leishuiyu created FLINK-37646:
-
Summary: flink jobmanager oom
Key: FLINK-37646
URL: https://issues.apache.org/jira/browse/FLINK-37646
Project: Flink
Issue Type: Bug
Components: Runtime
Hi everyone,
I would like to start a discussion about FLIP-520: Simplify
StructuredType handling [1].
Flink SQL already supports structured types in the engine, serializers,
UDFs, and connector interfaces. However, currently only Table API was
able to make use of them. While UDFs can take ob
+1 (non-binding)
I’ve done the following checklist:
- Tarball checksum and GPG signatures are valid
- Jar files were built with JDK 1.8.0_381 and Apache Maven 3.8.6, bytecode
version is 52.0
- Compiled source cod & ran unit test cases locally with JDK 8, 11, and 17
- Reviewed release note and fl
+1 (binding)
On Tue, Apr 8, 2025 at 12:14 PM Őrhidi Mátyás
wrote:
> Hi Leonard,
>
> Your findings are correct. Users have to implement their own
> KafkaMetadataService. KafkaMetadataServiceImpl is an example for API usage
> in the docs.
>
> Best,
> Matyas
>
> On Tue, Apr 8, 2025 at 1:32 AM Leona
Hi Yanquan,
I think you are completely right. This current state doesn't work. So I'm
calling RC2 off until we fix this issue.
We added one code path in 3.4 (!) that made this issue probably more
obvious but the unchecked use of guava has been going on since 2 years. We
have more than 10 instance
Yanquan Lv created FLINK-37644:
--
Summary: Include guava dependency to package of Kafka connector
release and shade
Key: FLINK-37644
URL: https://issues.apache.org/jira/browse/FLINK-37644
Project: Flink
Thank you all who voted, the vote is now closed.
I will announce the results in another thread.
Best,
Ferenc
On Wednesday, April 9th, 2025 at 13:05, Márton Balassi
wrote:
>
>
> Thanks.
>
> +1 (binding)
>
> Verified:
> - Checkpoints, signatures
> - Checked notice files + no binaries in sou
Hi everyone,
I am happy to announce that we have unanimously approved
this release.
There are 6 approving votes, 3 of which are binding:
- Gyula Fora (binding)
- Zakelly Lan (binding)
- Mate Czagany (non-binding)
- Xiqian Yu (non-binding)
- Yanquan Lv (non-binding)
- Marton Balassi (binding)
Th
I checked:
- Review JIRA release notes
- Verify hashes and verify signatures
- Build success from source with JDK17 & maven3.8.6
- Source code artifacts matching the current release
- Read the announcement blog
- Verify that the major version is 55
I am trying to use the jar flink-sql-connector-k
Dawid Wysakowicz created FLINK-37643:
Summary: Support partial deletes when converting to external data
structures
Key: FLINK-37643
URL: https://issues.apache.org/jira/browse/FLINK-37643
Project:
Hi all,
I have also added a 'Follow up' section at the end of the FLIP-425[1]
describing this.
[1] https://cwiki.apache.org/confluence/x/S4p3EQ
Best,
Zakelly
On Wed, Apr 9, 2025 at 12:42 PM Zakelly Lan wrote:
> Thanks for driving this!
>
> +1 for the FLIP given there is a solid user case beh
Thank you all for voting.
I'm closing the voting and the results will be sent out in a separate email.
> 2025年4月3日 12:10，Han Yin 写道：
>
> Hi everyone,
>
> I'd like to start a vote on the FLIP-432: Faster Checkpoint & Recovery for
> Disaggregated State [1]. The discussion thread is here [2].
>
Hi everyone,
Please review and vote on release candidate #1 for
flink-connector-jdbc v3.3.0, as follows:
[ ] +1, Approve the release
[ ] -1, Do not approve the release (please provide specific comments)
The complete staging area is available for your review, which includes:
* JIRA release notes [
Jiaan Geng created FLINK-37645:
--
Summary: Improve existsUnboundedSource by avoiding duplicate check
transformations
Key: FLINK-37645
URL: https://issues.apache.org/jira/browse/FLINK-37645
Project: Flink
Dear devs, The voting[1] of FLIP-432[2] concluded with the following result.
Approving votes:
Yuan Mei (non-binding)
Zakelly Lan (binding)
Yanfei Lei (binding)
Hangxiang Yu (binding)
There are no disproving votes, so I'm happy to announce that FLIP-432 has been
approved. Thanks Han Yi
Bonnie Varghese created FLINK-37647:
---
Summary: Add support for NonTimeRowsUnboundedPrecedingFunction for
OverAggregate
Key: FLINK-37647
URL: https://issues.apache.org/jira/browse/FLINK-37647
Project
Hi Leonard,
Your findings are correct. Users have to implement their own
KafkaMetadataService. KafkaMetadataServiceImpl is an example for API usage
in the docs.
Best,
Matyas
On Tue, Apr 8, 2025 at 1:32 AM Leonard Xu wrote:
> Hi @Matyas,
>
> First of all, apologies for jumping into the vote thr
Zakelly Lan created FLINK-37642:
---
Summary: Use the mailboxExecutor for operator in Async state
processing
Key: FLINK-37642
URL: https://issues.apache.org/jira/browse/FLINK-37642
Project: Flink
+1 (non-binding)
- verified checksum and signature
- checked no binaries in the source archive
- built source with JDK11 + Maven 3.8.6
- verified release tag, CI, and staging repo
- reviewed web PR
Best,
Ferenc
On Wednesday, April 9th, 2025 at 12:25, Xiqian YU wrote:
>
>
> +1 (non-binding)
Hi Hong,
Thanks for the comments and suggestions, really appreciate it!
