You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user-zh@flink.apache.org by "yinghua_zh@163.com" <yi...@163.com> on 2021/01/14 10:02:59 UTC

回复: 回复: 请教个Flink checkpoint的问题

代码如下:
    streamEnv.enableCheckpointing(5 * 60 * 1000);
    CheckpointConfig checkPointConfig = streamEnv.getCheckpointConfig();
    checkPointConfig.setCheckpointingMode(CheckpointingMode.EXACTLY_ONCE);
    checkPointConfig.setCheckpointTimeout(1 * 60 * 1000);
    checkPointConfig.setMinPauseBetweenCheckpoints((1 * 30 * 1000));
    checkPointConfig.setMaxConcurrentCheckpoints(1);
    checkPointConfig.setTolerableCheckpointFailureNumber(3);
    checkPointConfig
            .enableExternalizedCheckpoints(ExternalizedCheckpointCleanup.RETAIN_ON_CANCELLATION);

    String checkpointPath = this.execOptions.get(CHECKPOINT_PATH);
    try {
      StateBackend rocksDBStateBackend = new RocksDBStateBackend(checkpointPath);
      streamEnv.setStateBackend(rocksDBStateBackend);



yinghua_zh@163.com
 
发件人: Evan
发送时间: 2021-01-14 17:55
收件人: user-zh
主题: 回复: 请教个Flink checkpoint的问题
代码图挂掉了,看不到代码
 
 
 
发件人: yinghua_zh@163.com
发送时间: 2021-01-14 17:26
收件人: user-zh
主题: 请教个Flink checkpoint的问题
我在yarn上提交任务时,设置flink的checkpoint是5分钟一次,checkpoint使用RocksDBStateBackend保存在HDFS上且任务取消后不删除checkpoint,代码如下
现在我发现,当我停止任务时使用stopWithSavepoint发现任务停止后把最近一次的checkpoint信息给删除了?目前机制是这样的吗还是我使用有问题?是不是调用cancelWithSavepoint停止任务时就不会删除最近一次的checkpoint信息?
 
 
yinghua_zh@163.com

Re: 回复: 请教个Flink checkpoint的问题

Posted by 赵一旦 <hi...@gmail.com>.
Evan说的这个是一个设置,但也仅影响cancel那个命令,stop还是会删除。这个点其实做的不是很好,不清楚为啥,之前Q过,没人鸟。。。
所以按照我的经验,如果是需要停止并基于保存点重启,那还好。如果计划基于检查点重启,无比提前备份检查点,然后停任务,然后复制备份回去。
在或者,直接cancel,不用stop。

Evan <ch...@foxmail.com> 于2021年1月14日周四 下午6:49写道:

> 是的,应该是机制问题,链接[1]打开有这样一句解释:
>
>  If you choose to retain externalized checkpoints on cancellation you have
> to handle checkpoint clean up manually when you cancel the job as well
> (terminating with job status JobStatus#CANCELED).
>
> [1]
> https://ci.apache.org/projects/flink/flink-docs-release-1.12/zh/deployment/config.html#execution-checkpointing-externalized-checkpoint-retention
>
> 如回答有误,请指正。
>
>
>
>
>
>
> 发件人: yinghua_zh@163.com
> 发送时间: 2021-01-14 18:02
> 收件人: user-zh
> 主题: 回复: 回复: 请教个Flink checkpoint的问题
> 代码如下:
>     streamEnv.enableCheckpointing(5 * 60 * 1000);
>     CheckpointConfig checkPointConfig = streamEnv.getCheckpointConfig();
>     checkPointConfig.setCheckpointingMode(CheckpointingMode.EXACTLY_ONCE);
>     checkPointConfig.setCheckpointTimeout(1 * 60 * 1000);
>     checkPointConfig.setMinPauseBetweenCheckpoints((1 * 30 * 1000));
>     checkPointConfig.setMaxConcurrentCheckpoints(1);
>     checkPointConfig.setTolerableCheckpointFailureNumber(3);
>     checkPointConfig
>
> .enableExternalizedCheckpoints(ExternalizedCheckpointCleanup.RETAIN_ON_CANCELLATION);
>
>     String checkpointPath = this.execOptions.get(CHECKPOINT_PATH);
>     try {
>       StateBackend rocksDBStateBackend = new
> RocksDBStateBackend(checkpointPath);
>       streamEnv.setStateBackend(rocksDBStateBackend);
>
>
>
> yinghua_zh@163.com
>
> 发件人: Evan
> 发送时间: 2021-01-14 17:55
> 收件人: user-zh
> 主题: 回复: 请教个Flink checkpoint的问题
> 代码图挂掉了,看不到代码
>
>
>
> 发件人: yinghua_zh@163.com
> 发送时间: 2021-01-14 17:26
> 收件人: user-zh
> 主题: 请教个Flink checkpoint的问题
>
> 我在yarn上提交任务时,设置flink的checkpoint是5分钟一次,checkpoint使用RocksDBStateBackend保存在HDFS上且任务取消后不删除checkpoint,代码如下
>
> 现在我发现,当我停止任务时使用stopWithSavepoint发现任务停止后把最近一次的checkpoint信息给删除了?目前机制是这样的吗还是我使用有问题?是不是调用cancelWithSavepoint停止任务时就不会删除最近一次的checkpoint信息?
>
>
> yinghua_zh@163.com
>

