Canal ha mode configuration

Canal HA Mode configuration

Server side HA Mode configuration

canal It's supporting HA Of , In fact, the implementation mechanism also depends on zookeeper To achieve , The features used are watcher and EPHEMERAL node ( and session Life cycle binding ), And HDFS Of HA similar .

canal Of ha In two parts ,canal server and canal client There are corresponding ha Realization

• canal server: In order to reduce the mysql dump Request , Different server Upper instance( Different server The same on instance) You can only be at the same time running, The others are in standby state (standby yes instance The state of ).
• canal client: In order to ensure order , One copy instance Only one at a time canal client Conduct get/ack/rollback operation , Otherwise, the order of client receiving cannot be guaranteed .

Machine preparation

• function canal Machine ： node1 , node2
• zookeeper The address is node1:2181,node2:2181,node3:2181
• mysql Address ：node1:3306

Canal HA Server configuration

According to deployment and configuration , Complete the configuration on a single machine , Demonstration time instance name by example

modify canal.properties, add zookeeper To configure

canal.zkServers=node1:2181,node2,node3
canal.instance.global.spring.xml = classpath:spring/default-instance.xml

establish example Catalog , And modify it instance.properties

canal.instance.mysql.slaveId = 1234 ## The other machine is changed to 1235, Guarantee slaveId Don't repeat 
canal.instance.master.address = node1:3306

Be careful ： On two machines instance The names of the directories need to be exactly the same ,HA Patterns depend on instance name Conduct management , At the same time, we must all choose default-instance.xml To configure

Start the... Of both machines canal

-------
ssh node1
sh bin/startup.sh
--------
ssh node2
sh bin/startup.sh

After starting , You can see logs/example/example.log, You will only see a successful startup log on one machine .

For example, the successful startup here is node1

2013-03-19 18:18:20.590 [main] INFO  c.a.o.c.i.spring.support.PropertyPlaceholderConfigurer - Loading properties file from class path resource [canal.properties]
2013-03-19 18:18:20.596 [main] INFO  c.a.o.c.i.spring.support.PropertyPlaceholderConfigurer - Loading properties file from class path resource [example/instance.properties]
2013-03-19 18:18:20.831 [main] INFO  c.a.otter.canal.instance.spring.CanalInstanceWithSpring - start CannalInstance for 1-example 
2013-03-19 18:18:20.845 [main] INFO  c.a.otter.canal.instance.spring.CanalInstanceWithSpring - start successful...

Check it out. zookeeper Node information in , You can also know that the current working node is node1:11111

[zk: localhost:2181(CONNECTED) 15] get /otter/canal/destinations/example/running  
{
    "active":true,"address":"192.168.88.120:11111","cid":1}

Client link , Consumption data

1、 You can specify zookeeper Address and instance name,canal client Automatically from zookeeper Medium running node , Get the working node of the current service , And then build a link with it ：

CanalConnector connector = CanalConnectors.newClusterConnector("node1:2181", "example", "canal", "canal");

2、 After the link is successful ,canal server Will record what is currently working canal client Information , For example, client ip, Port information of link etc

[zk: localhost:2181(CONNECTED) 17] get /otter/canal/destinations/example/1001/running
{
    "active":true,"address":"192.168.88.1:60957","clientId":1001}

3、 After the success of data consumption ,canal server Will be in zookeeper The last successful consumption is recorded in binlog site . ( Next restart client when , Will continue to consume from this last point )

[zk: localhost:2181(CONNECTED) 16] get /otter/canal/destinations/example/1001/cursor
{
    "@type":"com.alibaba.otter.canal.protocol.position.LogPosition","identity":{
    "slaveId":-1,"sourceAddress":{
    "address":"node1","port":3306}},"postion":{
    "included":false,"journalName":"mysql-bin.000025","position":51944693,"serverId":1,"timestamp":1575424854000}}

Restart it canal server

Stop working node1 Of canal server

ssh node1 
sh bin/stop.sh

At this time node2 It'll start right away example instance, Provide new data services

[zk: localhost:2181(CONNECTED) 19] get /otter/canal/destinations/example/running
{"active":true,"address":"192.168.88.121:11111","cid":1}

meanwhile , The client will follow canal server Handoff , By acquiring zookeeper The latest address in , With the new canal server Building links , Continue to consume data , The whole process is done automatically

Canal Server HA Flow chart of

1. canal server To start something canal instance Always go first to zookeeper Make an attempt to start the judgment ( Realization ： establish EPHEMERAL node , Whoever creates successfully will be allowed to start )
2. establish zookeeper After node success , Corresponding canal server Start the corresponding canal instance, No successful canal instance Will be in standby state
3. once zookeeper Find out canal server A After the created node disappears , Inform the others immediately canal server Step again 1 The operation of , Choose a new one canal server start-up instance.
4. canal client Every time connect when , First of all zookeeper Ask who is currently starting canal instance, And then link to it , Once the link is not available , Will try again connect.

