Replica Set
今天來學習如何透過 docker compose 建立出一個 MongoDB Replica Set 這次將使用 play-with-docker 提供的 Instance 來建立我們環境 我們建立三台 Instance 來模擬三台虛擬機的情況
- node1 192.168.0.8
- node2 192.168.0.7
- node3 192.168.0.6
首先先分別把這三台 Instance 的 IP 加入到各自 /etc/hosts 檔案內
#/etc/hosts
192.168.0.8 node1
192.168.0.7 node2
192.168.0.6 node3
建立完成後可以使用 ping 命令進行測試
接下來在這三台 Instance 建立 docker-compose.yaml 以及啟動 mongodb
mkdir -p mongo && cd $_
vi docker-compose.yml
# docker-compose.yml
services:
mongo:
image: mongo
restart: always
command: [ "mongod", "--bind_ip_all", "--replSet", "dbrs" ]
network_mode: "host"
在上一篇的文章中，我們知道 mongodb Dockerfile 預設是運行 "docker-entrypoint.sh mongod"
在今天的設定檔中我們修改了預設的 command 額外添加了幾個參數 --bind_ip_all、--replSet
我們先看一下官方預設的 mongod.conf 設定檔 Github，的兩個設定值
# network interfaces
net:
port: 27017
bindIp: 127.0.0.1 # Enter 0.0.0.0,:: to bind to all IPv4 and IPv6 addresses or, alternatively, use the net.bindIpAll setting.
#replication:
這邊 net 設定值只會監聽 127.0.0.1:27017，我們必須把IP 修改成 0.0.0.0:27017，一種作法是另外提供一個設定檔另一種是使用 bind_ip_all 參數也可以達到同樣的效果
並且預設沒有 replication 相關的設定，這裡我們也使用 --replSet 參數指定 Replica Set 的名稱
接下來我們分別在這三台 Instance 輸入命令 docker compose up -d 啟動 container
- 0df0212f690b node1
- 133a172f223d node2
- 772fe1388842 node3
都正常啟動之後我們選擇 node1 做為 primary node，使用命令進入 node1 mongodb container 內
docker exec -it 0df0212f690b mongosh
使用 initiate 方法初始化 Replica Set
rs.initiate()
成功後會發現我們目前位於 primary 節點上
test> rs.initiate()
{
info2: 'no configuration specified. Using a default configuration for the set',
me: 'node1:27017',
ok: 1
}
dbrs [direct: other] test>
dbrs [direct: primary] test>
不過目前並沒有其他節點我們還需要將其他兩台 Instance 添加進來
rs.add()
dbrs [direct: primary] test> rs.add("node2:27017")
{
ok: 1,
'$clusterTime': {
clusterTime: Timestamp({ t: 1679838515, i: 1 }),
signature: {
hash: Binary(Buffer.from("0000000000000000000000000000000000000000", "hex"), 0),
keyId: Long("0")
}
},
operationTime: Timestamp({ t: 1679838515, i: 1 })
}
dbrs [direct: primary] test> rs.add("node3:27017")
{
ok: 1,
'$clusterTime': {
clusterTime: Timestamp({ t: 1679838527, i: 1 }),
signature: {
hash: Binary(Buffer.from("0000000000000000000000000000000000000000", "hex"), 0),
keyId: Long("0")
}
},
operationTime: Timestamp({ t: 1679838527, i: 1 })
}
添加成功後我們可以是使用命令 rs.status() 檢查 Replica Set 的運作情況
dbrs [direct: primary] test> rs.status()
{
set: 'dbrs',
...
members: [
{
_id: 0,
name: 'node1:27017',
health: 1,
state: 1,
stateStr: 'PRIMARY',
uptime: 455,
optime: { ts: Timestamp({ t: 1679838570, i: 1 }), t: Long("1") },
optimeDate: ISODate("2023-03-26T13:49:30.000Z"),
lastAppliedWallTime: ISODate("2023-03-26T13:49:30.004Z"),
lastDurableWallTime: ISODate("2023-03-26T13:49:30.004Z"),
syncSourceHost: '',
syncSourceId: -1,
infoMessage: '',
electionTime: Timestamp({ t: 1679838349, i: 2 }),
electionDate: ISODate("2023-03-26T13:45:49.000Z"),
configVersion: 5,
configTerm: 1,
self: true,
lastHeartbeatMessage: ''
},
{
_id: 1,
name: 'node2:27017',
health: 1,
state: 2,
stateStr: 'SECONDARY',
uptime: 60,
optime: { ts: Timestamp({ t: 1679838570, i: 1 }), t: Long("1") },
optimeDurable: { ts: Timestamp({ t: 1679838570, i: 1 }), t: Long("1") },
optimeDate: ISODate("2023-03-26T13:49:30.000Z"),
optimeDurableDate: ISODate("2023-03-26T13:49:30.000Z"),
lastAppliedWallTime: ISODate("2023-03-26T13:49:30.004Z"),
lastDurableWallTime: ISODate("2023-03-26T13:49:30.004Z"),
lastHeartbeat: ISODate("2023-03-26T13:49:34.002Z"),
lastHeartbeatRecv: ISODate("2023-03-26T13:49:33.999Z"),
pingMs: Long("0"),
lastHeartbeatMessage: '',
syncSourceHost: 'node1:27017',
syncSourceId: 0,
infoMessage: '',
configVersion: 5,
configTerm: 1
},
{
_id: 2,
name: 'node3:27017',
health: 1,
state: 2,
stateStr: 'SECONDARY',
uptime: 47,
optime: { ts: Timestamp({ t: 1679838570, i: 1 }), t: Long("1") },
optimeDurable: { ts: Timestamp({ t: 1679838570, i: 1 }), t: Long("1") },
optimeDate: ISODate("2023-03-26T13:49:30.000Z"),
optimeDurableDate: ISODate("2023-03-26T13:49:30.000Z"),
lastAppliedWallTime: ISODate("2023-03-26T13:49:30.004Z"),
lastDurableWallTime: ISODate("2023-03-26T13:49:30.004Z"),
lastHeartbeat: ISODate("2023-03-26T13:49:34.002Z"),
lastHeartbeatRecv: ISODate("2023-03-26T13:49:34.526Z"),
pingMs: Long("0"),
lastHeartbeatMessage: '',
syncSourceHost: 'node2:27017',
syncSourceId: 1,
infoMessage: '',
configVersion: 5,
configTerm: 1
}
]
...