Re:Re: Re: 请教个Flink checkpoint的问题

Posted by 邮件帮助中心 <yi...@163.com>.
1:先前测试过使用stopWithSavepoint时会将之前成功保存的checkpoint数据给删除掉,后来我们查看了下源码,里面描述如下,就是调用该方法时Flink会将程序设置成Finished态的,可能和实际使用场景有出入。
/**
 * Stops a program on Flink cluster whose job-manager is configured in this client's configuration.
 * Stopping works only for streaming programs. Be aware, that the program might continue to run for
 * a while after sending the stop command, because after sources stopped to emit data all operators
 * need to finish processing.
 *
 * @param jobId the job ID of the streaming program to stop
 * @param advanceToEndOfEventTime flag indicating if the source should inject a {@code MAX_WATERMARK} in the pipeline
 * @param savepointDirectory directory the savepoint should be written to
 * @return a {@link CompletableFuture} containing the path where the savepoint is located
 */

CompletableFuture<String> stopWithSavepoint(final JobID jobId, final boolean advanceToEndOfEventTime, @Nullable final String savepointDirectory);

















在 2021-01-17 16:48:22,"Congxian Qiu" <qc...@gmail.com> 写道:
>Hi
>    你可以看一下是不是 作业的状态变成了 FINISH 了,现在 Flink 在 FINISH 的时候,会删除 checkpoint(就算设置了
>retain on cancel 也会删除)
>    PS :FLINK-18263 正在讨论是否在 Job 变为 FINISH 状态后保留 checkpoint
>[1] https://issues.apache.org/jira/browse/FLINK-18263
>Best,
>Congxian
>
>
>yinghua_zh@163.com <yi...@163.com> 于2021年1月15日周五 上午11:23写道:
>
>> 感谢您的答复,今天我使用cancelWithSavepoint时也还是删除了,通过你的答复知道机制如此,我再使用另外的参数来观察下。
>>
>>
>>
>> yinghua_zh@163.com
>>
>> 发件人: Yun Tang
>> 发送时间: 2021-01-15 11:02
>> 收件人: user-zh
>> 主题: Re: 回复: 请教个Flink checkpoint的问题
>> Hi
>>
>> 这是因为在Flink-1.7.0 之后,savepoint也被当做是retained checkpoint了 [1],当你stop with
>> savepoint 成功时,新的savepoint创建之后,旧的checkpoint因为默认retain
>> checkpoint的数量为1而被subsume掉了,也就是被删掉了。
>>
>> 如果你还想保留之前的一个旧的checkpoint,可以将默认retain的checkpoint数目设置为2 [2]。
>>
>> 另外说一句,即使是已经deprecated的cancel with
>> savepoint的用法,当新的savepoint创建成功后,旧的checkpoint在默认情况下也应该会被删除,除非增大retain的checkpoint数量。
>>
>>
>> [1] https://issues.apache.org/jira/browse/FLINK-10354
>> [2]
>> https://ci.apache.org/projects/flink/flink-docs-stable/dev/stream/state/checkpointing.html#state-checkpoints-num-retained
>>
>> 祝好
>> 唐云
>> ________________________________
>> From: yinghua_zh@163.com <yi...@163.com>
>> Sent: Thursday, January 14, 2021 19:00
>> To: user-zh <us...@flink.apache.org>
>> Subject: 回复: 回复: 请教个Flink checkpoint的问题
