[ 
https://issues.apache.org/jira/browse/FLINK-17312?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17091693#comment-17091693
 ] 

lun zhang edited comment on FLINK-17312 at 4/24/20, 4:11 PM:
-------------------------------------------------------------

[~ykt836] Thank you for your reply.You may misunderstand my purpose.I want to 
restart one "insert" DML with a restore savepoint file.Because when you update 
a "insert" DML. You will get one single flink job in flink job manager. So when 
you stop or cancel this job with  *savepoint*,it will create only one single 
savepoint file contain meta data.After you *reinsert* with same env and same 
"insert" sql DML.I just want to restart  the cancel job with savepoint can run 
again from last stop.This savepoint dir will not change the sql job savepoint 
file.It's just work in the start.

 

But what you say run multiple *insert DML* in one *sql client session* which I 
don't expect, ** it may cause start from restore savepoint meet some unexpect 
error.

I actually  complete the pull request .It's able to restart the stop job with 
last savepoint.It's same as "*flink run -s savepoint*".Could you help me see my 
pull request.It's my first try to add a new future in flink.Help for your help.

Thank you for [~lzljs3620320], I think it's a better way if we can set it in 
DML or "set" command.

 


was (Author: zhanglun):
[~ykt836] Thank you for your reply.You may misunderstand my purpose.I want to 
restart one "insert" DML with a restore savepoint file.Because when you update 
a "insert" DML. You will get one single flink job in flink job manager. So when 
you stop or cancel this job with  *savepoint*,it will create only one single 
savepoint file contain meta data.After you *reinsert* with same env and same 
"insert" sql DML.I don't want to set a same savepont dir for multip job.I just 
want to restart  the cancel job with savepoint can run again from last stop.

But what you say run multiple *insert DML* in one *sql client session* which I 
don't expect, ** it may cause start from restore savepoint meet some unexpect 
error.

I actually  complete the pull request .It's able to restart the stop job with 
last savepoint.It's same as "*flink run -s savepoint*".Could you help me see my 
pull request.It's my first try to add a new future in flink.Help for your help.

Thank you for [~lzljs3620320], I think it's a better way if we can set it in 
DML or "set" command.

 

> Support sql client start with restore savepoint
> -----------------------------------------------
>
>                 Key: FLINK-17312
>                 URL: https://issues.apache.org/jira/browse/FLINK-17312
>             Project: Flink
>          Issue Type: Improvement
>          Components: Table SQL / Client
>    Affects Versions: 1.10.0, 1.11.0
>            Reporter: lun zhang
>            Priority: Major
>              Labels: pull-request-available
>
> Sql client  not support  *insert sql job* restart with *restore* *savepoint* 
> current.It's very helpful when you can stop your  flink *insert sql job* and 
> restart with a restore savepoint .



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to