拓展阅读

ETL-10-apache SeaTunnel Connector v2 source mysql cdc

mysql binlog

说明

mysql cdc json 格式，发送到 neo4j 持久化。

但是执行一段时间以后，会报错，服务直接挂掉。

https://github.com/apache/seatunnel/issues/5694

https://github.com/apache/seatunnel/pull/5695

STREAMING 任务配置

  • cdc.conf
# Defining the runtime environment
env {
  # You can set flink configuration here
  parallelism = 1
  job.mode = "STREAMING"
  job.name = "allInOne-CDC-JSON-STREAMING"
  checkpoint.interval = 10000
}
source{
    MySQL-CDC {
        base-url = "jdbc:mysql://127.0.0.1:3306/test?useSSL=false&serverTimezone=Asia/Shanghai"
        driver = "com.mysql.jdbc.Driver"
        username = "admin"
        password = "123456"
        table-names = ["test.role", "test.role_extra"]

        startup.mode = "initial"
        format = compatible_debezium_json
        debezium = {
           # include schema into kafka message
           key.converter.schemas.enable = false
           value.converter.schemas.enable = false
           # include dd1
           include.schema.changes = false
           # topic.prefix
           database.server.name = "merge"
        }
        result_table_name="allInOne-CDC-JSON-result"
    }

}

transform {
    # If you would like to get more information about how to configure seatunnel and see full list of transform plugins,
    # please go to https://seatunnel.apache.org/docs/transform-v2/sql
}

sink {
    MysqlToNeo4j {
            source_table_name = "allInOne-CDC-JSON-result"

            uri = "bolt://localhost:7687"
            username = "neo4j"
            password = "12345678"
            database = "neo4j"

            max_transaction_retry_time = 30000
            max_connection_timeout = 30000

            format = compatible_debezium_json

            queryConfigList = [
                # 省略
            ]
    }
}

配置

  • seatunnel.yaml
seatunnel:
    engine:
        backup-count: 1
        queue-type: blockingqueue
        print-execution-info-interval: 60
        slot-service:
            dynamic-slot: true
        checkpoint:
            interval: 100
            timeout: 100
            storage:
                type: localfile
                max-retained: 3
                plugin-config:
                    namespace: C:\ProgramData\seatunnel\checkpoint\

#            storage:
#                type: hdfs
#                max-retained: 3
#                plugin-config:
#                    namespace: /tmp/seatunnel/checkpoint_snapshot/
#                    storage.type: hdfs
#                    fs.defaultFS: file:///tmp/

这两个参数比较小的时候，会报错 timeout。

checkpoint:
            interval: 100
            timeout: 100

报错

2024-02-21 10:49:07,695 INFO  org.apache.seatunnel.core.starter.seatunnel.command.ClientExecuteCommand - run shutdown hook because get close signal
2024-02-21 10:49:07,692 ERROR org.apache.seatunnel.core.starter.SeaTunnel - 

===============================================================================


2024-02-21 10:49:07,692 ERROR org.apache.seatunnel.core.starter.SeaTunnel - Fatal Error, 

2024-02-21 10:49:07,692 ERROR org.apache.seatunnel.core.starter.SeaTunnel - Please submit bug report in https://github.com/apache/seatunnel/issues

2024-02-21 10:49:07,692 ERROR org.apache.seatunnel.core.starter.SeaTunnel - Reason:SeaTunnel job executed failed 

2024-02-21 10:49:07,694 ERROR org.apache.seatunnel.core.starter.SeaTunnel - Exception StackTrace:org.apache.seatunnel.core.starter.exception.CommandExecuteException: SeaTunnel job executed failed
	at org.apache.seatunnel.core.starter.seatunnel.command.ClientExecuteCommand.execute(ClientExecuteCommand.java:191)
	at org.apache.seatunnel.core.starter.SeaTunnel.run(SeaTunnel.java:40)
	at org.apache.seatunnel.example.engine.cdc.MysqlCdcDefaultToJdbcMultiTablesExampleRefKeyNullV013.main(MysqlCdcDefaultToJdbcMultiTablesExampleRefKeyNullV013.java:47)
Caused by: org.apache.seatunnel.engine.common.exception.SeaTunnelEngineException: org.apache.seatunnel.engine.server.checkpoint.CheckpointException: Checkpoint expired before completing. Please increase checkpoint timeout in the seatunnel.yaml
	at org.apache.seatunnel.engine.server.checkpoint.CheckpointCoordinator.handleCoordinatorError(CheckpointCoordinator.java:255)
	at org.apache.seatunnel.engine.server.checkpoint.CheckpointCoordinator.lambda$null$9(CheckpointCoordinator.java:532)
	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
	at java.util.concurrent.FutureTask.run$$$capture(FutureTask.java:266)
	at java.util.concurrent.FutureTask.run(FutureTask.java)
	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180)
	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:750)