>>
>> 好的,感谢您的回复!
>>
>>
>>
>> yinghua_zh@163.com
>>
>> 发件人: Evan
>> 发送时间: 2021-01-14 18:48
>> 收件人: user-zh
>> 主题: 回复: 回复: 请教个Flink checkpoint的问题
>> 是的,应该是机制问题,链接[1]打开有这样一句解释:
>>
>> If you choose to retain externalized checkpoints on cancellation you have
>> to handle checkpoint clean up manually when you cancel the job as well
>> (terminating with job status JobStatus#CANCELED).
>>
>> [1]
>> https://ci.apache.org/projects/flink/flink-docs-release-1.12/zh/deployment/config.html#execution-checkpointing-externalized-checkpoint-retention
>>
>> 如回答有误,请指正。
>>
>>
>>
>>
>>
>> 发件人: yinghua_zh@163.com
>> 发送时间: 2021-01-14 18:02
>> 收件人: user-zh
>> 主题: 回复: 回复: 请教个Flink checkpoint的问题
>> 代码如下:
>>     streamEnv.enableCheckpointing(5 * 60 * 1000);
>>     CheckpointConfig checkPointConfig = streamEnv.getCheckpointConfig();
>>     checkPointConfig.setCheckpointingMode(CheckpointingMode.EXACTLY_ONCE);
>>     checkPointConfig.setCheckpointTimeout(1 * 60 * 1000);
>>     checkPointConfig.setMinPauseBetweenCheckpoints((1 * 30 * 1000));
>>     checkPointConfig.setMaxConcurrentCheckpoints(1);
>>     checkPointConfig.setTolerableCheckpointFailureNumber(3);
>>     checkPointConfig
>>
>> .enableExternalizedCheckpoints(ExternalizedCheckpointCleanup.RETAIN_ON_CANCELLATION);
>>
>>     String checkpointPath = this.execOptions.get(CHECKPOINT_PATH);
>>     try {
>>       StateBackend rocksDBStateBackend = new
>> RocksDBStateBackend(checkpointPath);
>>       streamEnv.setStateBackend(rocksDBStateBackend);
>>
>>
>>
>> yinghua_zh@163.com
>> 发件人: Evan
>> 发送时间: 2021-01-14 17:55
>> 收件人: user-zh
>> 主题: 回复: 请教个Flink checkpoint的问题
>> 代码图挂掉了,看不到代码
>> 发件人: yinghua_zh@163.com
>> 发送时间: 2021-01-14 17:26
>> 收件人: user-zh
>> 主题: 请教个Flink checkpoint的问题
>>
>> 我在yarn上提交任务时,设置flink的checkpoint是5分钟一次,checkpoint使用RocksDBStateBackend保存在HDFS上且任务取消后不删除checkpoint,代码如下
>>
>> 现在我发现,当我停止任务时使用stopWithSavepoint发现任务停止后把最近一次的checkpoint信息给删除了?目前机制是这样的吗还是我使用有问题?是不是调用cancelWithSavepoint停止任务时就不会删除最近一次的checkpoint信息?
>> yinghua_zh@163.com
>>

Re: Re: 请教个Flink checkpoint的问题

Posted by Congxian Qiu <qc...@gmail.com>.
Hi
    你可以看一下是不是 作业的状态变成了 FINISH 了,现在 Flink 在 FINISH 的时候,会删除 checkpoint(就算设置了
retain on cancel 也会删除)
    PS :FLINK-18263 正在讨论是否在 Job 变为 FINISH 状态后保留 checkpoint
[1] https://issues.apache.org/jira/browse/FLINK-18263
Best,
Congxian


yinghua_zh@163.com <yi...@163.com> 于2021年1月15日周五 上午11:23写道:

> 感谢您的答复,今天我使用cancelWithSavepoint时也还是删除了,通过你的答复知道机制如此,我再使用另外的参数来观察下。
>
>
>
> yinghua_zh@163.com
>
> 发件人: Yun Tang
> 发送时间: 2021-01-15 11:02
> 收件人: user-zh
> 主题: Re: 回复: 请教个Flink checkpoint的问题
> Hi
>
> 这是因为在Flink-1.7.0 之后,savepoint也被当做是retained checkpoint了 [1],当你stop with
> savepoint 成功时,新的savepoint创建之后,旧的checkpoint因为默认retain
> checkpoint的数量为1而被subsume掉了,也就是被删掉了。
>
> 如果你还想保留之前的一个旧的checkpoint,可以将默认retain的checkpoint数目设置为2 [2]。
>
> 另外说一句,即使是已经deprecated的cancel with
> savepoint的用法,当新的savepoint创建成功后,旧的checkpoint在默认情况下也应该会被删除,除非增大retain的checkpoint数量。
>
>
> [1] https://issues.apache.org/jira/browse/FLINK-10354
> [2]
> https://ci.apache.org/projects/flink/flink-docs-stable/dev/stream/state/checkpointing.html#state-checkpoints-num-retained
>
> 祝好
> 唐云
> ________________________________
> From: yinghua_zh@163.com <yi...@163.com>
> Sent: Thursday, January 14, 2021 19:00
> To: user-zh <us...@flink.apache.org>
> Subject: 回复: 回复: 请教个Flink checkpoint的问题
>
> 好的,感谢您的回复!
>
>
>
> yinghua_zh@163.com
>
> 发件人: Evan
> 发送时间: 2021-01-14 18:48
> 收件人: user-zh
> 主题: 回复: 回复: 请教个Flink checkpoint的问题
> 是的,应该是机制问题,链接[1]打开有这样一句解释:
>
> If you choose to retain externalized checkpoints on cancellation you have
> to handle checkpoint clean up manually when you cancel the job as well
> (terminating with job status JobStatus#CANCELED).
>
> [1]
> https://ci.apache.org/projects/flink/flink-docs-release-1.12/zh/deployment/config.html#execution-checkpointing-externalized-checkpoint-retention
>
> 如回答有误,请指正。
>
>
>
>
>
> 发件人: yinghua_zh@163.com
> 发送时间: 2021-01-14 18:02
> 收件人: user-zh
> 主题: 回复: 回复: 请教个Flink checkpoint的问题
> 代码如下:
>     streamEnv.enableCheckpointing(5 * 60 * 1000);
>     CheckpointConfig checkPointConfig = streamEnv.getCheckpointConfig();
>     checkPointConfig.setCheckpointingMode(CheckpointingMode.EXACTLY_ONCE);
>     checkPointConfig.setCheckpointTimeout(1 * 60 * 1000);
>     checkPointConfig.setMinPauseBetweenCheckpoints((1 * 30 * 1000));
>     checkPointConfig.setMaxConcurrentCheckpoints(1);
>     checkPointConfig.setTolerableCheckpointFailureNumber(3);
>     checkPointConfig
>
> .enableExternalizedCheckpoints(ExternalizedCheckpointCleanup.RETAIN_ON_CANCELLATION);
>
>     String checkpointPath = this.execOptions.get(CHECKPOINT_PATH);
>     try {
>       StateBackend rocksDBStateBackend = new
> RocksDBStateBackend(checkpointPath);
>       streamEnv.setStateBackend(rocksDBStateBackend);
>
>
>
> yinghua_zh@163.com
> 发件人: Evan
> 发送时间: 2021-01-14 17:55
> 收件人: user-zh
> 主题: 回复: 请教个Flink checkpoint的问题
> 代码图挂掉了,看不到代码
> 发件人: yinghua_zh@163.com
> 发送时间: 2021-01-14 17:26
> 收件人: user-zh
> 主题: 请教个Flink checkpoint的问题
>
> 我在yarn上提交任务时,设置flink的checkpoint是5分钟一次,checkpoint使用RocksDBStateBackend保存在HDFS上且任务取消后不删除checkpoint,代码如下
>
> 现在我发现,当我停止任务时使用stopWithSavepoint发现任务停止后把最近一次的checkpoint信息给删除了?目前机制是这样的吗还是我使用有问题?是不是调用cancelWithSavepoint停止任务时就不会删除最近一次的checkpoint信息?
> yinghua_zh@163.com
>

回复: Re: 请教个Flink checkpoint的问题

Posted by "yinghua_zh@163.com" <yi...@163.com>.
感谢您的答复,今天我使用cancelWithSavepoint时也还是删除了,通过你的答复知道机制如此,我再使用另外的参数来观察下。



yinghua_zh@163.com
 
发件人: Yun Tang
发送时间: 2021-01-15 11:02
收件人: user-zh
主题: Re: 回复: 请教个Flink checkpoint的问题
Hi
 
