[
https://issues.apache.org/jira/browse/FLINK-18192?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17225247#comment-17225247
]
Robert Metzger commented on FLINK-18192:
----------------------------------------
We've spend some time preparing a backport for Flink 1.11 already, we could
just merge that. I'm personally not a fan of doing dependency version upgrades
in minor Flink releases.
But I understand that this version upgrade also means a security fix.
I'll open a PR with the 1.11 backport to get CI clearance, you guys can still
decide whether to merge it or not (but we need to be fast, as the 1.11.3
discussion seems to conclude)
> Upgrade to Avro version 1.10.0 from 1.8.2
> -----------------------------------------
>
> Key: FLINK-18192
> URL: https://issues.apache.org/jira/browse/FLINK-18192
> Project: Flink
> Issue Type: Improvement
> Components: Formats (JSON, Avro, Parquet, ORC, SequenceFile)
> Reporter: Lucas Heimberg
> Assignee: Dawid Wysakowicz
> Priority: Major
> Fix For: 1.12.0
>
>
> As of version 1.11, Flink (i.e., flink-avro) still uses Avro in version 1.8.2.
> Avro 1.9.2 contains many bugfixes, in particular in respect to the support
> for logical types. A further advantage would be that an upgrade to Avro 1.9.2
> would also allow to use the Confluent Schema Registry client and Avro
> deserializer in version 5.5.0, which finally support schema references.
> Therefore it would be great if Flink could make use of Avro 1.9.2 or higher
> in future releases.
>
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)