	at org.apache.seatunnel.engine.client.job.ClientJobProxy.waitForJobComplete(ClientJobProxy.java:122)
	at org.apache.seatunnel.core.starter.seatunnel.command.ClientExecuteCommand.execute(ClientExecuteCommand.java:184)
	... 2 more
 
2024-02-21 10:49:07,694 ERROR org.apache.seatunnel.core.starter.SeaTunnel - 
===============================================================================



Exception in thread "main" org.apache.seatunnel.core.starter.exception.CommandExecuteException: SeaTunnel job executed failed
	at org.apache.seatunnel.core.starter.seatunnel.command.ClientExecuteCommand.execute(ClientExecuteCommand.java:191)
	at org.apache.seatunnel.core.starter.SeaTunnel.run(SeaTunnel.java:40)
	at org.apache.seatunnel.example.engine.cdc.MysqlCdcDefaultToJdbcMultiTablesExampleRefKeyNullV013.main(MysqlCdcDefaultToJdbcMultiTablesExampleRefKeyNullV013.java:47)
Caused by: org.apache.seatunnel.engine.common.exception.SeaTunnelEngineException: org.apache.seatunnel.engine.server.checkpoint.CheckpointException: Checkpoint expired before completing. Please increase checkpoint timeout in the seatunnel.yaml
	at org.apache.seatunnel.engine.server.checkpoint.CheckpointCoordinator.handleCoordinatorError(CheckpointCoordinator.java:255)
	at org.apache.seatunnel.engine.server.checkpoint.CheckpointCoordinator.lambda$null$9(CheckpointCoordinator.java:532)
	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
	at java.util.concurrent.FutureTask.run$$$capture(FutureTask.java:266)
	at java.util.concurrent.FutureTask.run(FutureTask.java)
	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180)
	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:750)

	at org.apache.seatunnel.engine.client.job.ClientJobProxy.waitForJobComplete(ClientJobProxy.java:122)
	at org.apache.seatunnel.core.starter.seatunnel.command.ClientExecuteCommand.execute(ClientExecuteCommand.java:184)
	... 2 more

这个是任务在没有完成之前结束，Checkpoint expired before completing. Please increase checkpoint timeout in the seatunnel.yaml

那我调整大一些。

调整大一些

配置

  • seatunnel.yaml
seatunnel:
    engine:
        backup-count: 1
        queue-type: blockingqueue
        print-execution-info-interval: 60
        slot-service:
            dynamic-slot: true
        checkpoint:
            interval: 1000
            timeout: 1000
            storage:
                type: localfile
                max-retained: 3
                plugin-config:
                    namespace: C:\ProgramData\seatunnel\checkpoint\

#            storage:
#                type: hdfs
#                max-retained: 3
#                plugin-config:
#                    namespace: /tmp/seatunnel/checkpoint_snapshot/
#                    storage.type: hdfs
#                    fs.defaultFS: file:///tmp/

看源码：

    public static final Option<Integer> CHECKPOINT_INTERVAL =
            Options.key("interval")
                    .intType()
                    .defaultValue(300000)
                    .withDescription(
                            "The interval (in milliseconds) between two consecutive checkpoints.");

    public static final Option<Integer> CHECKPOINT_TIMEOUT =
            Options.key("timeout")
                    .intType()
                    .defaultValue(30000)
                    .withDescription("The timeout (in milliseconds) for a checkpoint.");