这是因为在Flink-1.7.0 之后,savepoint也被当做是retained checkpoint了 [1],当你stop with savepoint 成功时,新的savepoint创建之后,旧的checkpoint因为默认retain checkpoint的数量为1而被subsume掉了,也就是被删掉了。
 
如果你还想保留之前的一个旧的checkpoint,可以将默认retain的checkpoint数目设置为2 [2]。
 
另外说一句,即使是已经deprecated的cancel with savepoint的用法,当新的savepoint创建成功后,旧的checkpoint在默认情况下也应该会被删除,除非增大retain的checkpoint数量。
 
 
[1] https://issues.apache.org/jira/browse/FLINK-10354
[2] https://ci.apache.org/projects/flink/flink-docs-stable/dev/stream/state/checkpointing.html#state-checkpoints-num-retained
 
祝好
唐云
________________________________
From: yinghua_zh@163.com <yi...@163.com>
Sent: Thursday, January 14, 2021 19:00
To: user-zh <us...@flink.apache.org>
Subject: 回复: 回复: 请教个Flink checkpoint的问题
 
好的,感谢您的回复!
 
 
 
yinghua_zh@163.com
 
发件人: Evan
发送时间: 2021-01-14 18:48
收件人: user-zh
主题: 回复: 回复: 请教个Flink checkpoint的问题
是的,应该是机制问题,链接[1]打开有这样一句解释:
 
If you choose to retain externalized checkpoints on cancellation you have to handle checkpoint clean up manually when you cancel the job as well (terminating with job status JobStatus#CANCELED).
 
[1] https://ci.apache.org/projects/flink/flink-docs-release-1.12/zh/deployment/config.html#execution-checkpointing-externalized-checkpoint-retention
 
如回答有误,请指正。
 
 
 
 
 
发件人: yinghua_zh@163.com
发送时间: 2021-01-14 18:02
收件人: user-zh
主题: 回复: 回复: 请教个Flink checkpoint的问题
代码如下:
    streamEnv.enableCheckpointing(5 * 60 * 1000);
    CheckpointConfig checkPointConfig = streamEnv.getCheckpointConfig();
    checkPointConfig.setCheckpointingMode(CheckpointingMode.EXACTLY_ONCE);
    checkPointConfig.setCheckpointTimeout(1 * 60 * 1000);
    checkPointConfig.setMinPauseBetweenCheckpoints((1 * 30 * 1000));
    checkPointConfig.setMaxConcurrentCheckpoints(1);
    checkPointConfig.setTolerableCheckpointFailureNumber(3);
    checkPointConfig
            .enableExternalizedCheckpoints(ExternalizedCheckpointCleanup.RETAIN_ON_CANCELLATION);
 
    String checkpointPath = this.execOptions.get(CHECKPOINT_PATH);
    try {
      StateBackend rocksDBStateBackend = new RocksDBStateBackend(checkpointPath);
      streamEnv.setStateBackend(rocksDBStateBackend);
 
 
 
yinghua_zh@163.com
发件人: Evan
发送时间: 2021-01-14 17:55
收件人: user-zh
主题: 回复: 请教个Flink checkpoint的问题
代码图挂掉了,看不到代码
发件人: yinghua_zh@163.com
发送时间: 2021-01-14 17:26
收件人: user-zh
主题: 请教个Flink checkpoint的问题
我在yarn上提交任务时,设置flink的checkpoint是5分钟一次,checkpoint使用RocksDBStateBackend保存在HDFS上且任务取消后不删除checkpoint,代码如下
现在我发现,当我停止任务时使用stopWithSavepoint发现任务停止后把最近一次的checkpoint信息给删除了?目前机制是这样的吗还是我使用有问题?是不是调用cancelWithSavepoint停止任务时就不会删除最近一次的checkpoint信息?
yinghua_zh@163.com

Re: 回复: 请教个Flink checkpoint的问题

Posted by Yun Tang <my...@live.com>.
Hi

这是因为在Flink-1.7.0 之后,savepoint也被当做是retained checkpoint了 [1],当你stop with savepoint 成功时,新的savepoint创建之后,旧的checkpoint因为默认retain checkpoint的数量为1而被subsume掉了,也就是被删掉了。