[ Failed to transfer the external chain picture , The origin station may have anti-theft chain mechanism , It is suggested to save the pictures and upload them directly (img-smQFGnoR-1613006260310)(F:/ Project video / Actual combat of shucang project - Dual real-time part / Real time data warehouse / Hundreds of billions of dollars of data warehouse 02 God /1. The notes /assets/image-20200214112119544.png)]

client HA Mode configuration

Canal Client And canal server Similar way , It's also the use of zookeeper To seize EPHEMERAL The way nodes control .

HA The implementation of the , The client is ClientRunningMonitor, The server is ServerRunningMonitor.

Take several directly Canal client , If it starts at the same time , Only one client can be accessed from Canal The server gets binlog news , Other clients cannot pull to binlog news .

From the run configuration , Copy the same configuration . Then start running .

[ Failed to transfer the external chain picture , The origin station may have anti-theft chain mechanism , It is suggested to save the pictures and upload them directly (img-uN2GtWMl-1613006260312)(F:/ Project video / Actual combat of shucang project - Dual real-time part / Real time data warehouse / Hundreds of billions of dollars of data warehouse 02 God /1. The notes /assets/image-20191204100238491.png)]

Close the previous client , Then wait a little , Check whether another client can get binlog Log message .

Client1 Log ：

****************************************************
* Batch Id: [3] ,count : [3] , memsize : [198] , Time : 2017-10-12 17:59:59
* Start : [mysql-bin.000004:1656:1507802398000(2017-10-12 17:59:58)] 
* End : [mysql-bin.000004:1831:1507802398000(2017-10-12 17:59:58)] 
****************************************************

================> binlog[mysql-bin.000004:1656] , executeTime : 1507802398000 , delay : 1188ms
 BEGIN ----> Thread id: 768
----------------> binlog[mysql-bin.000004:1782] , name[canal_test,test] , eventType : UPDATE , executeTime : 1507802398000 , delay : 1199ms
uid : 1    type=int(4)
name : zqh    type=varchar(10)    update=true
----------------
 END ----> transaction id: 0
================> binlog[mysql-bin.000004:1831] , executeTime : 1507802398000 , delay : 1236ms
## stop the canal client## canal client is down.

stop it Client1 after ,Client2 Log ：

****************************************************
* Batch Id: [4] ,count : [3] , memsize : [198] , Time : 2017-10-12 18:02:15
* Start : [mysql-bin.000004:1906:1507802534000(2017-10-12 18:02:14)] 
* End : [mysql-bin.000004:2081:1507802534000(2017-10-12 18:02:14)] 
****************************************************

================> binlog[mysql-bin.000004:1906] , executeTime : 1507802534000 , delay : 1807ms
 BEGIN ----> Thread id: 768
----------------> binlog[mysql-bin.000004:2032] , name[canal_test,test] , eventType : UPDATE , executeTime : 1507802534000 , delay : 1819ms
uid : 1    type=int(4)
name : zqhx    type=varchar(10)    update=true
----------------
 END ----> transaction id: 0
================> binlog[mysql-bin.000004:2081] , executeTime : 1507802534000 , delay : 1855ms

Observe ZK In nodes instance Corresponding client node , stay Client When switching , Changes will be made .
For example, the following clients are from 56806 The port is switched to 56842 port .
Shut down all clients ,1001 Under no running. Express instance No client consumption binlog 了 .

 Start two clients , The first client （56806） Running 
[zk: 192.168.6.52:2181(CONNECTED) 29] get /otter/canal/destinations/example/1001/running
{"active":true,"address":"10.57.241.44:56806","clientId":1001}

 Stop the first client , Delete node 
[zk: 192.168.6.52:2181(CONNECTED) 30] get /otter/canal/destinations/example/1001/running
Node does not exist: /otter/canal/destinations/example/1001/running

 The second client （56842） Become lord 
[zk: 192.168.6.52:2181(CONNECTED) 31] get /otter/canal/destinations/example/1001/running
{"active":true,"address":"10.57.241.44:56842","clientId":1001}

[zk: 192.168.6.52:2181(CONNECTED) 32] ls /otter/canal/destinations/example/1001
[cursor]

The specific implementation related classes are ：ClientRunningMonitor/ClientRunningListener/ClientRunningData.

client running Related control , Mainly to solve client Self failover Mechanism .
canal client It is allowed to start multiple at the same time canal client,
adopt running Mechanism , There is only one guarantee client at work , other client As a cold standby .
When in operation client Hang up ,running Control the cold standby client Switch to working mode ,
This will ensure that canal client It will not be a single point . Ensure the high availability of the whole system .

The left side of the figure below is the client side HA Realization , On the right is the server HA Realization

ilover Mechanism .
canal client It is allowed to start multiple at the same time canal client,
adopt running Mechanism , There is only one guarantee client at work , other client As a cold standby .
When in operation client Hang up ,running Control the cold standby client Switch to working mode ,
This will ensure that canal client It will not be a single point . Ensure the high availability of the whole system .

The left side of the figure below is the client side HA Realization , On the right is the server HA Realization

[ Failed to transfer the external chain picture , The origin station may have anti-theft chain mechanism , It is suggested to save the pictures and upload them directly (img-vDw20GjK-1613006260314)(F:/ Project video / Actual combat of shucang project - Dual real-time part / Real time data warehouse / Hundreds of billions of dollars of data warehouse 02 God /1. The notes /assets/image-20200214112840115.png)]