Re: How to avoid breaking states when upgrading Flink job?

Posted by Maximilian Michels on
URL: http://deprecated-apache-flink-user-mailing-list-archive.369.s1.nabble.com/How-to-avoid-breaking-states-when-upgrading-Flink-job-tp7726p7742.html

Hi Josh,

You have to assign UIDs to all operators to change the topology. Plus,
you have to add dummy operators for all UIDs which you removed; this
is a limitation currently because Flink will attempt to find all UIDs
of the old job.

Cheers,
Max

On Wed, Jun 29, 2016 at 9:00 PM, Josh <[hidden email]> wrote:

> Hi all,
> Is there any information out there on how to avoid breaking saved
> states/savepoints when making changes to a Flink job and redeploying it?
>
> I want to know how to avoid exceptions like this:
>
> java.lang.RuntimeException: Failed to deserialize state handle and setup
> initial operator state.
> at org.apache.flink.runtime.taskmanager.Task.run(Task.java:551)
> at java.lang.Thread.run(Thread.java:745)
> Caused by: java.lang.ClassNotFoundException:
> com.me.flink.MyJob$$anon$1$$anon$7$$anon$4
>
>
> The best information I could find in the docs is here:
>
> https://ci.apache.org/projects/flink/flink-docs-master/apis/streaming/savepoints.html
>
>
> Having made the suggested changes to my job (i.e. giving a uid to every
> stateful sink and map function), what changes to the job/topology are then
> allowed/not allowed?
>
>
> If I'm 'naming' my states by providing uids, why does Flink need to look for
> a specific class, like com.me.flink.MyJob$$anon$1$$anon$7$$anon$4 ?
>
>
> Thanks for any advice,
>
> Josh