如果你还想保留之前的一个旧的checkpoint,可以将默认retain的checkpoint数目设置为2 [2]。

另外说一句,即使是已经deprecated的cancel with savepoint的用法,当新的savepoint创建成功后,旧的checkpoint在默认情况下也应该会被删除,除非增大retain的checkpoint数量。


[1] https://issues.apache.org/jira/browse/FLINK-10354
[2] https://ci.apache.org/projects/flink/flink-docs-stable/dev/stream/state/checkpointing.html#state-checkpoints-num-retained

祝好
唐云
________________________________
From: yinghua_zh@163.com <yi...@163.com>
Sent: Thursday, January 14, 2021 19:00
To: user-zh <us...@flink.apache.org>
Subject: 回复: 回复: 请教个Flink checkpoint的问题

好的,感谢您的回复!



yinghua_zh@163.com

发件人: Evan
发送时间: 2021-01-14 18:48
收件人: user-zh
主题: 回复: 回复: 请教个Flink checkpoint的问题
是的,应该是机制问题,链接[1]打开有这样一句解释:

If you choose to retain externalized checkpoints on cancellation you have to handle checkpoint clean up manually when you cancel the job as well (terminating with job status JobStatus#CANCELED).

[1] https://ci.apache.org/projects/flink/flink-docs-release-1.12/zh/deployment/config.html#execution-checkpointing-externalized-checkpoint-retention

如回答有误,请指正。





发件人: yinghua_zh@163.com
发送时间: 2021-01-14 18:02
收件人: user-zh
主题: 回复: 回复: 请教个Flink checkpoint的问题
代码如下:
    streamEnv.enableCheckpointing(5 * 60 * 1000);
    CheckpointConfig checkPointConfig = streamEnv.getCheckpointConfig();
    checkPointConfig.setCheckpointingMode(CheckpointingMode.EXACTLY_ONCE);
    checkPointConfig.setCheckpointTimeout(1 * 60 * 1000);
    checkPointConfig.setMinPauseBetweenCheckpoints((1 * 30 * 1000));
    checkPointConfig.setMaxConcurrentCheckpoints(1);
    checkPointConfig.setTolerableCheckpointFailureNumber(3);
    checkPointConfig
            .enableExternalizedCheckpoints(ExternalizedCheckpointCleanup.RETAIN_ON_CANCELLATION);

    String checkpointPath = this.execOptions.get(CHECKPOINT_PATH);
    try {
      StateBackend rocksDBStateBackend = new RocksDBStateBackend(checkpointPath);
      streamEnv.setStateBackend(rocksDBStateBackend);



yinghua_zh@163.com
发件人: Evan
发送时间: 2021-01-14 17:55
收件人: user-zh
主题: 回复: 请教个Flink checkpoint的问题
代码图挂掉了,看不到代码
发件人: yinghua_zh@163.com
发送时间: 2021-01-14 17:26
收件人: user-zh
主题: 请教个Flink checkpoint的问题
我在yarn上提交任务时,设置flink的checkpoint是5分钟一次,checkpoint使用RocksDBStateBackend保存在HDFS上且任务取消后不删除checkpoint,代码如下
现在我发现,当我停止任务时使用stopWithSavepoint发现任务停止后把最近一次的checkpoint信息给删除了?目前机制是这样的吗还是我使用有问题?是不是调用cancelWithSavepoint停止任务时就不会删除最近一次的checkpoint信息?
yinghua_zh@163.com

回复: 回复: 请教个Flink checkpoint的问题

Posted by "yinghua_zh@163.com" <yi...@163.com>.
好的,感谢您的回复!



yinghua_zh@163.com
 
发件人: Evan
发送时间: 2021-01-14 18:48
收件人: user-zh
主题: 回复: 回复: 请教个Flink checkpoint的问题
是的,应该是机制问题,链接[1]打开有这样一句解释:
 
If you choose to retain externalized checkpoints on cancellation you have to handle checkpoint clean up manually when you cancel the job as well (terminating with job status JobStatus#CANCELED).
 