}
這邊確認 stateStr 的值顯示為 'PRIMARY' 和 'SECONDARY' 就代表成功了
我們嘗試在主節點上添加一筆資料，看看會不會自動複製資料
db.test.insertOne({"node": "node1"})
dbrs [direct: primary] test> db.test.insertOne({"node": "node1"})
{
acknowledged: true,
insertedId: ObjectId("64204eb44839d618edd828f1")
}
接下來我們到 node2 或 node3 上面看看資料
docker exec -it 133a172f223d mongosh
terminal 會顯示我們目前的節點為 secondary
dbrs [direct: secondary] test>
如果我們這邊直接輸入 db.test.find() 會發生報錯
dbrs [direct: secondary] test> db.test.find()
MongoServerError: not primary and secondaryOk=false - consider using db.getMongo().setReadPref() or readPreference in the connection string
是因為 secondary 節點沒有開放讀取，所以我們需要先調整設定值
db.getMongo().setReadPref('primaryPreferred')
就可以在 secondary 節點讀取到我們剛剛插入的資料了
dbrs [direct: secondary] test> db.test.find()
[ { _id: ObjectId("64204eb44839d618edd828f1"), node: 'node1' } ]
最後我們回到 node1 ，直接將 node1 的 mongodb 服務關閉模擬異常情況，看看會不會進行自動轉移
[node1] (local) root@192.168.0.8 ~/mongo
$ docker compose down
[+] Running 1/1
⠿ Container mongo-mongo-1 Removed
再回到 node2 使用命令 rs.status() 進行檢查
dbrs [direct: primary] test> rs.status()
{
set: 'dbrs',
...
},
members: [
{
_id: 0,
name: 'node1:27017',
health: 0,
state: 8,
stateStr: '(not reachable/healthy)',
uptime: 0,
optime: { ts: Timestamp({ t: 0, i: 0 }), t: Long("-1") },
optimeDurable: { ts: Timestamp({ t: 0, i: 0 }), t: Long("-1") },
optimeDate: ISODate("1970-01-01T00:00:00.000Z"),
optimeDurableDate: ISODate("1970-01-01T00:00:00.000Z"),
lastAppliedWallTime: ISODate("2023-03-26T14:01:22.573Z"),
lastDurableWallTime: ISODate("2023-03-26T14:01:22.573Z"),
lastHeartbeat: ISODate("2023-03-26T14:01:54.652Z"),
lastHeartbeatRecv: ISODate("2023-03-26T14:01:31.584Z"),
pingMs: Long("0"),
lastHeartbeatMessage: 'Error connecting to node1:27017 (192.168.0.8:27017) :: caused by :: Connection refused',
syncSourceHost: '',
syncSourceId: -1,
infoMessage: '',
configVersion: 5,
configTerm: 2
},
{
_id: 1,
name: 'node2:27017',
health: 1,
state: 1,
stateStr: 'PRIMARY',
uptime: 1207,
optime: { ts: Timestamp({ t: 1679839310, i: 4 }), t: Long("2") },
optimeDate: ISODate("2023-03-26T14:01:50.000Z"),
lastAppliedWallTime: ISODate("2023-03-26T14:01:50.405Z"),
lastDurableWallTime: ISODate("2023-03-26T14:01:50.405Z"),
syncSourceHost: '',
syncSourceId: -1,
infoMessage: '',
electionTime: Timestamp({ t: 1679839282, i: 1 }),
electionDate: ISODate("2023-03-26T14:01:22.000Z"),
configVersion: 5,
configTerm: 2,
self: true,
lastHeartbeatMessage: ''
},
{
_id: 2,
name: 'node3:27017',
health: 1,
state: 2,
stateStr: 'SECONDARY',
uptime: 787,
optime: { ts: Timestamp({ t: 1679839310, i: 4 }), t: Long("2") },
optimeDurable: { ts: Timestamp({ t: 1679839310, i: 4 }), t: Long("2") },
optimeDate: ISODate("2023-03-26T14:01:50.000Z"),
optimeDurableDate: ISODate("2023-03-26T14:01:50.000Z"),
lastAppliedWallTime: ISODate("2023-03-26T14:01:50.405Z"),
lastDurableWallTime: ISODate("2023-03-26T14:01:50.405Z"),
lastHeartbeat: ISODate("2023-03-26T14:01:54.595Z"),
lastHeartbeatRecv: ISODate("2023-03-26T14:01:54.600Z"),
pingMs: Long("0"),
lastHeartbeatMessage: '',
syncSourceHost: 'node2:27017',
syncSourceId: 1,
infoMessage: '',
configVersion: 5,
configTerm: 2
}
],
...
}
發現我們的 node1 目前狀態為 (not reachable/healthy)，並且主節點已經成功切換到 node2 上了
Summary
今天測試了如何在 mongodb 建立 Replica Set，增加資料的安全性避免單一節點故障導致服務下線或者資料遺失