appear after the chunk counts for each sharded collection. 区块范围显示在每个分片集合的区块计数之后。For example, the following are the chunk ranges for the test.members collection:例如，以下是test.members集合的区块范围：

{ "username" : { "$minKey" : 1 } } -->> { "username" : "user16643" } on : shardB Timestamp(13, 0)
{ "username" : "user16643" } -->> { "username" : "user2329" } on : shardB Timestamp(16, 0)
{ "username" : "user2329" } -->> { "username" : "user29937" } on : shardB Timestamp(17, 0)
{ "username" : "user29937" } -->> { "username" : "user36583" } on : shardB Timestamp(18, 0)
{ "username" : "user36583" } -->> { "username" : "user43229" } on : shardB Timestamp(19, 0)
{ "username" : "user43229" } -->> { "username" : "user49877" } on : shardB Timestamp(20, 0)
{ "username" : "user49877" } -->> { "username" : "user56522" } on : shardB Timestamp(21, 0)
{ "username" : "user56522" } -->> { "username" : "user63169" } on : shardA Timestamp(21, 1)
{ "username" : "user63169" } -->> { "username" : "user69816" } on : shardA Timestamp(10, 1)
{ "username" : "user69816" } -->> { "username" : "user76462" } on : shardA Timestamp(11, 1)
{ "username" : "user76462" } -->> { "username" : "user83108" } on : shardA Timestamp(12, 1)
{ "username" : "user83108" } -->> { "username" : "user89756" } on : shardA Timestamp(14, 1)
{ "username" : "user89756" } -->> { "username" : "user96401" } on : shardA Timestamp(15, 1)
{ "username" : "user96401" } -->> { "username" : { "$maxKey" : 1 } } on : shardA Timestamp(15, 2)

Merge Chunks合并区块

Merge contiguous chunks on the same shard.合并同一分片上的连续块。

For example, consider the following chunk ranges on shardA:例如，考虑shardA上的以下区块范围：

Note注意

The chunks to be merged are highlighted.将突出显示要合并的块。

{ "username" : "user56522" } -->> { "username" : "user63169" } on : shardA Timestamp(21, 1)
{ "username" : "user63169" } -->> { "username" : "user69816" } on : shardA Timestamp(10, 1)
{ "username" : "user69816" } -->> { "username" : "user76462" } on : shardA Timestamp(11, 1)
{ "username" : "user76462" } -->> { "username" : "user83108" } on : shardA Timestamp(12, 1)
{ "username" : "user83108" } -->> { "username" : "user89756" } on : shardA Timestamp(14, 1)
{ "username" : "user89756" } -->> { "username" : "user96401" } on : shardA Timestamp(15, 1)
{ "username" : "user96401" } -->> { "username" : { "$maxKey" : 1 } } on : shardA Timestamp(15, 2)

To merge the highlighted contiguous chunks, issue the mergeChunks command against the admin database:要合并突出显示的连续块，请对管理数据库发出mergeChunks命令：

db.adminCommand( {
   mergeChunks: "test.members",
   bounds: [ { "username" : "user69816" },
             { "username" : "user96401" } ]
} )

On success, mergeChunks produces the following output:成功后，mergeChunks生成以下输出：

{
   "ok" : 1,
   "operationTime" : Timestamp(1589580356, 14),
   "$clusterTime" : {
      "clusterTime" : Timestamp(1589580356, 14),
      "signature" : {
         "hash" : BinData(0,"up5VKd49G/uPCq1iger2nOtfIHw="),
         "keyId" : NumberLong("6827188741371592725")
      }
   }
}

On any failure condition, mergeChunks returns a document where the value of the ok field is 0.在任何失败条件下，mergeChunks都会返回ok字段值为0的文档。

View Merged Chunks Ranges查看合并区块范围

After merging the identified chunks, confirm the new chunk, as follows:合并识别的块后，按如下方式确认新块：

sh.status()

The output of sh.status() should resemble:sh.status()的输出应该类似于：

--- Sharding Status ---
  sharding version: {
     "_id" : 1,
     "minCompatibleVersion" : 5,
     "currentVersion" : 6,
     "clusterId" : ObjectId("5ebef5447fa151d4bd79dd72")
  }
  shards:
        {  "_id" : "shardA",  "host" : "shardA/shardA-m1.example.net:27018,shardA-m2.example.net:27018,shardA-m3.example.net:27018",  "state" : 1 }
        {  "_id" : "shardB",  "host" : "shardB/shardB-m1.example.net:27018,shardB-m2.example.net:27018,shardB-m3.example.net:27018",  "state" : 1 }
  active mongoses:
        "4.4.0" : 1
  autosplit:
        Currently enabled: yes
  balancer:
        Currently enabled:  yes
        Currently running:  no
        Failed balancer rounds in last 5 attempts:  0
        Migration Results for the last 24 hours:
                519 : Success
  databases:
        {  "_id" : "config",  "primary" : "config",  "partitioned" : true }
                config.system.sessions
                        shard key: { "_id" : 1 }
                        unique: false
                        balancing: true
                        chunks:
                                shardA   512
                                shardB   512
                       too many chunks to print, use verbose if you want to force print
        {  "_id" : "test",  "primary" : "shardA",  "partitioned" : true,  "version" : {  "uuid" : UUID("22c042fc-7e3d-4c6d-992d-f3d714759781"),  "lastMod" : 1 } }
                test.members
                        shard key: { "username" : 1 }
                        unique: false
                        balancing: true
                        chunks:
                                shardA
       5
                                shardB
       6
                        { "username" : { "$minKey" : 1 } } -->> { "username" : "user16643" } on : shardA Timestamp(22, 0)
                        { "username" : "user16643" } -->> { "username" : "user2329" } on : shardB Timestamp(22, 1)
                        { "username" : "user2329" } -->> { "username" : "user29937" } on : shardB Timestamp(17, 0)
                        { "username" : "user29937" } -->> { "username" : "user36583" } on : shardB Timestamp(18, 0)
                        { "username" : "user36583" } -->> { "username" : "user43229" } on : shardB Timestamp(19, 0)
                        { "username" : "user43229" } -->> { "username" : "user49877" } on : shardB Timestamp(20, 0)
                        { "username" : "user49877" } -->> { "username" : "user56522" } on : shardB Timestamp(21, 0)
                        { "username" : "user56522" } -->> { "username" : "user63169" } on : shardA Timestamp(21, 1)
                        { "username" : "user63169" } -->> { "username" : "user69816" } on : shardA Timestamp(10, 1)
                        { "username" : "user69816" } -->> { "username" : "user96401" } on : shardA Timestamp(21, 2)
                        { "username" : "user96401" } -->> { "username" : { "$maxKey" : 1 } } on : shardA Timestamp(15, 2)

After the merge, the balancer may migrate chunks across shards to ensure a more even distribution of chunks.合并后，平衡器可以跨分片迁移块，以确保块的分布更加均匀。

←  Split Chunks in a Sharded ClusterModify Chunk Size in a Sharded Cluster →