In theory, the serializer should be incompatible if you don't change
anything.
So Could you share more so that we could find the root cause:
1. the exception stack where we could see the detailed error serializer
2. flink conf where we could see some potential problematic parameters

On Mon, Jul 10, 2023 at 12:23 PM 湘晗刚 <1016465...@qq.com> wrote:

> Hi ,
> i am using pojo type ,
> flink version is  1.10,
> but the pojo type is the same as before,
> What do.you mean by saying “modify schema of kryo types”?
>
> ---Original---
> *From:* "Hangxiang Yu"<master...@gmail.com>
> *Date:* Mon, Jul 10, 2023 11:58 AM
> *To:* "湘晗刚"<1016465...@qq.com>;"user"<user@flink.apache.org>;
> *Subject:* Re: the new state serializer can not be incompatible
>
> Hi,
> Could you share the detailed exception stack ? Or Did you modify any job
> logic or parameters ?
> Currently, Flink only supports simple schema evolution (e.g. add or remove
> fields for pojo types) for DataStream Jobs[1].
> Other modifications may cause this exception, for example:
> 1. modify some schemas for SQL Jobs
> 2. some complex modifications for DataStream Jobs (e.g. modify schema of
> kryo types)
> 3. modify some parameters (e.g. enable TTL or disable TTL[2])
>
> [1]
> https://nightlies.apache.org/flink/flink-docs-release-1.17/docs/dev/datastream/fault-tolerance/serialization/schema_evolution/
> [2]
> https://nightlies.apache.org/flink/flink-docs-release-1.17/docs/dev/datastream/fault-tolerance/state/#state-time-to-live-ttl
>
> On Mon, Jul 3, 2023 at 7:56 PM 湘晗刚 <1016465...@qq.com> wrote:
>
>>
>> Hi team : restore.from savepoint i meet:
>> the new state serializer can not be incompatible
>>
>> Who can help me?
>>
>> Thanks in advance
>> Kobe24
>>
>
>
> --
> Best,
> Hangxiang.
>


-- 
Best,
Hangxiang.

Reply via email to