[1] https://ci.apache.org/projects/flink/flink-docs-release-1.12/zh/deployment/config.html#execution-checkpointing-externalized-checkpoint-retention 
 
如回答有误,请指正。
 
 
 
 
 
发件人: yinghua_zh@163.com
发送时间: 2021-01-14 18:02
收件人: user-zh
主题: 回复: 回复: 请教个Flink checkpoint的问题
代码如下:
    streamEnv.enableCheckpointing(5 * 60 * 1000);
    CheckpointConfig checkPointConfig = streamEnv.getCheckpointConfig();
    checkPointConfig.setCheckpointingMode(CheckpointingMode.EXACTLY_ONCE);
    checkPointConfig.setCheckpointTimeout(1 * 60 * 1000);
    checkPointConfig.setMinPauseBetweenCheckpoints((1 * 30 * 1000));
    checkPointConfig.setMaxConcurrentCheckpoints(1);
    checkPointConfig.setTolerableCheckpointFailureNumber(3);
    checkPointConfig
            .enableExternalizedCheckpoints(ExternalizedCheckpointCleanup.RETAIN_ON_CANCELLATION);
 
    String checkpointPath = this.execOptions.get(CHECKPOINT_PATH);
    try {
      StateBackend rocksDBStateBackend = new RocksDBStateBackend(checkpointPath);
      streamEnv.setStateBackend(rocksDBStateBackend);
 
 
 
yinghua_zh@163.com
发件人: Evan
发送时间: 2021-01-14 17:55
收件人: user-zh
主题: 回复: 请教个Flink checkpoint的问题
代码图挂掉了,看不到代码
发件人: yinghua_zh@163.com
发送时间: 2021-01-14 17:26
收件人: user-zh
主题: 请教个Flink checkpoint的问题
我在yarn上提交任务时,设置flink的checkpoint是5分钟一次,checkpoint使用RocksDBStateBackend保存在HDFS上且任务取消后不删除checkpoint,代码如下
现在我发现,当我停止任务时使用stopWithSavepoint发现任务停止后把最近一次的checkpoint信息给删除了?目前机制是这样的吗还是我使用有问题?是不是调用cancelWithSavepoint停止任务时就不会删除最近一次的checkpoint信息?
yinghua_zh@163.com

回复: 回复: 请教个Flink checkpoint的问题

Posted by Evan <ch...@foxmail.com>.
是的,应该是机制问题,链接[1]打开有这样一句解释:

 If you choose to retain externalized checkpoints on cancellation you have to handle checkpoint clean up manually when you cancel the job as well (terminating with job status JobStatus#CANCELED).

[1] https://ci.apache.org/projects/flink/flink-docs-release-1.12/zh/deployment/config.html#execution-checkpointing-externalized-checkpoint-retention 

如回答有误,请指正。





 
发件人: yinghua_zh@163.com
发送时间: 2021-01-14 18:02
收件人: user-zh
主题: 回复: 回复: 请教个Flink checkpoint的问题
代码如下:
    streamEnv.enableCheckpointing(5 * 60 * 1000);
    CheckpointConfig checkPointConfig = streamEnv.getCheckpointConfig();
    checkPointConfig.setCheckpointingMode(CheckpointingMode.EXACTLY_ONCE);
    checkPointConfig.setCheckpointTimeout(1 * 60 * 1000);
    checkPointConfig.setMinPauseBetweenCheckpoints((1 * 30 * 1000));
    checkPointConfig.setMaxConcurrentCheckpoints(1);
    checkPointConfig.setTolerableCheckpointFailureNumber(3);
    checkPointConfig
            .enableExternalizedCheckpoints(ExternalizedCheckpointCleanup.RETAIN_ON_CANCELLATION);

    String checkpointPath = this.execOptions.get(CHECKPOINT_PATH);
    try {
      StateBackend rocksDBStateBackend = new RocksDBStateBackend(checkpointPath);
      streamEnv.setStateBackend(rocksDBStateBackend);



yinghua_zh@163.com
 
发件人: Evan
发送时间: 2021-01-14 17:55
收件人: user-zh
主题: 回复: 请教个Flink checkpoint的问题
代码图挂掉了,看不到代码
 
 
 
发件人: yinghua_zh@163.com
发送时间: 2021-01-14 17:26
收件人: user-zh
主题: 请教个Flink checkpoint的问题
我在yarn上提交任务时,设置flink的checkpoint是5分钟一次,checkpoint使用RocksDBStateBackend保存在HDFS上且任务取消后不删除checkpoint,代码如下
现在我发现,当我停止任务时使用stopWithSavepoint发现任务停止后把最近一次的checkpoint信息给删除了?目前机制是这样的吗还是我使用有问题?是不是调用cancelWithSavepoint停止任务时就不会删除最近一次的checkpoint信息?
 
 
yinghua_zh@163.com