但是我们执行的是一个流任务这个单位应该是毫秒。

然后发现会有另外一个错误：


2024-02-21 11:04:22,713 WARN  org.apache.seatunnel.engine.server.dag.physical.SubPlan - Job allInOne-CDC-JSON-STREAMING-refkey-null-v0_13.conf (812519818174398465), Pipeline: [(1/1)] cancel error will retry
java.lang.InterruptedException: null
	at java.util.concurrent.CompletableFuture.reportGet(CompletableFuture.java:347) ~[?:1.8.0_371]
	at java.util.concurrent.CompletableFuture.get(CompletableFuture.java:1908) ~[?:1.8.0_371]
	at org.apache.seatunnel.engine.server.dag.physical.SubPlan.cancelPipelineTasks(SubPlan.java:461) ~[classes/:?]
	at org.apache.seatunnel.engine.server.dag.physical.SubPlan.cancelPipeline(SubPlan.java:417) ~[classes/:?]
	at org.apache.seatunnel.engine.server.dag.physical.SubPlan.handleCheckpointError(SubPlan.java:659) ~[classes/:?]
	at org.apache.seatunnel.engine.server.master.JobMaster.lambda$handleCheckpointError$2(JobMaster.java:341) ~[classes/:?]
	at java.util.ArrayList.forEach(ArrayList.java:1259) ~[?:1.8.0_371]
	at org.apache.seatunnel.engine.server.master.JobMaster.handleCheckpointError(JobMaster.java:338) ~[classes/:?]
	at org.apache.seatunnel.engine.server.checkpoint.CheckpointManager.handleCheckpointError(CheckpointManager.java:180) ~[classes/:?]
	at org.apache.seatunnel.engine.server.checkpoint.CheckpointCoordinator.handleCoordinatorError(CheckpointCoordinator.java:266) ~[classes/:?]
	at org.apache.seatunnel.engine.server.checkpoint.CheckpointCoordinator.lambda$null$9(CheckpointCoordinator.java:532) ~[classes/:?]
	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) [?:1.8.0_371]
	at java.util.concurrent.FutureTask.run$$$capture(FutureTask.java:266) [?:1.8.0_371]
	at java.util.concurrent.FutureTask.run(FutureTask.java) [?:1.8.0_371]
	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180) [?:1.8.0_371]
	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293) [?:1.8.0_371]
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) [?:1.8.0_371]
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) [?:1.8.0_371]
	at java.lang.Thread.run(Thread.java:750) [?:1.8.0_371]
2024-02-21 11:04:22,715 WARN  org.apache.seatunnel.engine.server.dag.physical.SubPlan - start cancel job Job allInOne-CDC-JSON-STREAMING-refkey-null-v0_13.conf (812519818174398465), Pipeline: [(1/1)] count = 1
2024-02-21 11:04:22,721 WARN  org.apache.seatunnel.engine.server.dag.physical.SubPlan - start cancel job Job allInOne-CDC-JSON-STREAMING-refkey-null-v0_13.conf (812519818174398465), Pipeline: [(1/1)] count = 0
2024-02-21 11:04:22,724 ERROR org.apache.seatunnel.engine.server.scheduler.PipelineBaseScheduler - scheduler Job allInOne-CDC-JSON-STREAMING-refkey-null-v0_13.conf (812519818174398465), Pipeline: [(1/1)] stop. Because org.apache.seatunnel.engine.common.exception.SchedulerNotAllowException: Job allInOne-CDC-JSON-STREAMING-refkey-null-v0_13.conf (812519818174398465), Pipeline: [(1/1)] turn to state CANCELING, skip SCHEDULED this pipeline.
	at org.apache.seatunnel.engine.server.scheduler.PipelineBaseScheduler.handlePipelineStateTurnError(PipelineBaseScheduler.java:370)
	at org.apache.seatunnel.engine.server.scheduler.PipelineBaseScheduler.schedulerPipeline(PipelineBaseScheduler.java:92)
	at org.apache.seatunnel.engine.server.scheduler.PipelineBaseScheduler.reSchedulerPipeline(PipelineBaseScheduler.java:362)
	at org.apache.seatunnel.engine.server.master.JobMaster.reSchedulerPipeline(JobMaster.java:395)
	at org.apache.seatunnel.engine.server.dag.physical.SubPlan.restorePipeline(SubPlan.java:580)
	at org.apache.seatunnel.engine.server.dag.physical.SubPlan.lambda$addPhysicalVertexCallBack$2(SubPlan.java:208)
	at java.util.concurrent.CompletableFuture.uniAccept(CompletableFuture.java:670)
	at java.util.concurrent.CompletableFuture$UniAccept.tryFire$$$capture(CompletableFuture.java:646)
	at java.util.concurrent.CompletableFuture$UniAccept.tryFire(CompletableFuture.java)
	at java.util.concurrent.CompletableFuture$Completion.run(CompletableFuture.java:456)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:750)

