The purpose of this topic was to setup locally a MongoDB shard, with 2 shards and then run various commands to check the actual behaviour of this system.

Environment Setup
Sharding Setup
DB Setup
Playing around
Configuring zones

Setup

The setup will have.

• 2 replicasets, each with one primary and 2 secondaries.
• 2 routers.
• 3 config server.

docker-compose.yaml

version: '3'
services:
# replica set 1
  mongors1n1:
    container_name: mongors1n1
    image: mongo
    command: mongod --shardsvr --replSet mongors1 --dbpath /data/db --port 27017
    ports:
      - 27017:27017
    expose:
      - "27017"
    volumes:
      - ~/mongo_cluster/data1:/data/db

  mongors1n2:
    container_name: mongors1n2
    image: mongo
    command: mongod --shardsvr --replSet mongors1 --dbpath /data/db --port 27017
    ports:
      - 27027:27017
    expose:
      - "27017"
    volumes:
      - ~/mongo_cluster/data2:/data/db

  mongors1n3:
    container_name: mongors1n3
    image: mongo
    command: mongod --shardsvr --replSet mongors1 --dbpath /data/db --port 27017
    ports:
      - 27037:27017
    expose:
      - "27017"

    volumes:
      - ~/mongo_cluster/data3:/data/db

# replica set 2
  mongors2n1:
    container_name: mongors2n1
    image: mongo
    command: mongod --shardsvr --replSet mongors2 --dbpath /data/db --port 27017
    ports:
      - 27047:27017
    expose:
      - "27017"
    volumes:
      - ~/mongo_cluster/data4:/data/db

  mongors2n2:
    container_name: mongors2n2
    image: mongo
    command: mongod --shardsvr --replSet mongors2 --dbpath /data/db --port 27017
    ports:
      - 27057:27017
    expose:
      - "27017"
    volumes:
      - ~/mongo_cluster/data5:/data/db

  mongors2n3:
    container_name: mongors2n3
    image: mongo
    command: mongod --shardsvr --replSet mongors2 --dbpath /data/db --port 27017
    ports:
      - 27067:27017
    expose:
      - "27017"

    volumes:
      - ~/mongo_cluster/data6:/data/db

  # mongo config server
  mongocfg1:
    container_name: mongocfg1
    image: mongo
    command: mongod --configsvr --replSet mongors1conf --dbpath /data/db --port 27017
    expose:
      - "27017"
    volumes:
      - ~/mongo_cluster/config1:/data/db

  mongocfg2:
    container_name: mongocfg2
    image: mongo
    command: mongod --configsvr --replSet mongors1conf --dbpath /data/db --port 27017
    expose:
      - "27017"
    volumes:
      - ~/mongo_cluster/config2:/data/db

  mongocfg3:
    container_name: mongocfg3
    image: mongo
    command: mongod --configsvr --replSet mongors1conf --dbpath /data/db --port 27017

    expose:
      - "27017"
    volumes:
      - ~/mongo_cluster/config3:/data/db

# mongos router
  mongos1:
    container_name: mongos1
    image: mongo
    depends_on:
      - mongocfg1
      - mongocfg2
    command: mongos --configdb mongors1conf/mongocfg1:27017,mongocfg2:27017,mongocfg3:27017 --port 27017
    ports:
      - 27019:27017
    expose:
      - "27017"

  mongos2:
    container_name: mongos2
    image: mongo
    depends_on:
      - mongocfg1
      - mongocfg2
    command: mongos --configdb mongors1conf/mongocfg1:27017,mongocfg2:27017,mongocfg3:27017 --port 27017
    ports:
      - 27020:27017
    expose:
      - "27017"
                                        




$ docker-compose up
                                        

Sharding Setup

$ docker exec -it mongocfg1 bash -c "echo 'rs.initiate({_id: "mongors1conf",configsvr: true, members: [{ _id : 0, host : "mongocfg1" },{ _id : 1, host : "mongocfg2" }, { _id : 2, host : "mongocfg3" }]})' | mongo"

# building replica shard
$ docker exec -it mongors1n1 bash -c "echo 'rs.initiate({_id : "mongors1", members: [{ _id : 0, host : "mongors1n1" },{ _id : 1, host : "mongors1n2" },{ _id : 2, host : "mongors1n3" }]})' | mongo"
$ docker exec -it mongors2n1 bash -c "echo 'rs.initiate({_id : "mongors2", members: [{ _id : 0, host : "mongors2n1" },{ _id : 1, host : "mongors2n2" },{ _id : 2, host : "mongors2n3" }]})' | mongo"


# we add shard to the routers
$ docker exec -it mongos1 bash -c "echo 'sh.addShard("mongors1/mongors1n1")' | mongo "
$ docker exec -it mongos1 bash -c "echo 'sh.addShard("mongors2/mongors2n1")' | mongo "
                                        

DB Setup

$ docker exec -it mongors1n1 bash -c "echo 'use testDb' | mongo"
$ docker exec -it mongors1n1 bash -c "echo 'db.createCollection(\"testDb.testCollection\")' | mongo "
$ docker exec -it mongos1 bash -c "echo 'sh.shardCollection(\"testDb.testCollection\", {\"shardingField\" : 1})' | mongo "
                                        

Playing around

#login to mongos1 (router)
$ docker exec -it mongos1 mongo

#force split chunks by sharding key value.
mongos> sh.splitAt('testDb.testCollection', {shardingField: 100});
mongos> db.testCollection.insertOne({shardingField: 55});
mongos> db.testCollection.insertOne({shardingField: 155});

