[
https://issues.apache.org/jira/browse/FLINK-32658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17750097#comment-17750097
]
Rui Fan commented on FLINK-32658:
---------------------------------
Hi [~Zakelly], thanks for your reply.
{quote}you mean there is a problem when the state changes (or be removed) but
the operator still exists (the operator id remains the same), Flink will not
report an error when `ignore-unclaimed-state` is set to `false` ? {quote}
That's exactly what I mean. If Flink reports an error in this case, it is more
friendly to the flink user. Also, if `ignore-unclaimed-state` is set to `true`
flink should discard these removed state inside of the operator to prevent
state leak.
> State should not be silently removed when ignore-unclaimed-state is false
> -------------------------------------------------------------------------
>
> Key: FLINK-32658
> URL: https://issues.apache.org/jira/browse/FLINK-32658
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Checkpointing
> Affects Versions: 1.18.0, 1.17.1
> Reporter: Rui Fan
> Assignee: Rui Fan
> Priority: Major
> Attachments: screenshot-1.png
>
>
> When ignore-unclaimed-state is false and the old state is removed, flink
> should throw exception. It's similar to removing a stateful operator.
> This case occurs not only when the user removes state, but also when the
> operator is replaced.
> For example: upgrade FlinkKafkaConsumer to KafkaSource. All logical are not
> changed, so the operator id isn't changed. The KafkaSource cannot resume from
> the state of FlinkKafkaConsumer. However, the new flink job can start, and
> the state is silently removed in the new job.(The old state is not physically
> discarded, it is still stored in the state backend, but the new code will
> never use it.)
> It also brings an additional problem: the KafkaSource will snapshot 2 states,
> it includes the new state of KafkaSource, and the union list state of
> FlinkKafkaConsumer. Whenever a job resumes from checkpoint, the union List
> state is inflated. Eventually the state size of kafka offset exceeded 200MB.
> !screenshot-1.png!
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