最大值的类别

看代码类别虽然是 long，但是实际上会做 int 的最大值校验。

所以最大值最多只能是 int.maxValue。

所以这个值应该怎么设置呢？

直接设置的非常大？

TODO 还是感觉没有特别理解这个参数的含义。

后来发现设置了最大值，依然报错，所以应该不是 checkpoint 的问题。

其他的问题

后来在 seatunnel 的 logs 下面，看到了

Caused by io.debezizium.DebeziziumException：Error reading MySQL variables: The last packet

大概意思就是读取信息超时了。

建议给 JDBC url 添加上对应的属性 autoReconnect=true 来解决这个问题。

mysql 自动断开

28800单位是秒转化成小时就是8小时，看出MySQL的默认设置，当一个连接的空闲时间超过8小时后，MySQL就会断开该连接。

所以发现问题出在如果超过这个wait_timeout时间(默认是8小时)对数据库没有任何操作，那么MySQL会自动关闭数据库连接以节省资源。

数据库连接自动断开的问题确实是在第二天发生了，也就是在一个晚上没有对数据库进行操作(显然超过了8小时)的情况下发生的这个问题。

show global variables like '%wait_timeout%';

对应的值：

+--------------------------+----------+
| Variable_name            | Value    |
+--------------------------+----------+
| innodb_lock_wait_timeout | 50       |
| lock_wait_timeout        | 31536000 |
| mysqlx_wait_timeout      | 28800    |
| wait_timeout             | 28800    |
+--------------------------+----------+
4 rows in set (0.01 sec)

这个测试环境可能设置的更加短，也不太好修改，应该是为了解决无用链接占用问题。

看了一下测试环境这个属性只有 1800，也就是 30min。

mysql 版本

SELECT VERSION();

发现本地是 5.7.34

5.7

看看官网 mysql5.7上关于 autoReconnect的设置：

27.8.20 C API Automatic Reconnection Control

The MySQL client library can perform an automatic reconnection to the server if it finds that the connection is down when you attempt to send a statement to the server to be executed. If auto-reconnect is enabled, the library tries once to reconnect to the server and send the statement again.

Auto-reconnect is disabled by default.

If it is important for your application to know that the connection has been dropped (so that it can exit or take action to adjust for the loss of state information), be sure that auto-reconnect is disabled. To ensure this, call mysql_options() with the MYSQL_OPT_RECONNECT option:

翻译：

27.8.20 C API自动重新连接控制

当你尝试发送一个语句给服务器执行时，如果MySQL客户端库发现连接断开，它可以自动重新连接到服务器。如果启用了自动重新连接，则库会尝试一次重新连接到服务器并重新发送该语句。

自动重新连接默认情况下是禁用的。

如果你的应用程序需要知道连接已经断开（以便它可以退出或采取行动来调整丢失的状态信息），请确保禁用自动重新连接。为了确保这一点，使用MYSQL_OPT_RECONNECT选项调用mysql_options()。

解决方式

5.0 以上：autoReconnect=true

这个要注意一下数据库版本。

8.3 https://dev.mysql.com/doc/c-api/8.3/en/c-api-auto-reconnect.html

jdbc:mysql://127.0.0.1:3306/stock_tweet?autoReconnect=true

这个参数表示在mysql超时断开连接后会自动重新连接，配置的话，只需要在连接mysql的语句写上autoReconnect=true：

同时可以看到官网不推荐使用这个参数，因为它有一些副作用，具体介绍下：

  • 原有连接上的事务将会被回滚，事务的提交模式将会丢失；

  • 原有连接持有的表的锁将会全部释放；

  • 原有连接关联的会话Session将会丢失，重新恢复的连接关联的将会是一个新的会话Session；

  • 原有连接定义的用户变量将会丢失；

  • 原有连接定义的预编译SQL将会丢失；

  • 原有连接失效，新的连接恢复后，MySQL将会使用新的记录行来存储连接中的性能数据；

官方属性说明

官方 mysql 文档

  • 自动重新连接 autoReconnect

驱动程序是否应尝试重新建立陈旧和/或失效的连接？如果启用，则驱动程序将对在当前事务中属于陈旧或失效连接的查询引发异常，但会在下一个在新事务中发出的连接上的查询之前尝试重新连接。