#now we have 2 shards, each storing one chunk with 1 document
mongos> db.testCollection.insertOne({shardingField: 55})
{
	"acknowledged" : true,
	"insertedId" : ObjectId("61150783c8a0ff22b82d5ae6")
}

mongos> db.testCollection.insertOne({shardingField: 155})
{
	"acknowledged" : true,
	"insertedId" : ObjectId("61150788c8a0ff22b82d5ae7")
}

mongos> db.testCollection.getShardDistribution();

Shard mongors2 at mongors2/mongors2n1:27017,mongors2n2:27017,mongors2n3:27017
 data : 45B docs : 1 chunks : 1
 estimated data per chunk : 45B
 estimated docs per chunk : 1

Shard mongors1 at mongors1/mongors1n1:27017,mongors1n2:27017,mongors1n3:27017
 data : 45B docs : 1 chunks : 1
 estimated data per chunk : 45B
 estimated docs per chunk : 1

Totals
 data : 90B docs : 2 chunks : 2
 Shard mongors2 contains 50% data, 50% docs in cluster, avg obj size on shard : 45B
 Shard mongors1 contains 50% data, 50% docs in cluster, avg obj size on shard : 45B

mongos> sh.status()
--- Sharding Status --- 
  sharding version: {
  	"_id" : 1,
  	"minCompatibleVersion" : 5,
  	"currentVersion" : 6,
  	"clusterId" : ObjectId("6114bba388d545c510cfd551")
  }
  shards:
        {  "_id" : "mongors1",  "host" : "mongors1/mongors1n1:27017,mongors1n2:27017,mongors1n3:27017",  "state" : 1,  "topologyTime" : Timestamp(1628749545, 2) }
        {  "_id" : "mongors2",  "host" : "mongors2/mongors2n1:27017,mongors2n2:27017,mongors2n3:27017",  "state" : 1,  "topologyTime" : Timestamp(1628757106, 1) }
  ................
        {  "_id" : "testDb",  "primary" : "mongors1",  "partitioned" : true,  "version" : {  ...... }
                testDb.testCollection
                        shard key: { "shardingField" : 1 }
                        unique: false
                        balancing: true
                        chunks:
                                mongors1	1
                                mongors2	1
                        { "shardingField" : { "$minKey" : 1 } } -->> { "shardingField" : 100 } on : mongors2 Timestamp(2, 0) 
                        { "shardingField" : 100 } -->> { "shardingField" : { "$maxKey" : 1 } } on : mongors1 Timestamp(2, 1)

mongos> db.testCollection.find({});
{ "_id" : ObjectId("61150783c8a0ff22b82d5ae6"), "shardingField" : 55 }
{ "_id" : ObjectId("61150788c8a0ff22b82d5ae7"), "shardingField" : 155 }

# connect to one replica to verify what is stored (not recommended, just for testing purposes)
$ docker exec -it mongors1n1 mongo

mongors1:PRIMARY> use testDb
switched to db testDb
mongors1:PRIMARY> db.testCollection.find({})
{ "_id" : ObjectId("61150788c8a0ff22b82d5ae7"), "shardingField" : 155 }

# connect to the other replicaset
$ docker exec -it mongors2n1 mongo

mongors2:PRIMARY> use testDb
switched to db testDb
mongors2:PRIMARY> db.testCollection.find({})
{ "_id" : ObjectId("61150783c8a0ff22b82d5ae6"), "shardingField" : 55 }
                                        

Configuring Zones

mongos> sh.addShardToZone("mongors1", "testZone")
mongos> sh.addShardToZone("mongors2", "testZone")
mongos> sh.updateZoneKeyRange("testDb.testCollection", { shardingField: 20 }, { shardingField: 120 }, "testZone")
mongos> sh.status()
--- Sharding Status --- 
..................
shards:
        {  "_id" : "mongors1",  "host" : "mongors1/mongors1n1:27017,mongors1n2:27017,mongors1n3:27017",  "state" : 1,  "topologyTime" : Timestamp(1628749545, 2),  "tags" : [ "testZone" ] }
        {  "_id" : "mongors2",  "host" : "mongors2/mongors2n1:27017,mongors2n2:27017,mongors2n3:27017",  "state" : 1,  "topologyTime" : Timestamp(1628757106, 1),  "tags" : [ "testZone" ] }
.................
        databases:
.................
        {  "_id" : "testDb",  "primary" : "mongors1",  "partitioned" : true,  "version" : {  "uuid" : UUID("f1635ad8-d9b4-4ce6-9e3f-cef945fac96a"),  "timestamp" : Timestamp(1628749617, 1),  "lastMod" : 1 } }
                testDb.testCollection
                        shard key: { "shardingField" : 1 }
                        unique: false
                        balancing: true
                        chunks:
                                mongors1	2
                                mongors2	2
                        { "shardingField" : { "$minKey" : 1 } } -->> { "shardingField" : 20 } on : mongors2 Timestamp(2, 2) 
                        { "shardingField" : 20 } -->> { "shardingField" : 100 } on : mongors2 Timestamp(2, 3) 
                        { "shardingField" : 100 } -->> { "shardingField" : 120 } on : mongors1 Timestamp(2, 4) 
                        { "shardingField" : 120 } -->> { "shardingField" : { "$maxKey" : 1 } } on : mongors1 Timestamp(2, 5) 
                         tag: testZone  { "shardingField" : 20 } -->> { "shardingField" : 120 }