[
https://issues.apache.org/jira/browse/FLINK-10490?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16662292#comment-16662292
]
ASF GitHub Bot commented on FLINK-10490:
----------------------------------------
dawidwys commented on issue #6910: [FLINK-10490][tests] OperatorSnapshotUtil
should use SavepointV2Seria…
URL: https://github.com/apache/flink/pull/6910#issuecomment-432657823
The changes look good.
I am just wondering though, does it mean that all savepoints that we use in
migration tests were written with V1 format? If so, isn't it a problem? Do
those serializers are actually compatible in a way that savepoint written with
V1 format can be read with V2? (Apparently it does as all tests pass)
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
> OperatorSnapshotUtil should probably use SavepointV2Serializer
> --------------------------------------------------------------
>
> Key: FLINK-10490
> URL: https://issues.apache.org/jira/browse/FLINK-10490
> Project: Flink
> Issue Type: Test
> Components: Tests
> Affects Versions: 1.7.0
> Reporter: Stefan Richter
> Priority: Major
> Labels: pull-request-available
>
> {{OperatorSnapshotUtil}} is used for testing savepoint migration. This
> utility internally still uses {{SavepointV1Serializer}} and I would assume
> that it should use {{SavepointV2Serializer}}. I wonder if that means that
> some newer cases are actually not covered in the migration tests.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