不推荐使用此功能，因为当应用程序未正确处理SQLExceptions时，它会产生与会话状态和数据一致性相关的副作用，并且只设计用于在无法配置应用程序以正确处理由失效和陈旧连接引起的SQLExceptions时使用。

或者，作为最后一种选择，可以调查将MySQL服务器变量’wait_timeout’设置为较高值，而不是默认的8小时。

默认值：false 自版本：1.1

  • autoReconnectForPools

是否使用适合连接池的重新连接策略？

默认值：false 自版本：3.1.3

  • failOverReadOnly

在“autoReconnect”模式下故障转移时，连接是否应设置为“只读”？

默认值：true 自版本：3.0.12

  • maxReconnects

如果“autoReconnect”为“true”，尝试重新连接的最大次数。

默认值：3 自版本：1.1

  • reconnectAtTxEnd

如果将“autoReconnect”设置为“true”，是否应在每个事务结束时尝试重新连接？

默认值：false 自版本：3.0.10

  • initialTimeout

如果启用了“autoReconnect”，重新连接尝试之间的初始等待时间（以秒为单位，默认为“2”）。

默认值：2 自版本：1.1

更多的属性调整

连接数据库超时设置autoReconnect=true，默认重试次数调整

参数 说明

autoReconnect 驱动程序是否应该尝试重新建立连接? 如果启用了，驱动程序将会对陈旧或死亡连接上发出的查询抛出异常，这些查询属于当前事务，但是在新事务中对连接发出的下一个查询之前会尝试重新连接。 不推荐这个功能的使用,因为它有副作用与会话状态和数据一致性当应用程序不妥善处理异常,仅设计用于当你无法配置您的应用程序来处理异常造成死亡和陈旧的正确连接。 另外，作为最后一个选项，研究将MySQL服务器变量“wait_timeout”设置为一个较高的值，而不是默认的8小时。 maxReconnects 如果autoReconnect为true，则尝试重新连接的最大次数，默认为’3’。 initialTimeout 如果启用了autoReconnect，重新连接尝试之间的初始等待时间(以秒为单位，默认为’2’)。

连接池文档

MySQL Connector/J Documentation

测试

我们设置一下 2 个属性：

autoReconnect=true&initialTimeout=60

怀疑是超时时间可能不够，导致的问题，

如下：

source{
    MySQL-CDC {
        base-url = "jdbc:mysql://127.0.0.1:3306/test?autoReconnect=true&initialTimeout=60&useSSL=false&serverTimezone=Asia/Shanghai&autoReconnect=true"
        driver = "com.mysql.jdbc.Driver"
        username = "admin"
        password = "123456"
        table-names = ["test.role", "test.role_extra"]

        startup.mode = "initial"
        format = compatible_debezium_json
        debezium = {
           # include schema into kafka message
           key.converter.schemas.enable = false
           value.converter.schemas.enable = false
           # include dd1
           include.schema.changes = false
           # topic.prefix
           database.server.name = "merge"
        }
        result_table_name="allInOne-CDC-JSON-result"
    }

}

给这里的 cdc url 加上 autoReconnect=true 属性。

然后发现不行，这里引发出了另一个问题。

我们的另一个 cdc 就不存在这个样的问题。

NEXT

学习一下 checkpoint 的源码？

这部分的作用到底是什么？

参考资料

mysql connect 5.6_mysql5，mysql5.6，mysql5.7，mysql8设置autoReconnect=true解决连接数据库超时问m.

https://github.com/apache/seatunnel/issues/5777

https://github.com/apache/seatunnel/issues/6013

https://github.com/apache/seatunnel/issues/5555

https://github.com/apache/seatunnel/issues/5694

https://www.51cto.com/article/707891.html