> Clarify Cloudwatch API limitations and handling in the sink. [1] Great to see
> we're being explicit with the Java API model! Let's be explicit about the
> PutMetricsData API constraints and how we handle them (it would
Hi Ahmed,
Thanks for the detailed feedback and questions
Just a quick note that after gathering you and Hong’s feedback, I have updated
the design to remove “CloudWatchMetricInput” model and introduce
“MetricWriteRequest” instead, and we will expose an element converter that user
can implement
+1 (non-binding)
Checklist:
- Verified tarball checksum & signatures are valid
- Compiled source code locally with JDK 17
- Ran unit test cases locally and all passed
- Confirmed dist jars are targeted to 52.0 (Java 8)
- Reviewed release notes and flink-web PR
Best Regards,
Xiqian
> 2025年4月9日 1
Jiaan Geng created FLINK-37637:
--
Summary: Avoid dead lock for Configuration's addAll
Key: FLINK-37637
URL: https://issues.apache.org/jira/browse/FLINK-37637
Project: Flink
Issue Type: Bug
+1 (non-binding)
I checked:
- Review JIRA release notes
- Verify hashes and verify signatures
- Build success from source with JDK11 & maven3.8.6
- Source code artifacts matching the current release
- Read the announcement blog and LGTM
Ferenc Csaky 于2025年3月28日周五 21:34写道：
> Hi everyone,
>
> Ple
Jiaan Geng created FLINK-37638:
--
Summary: Avoid duplicate call on getEffectiveConfiguration
Key: FLINK-37638
URL: https://issues.apache.org/jira/browse/FLINK-37638
Project: Flink
Issue Type: Imp
Hi everyone,
Please review and vote on release candidate #2 for flink-connector-kafka
v4.0.0, as follows:
[ ] +1, Approve the release
[ ] -1, Do not approve the release (please provide specific comments)
The complete staging area is available for your review, which includes:
* JIRA release notes
Hi, Weijie.
Sorry for reply again, But I found that the major version of the release
jar is 52 using `javap -classpath
flink-sql-connector-elasticsearch7-4.0.0-2.0.jar -verbose
org.apache.flink.connector.elasticsearch.Elasticsearch7ApiCallBridge | grep
version`, Is this not as expected? Do we need
Selman Kayrancioglu created FLINK-37639:
---
Summary: KeyedStateBootstrapFunction hangs if an exception is
thrown within
Key: FLINK-37639
URL: https://issues.apache.org/jira/browse/FLINK-37639
Proj
Hi Arvid,
I think what Kunni is mentioning is the compiled JDK version (JDK 17 here), not
the targeted class file version (JDK 8 here). The target version could be lower
than the JDK version used.
Actual target version could be determined by executing `javap -verbose X.class
| grep version`, w
Sorry for the previous conclusion, I have confirmed again and found that
the major version is 52, as you and Xiqian mentioned.
But fortunately, we can confirm and discover the actual problem.
Arvid Heise 于2025年4月8日周二 22:09写道：
> Hi everyone,
> Please review and vote on release candidate #1 for fl
Glad that it's sorted out. META-INF now also shows 11.
Also, I forgot to mention that I also updated the NOTICE as Hongshun
pointed out.
On Wed, Apr 9, 2025 at 11:22 AM Yanquan Lv wrote:
> Sorry for the previous conclusion, I have confirmed again and found that
> the major version is 52, as you
+1 (non-binding)
I agree that a release that supports 1.20 (LTS version) is more important
because adding this SQL connector to the official release also requires
development time, and it is reasonable not to block the use of more users.
Just want to confirm that we haven't missed anything importa
Shuai Xu created FLINK-37640:
Summary: Type timestamp_ltz is converted timestamp wrongly when
scanreuse
Key: FLINK-37640
URL: https://issues.apache.org/jira/browse/FLINK-37640
Project: Flink
Iss
I have started another thread. Sources are unchanged and I just updated the
binaries.
However, I found something odd. The release scripts currently hard code
Java 8 as a target. So I'm not sure how the previous binary was JDK 17.
Unfortunately, I deleted the binaries before checking. @Yanquan Lv
Hongshun Wang created FLINK-37641:
-
Summary: Fix: SchemaEvolveE2eITCase.testLenientSchemaEvolution
timeout
Key: FLINK-37641
URL: https://issues.apache.org/jira/browse/FLINK-37641
Project: Flink
Thanks.
+1 (binding)
Verified:
- Checkpoints, signatures
- Checked notice files + no binaries in source release
- Built from source
- Verified release tag, release notes and website PR
Best,
Marton
On Wed, Apr 9, 2025 at 9:19 AM Yanquan Lv wrote:
> +1 (non-binding)
>
> I checked:
> - Revi
Hi, David and Yanquan.
It seems like these issues are not blockers. And the PRs need some
modifications to merge.
I will move these open issues to a new version and prepare the release
version jdbc 3.3.0.
Thanks for all helps~
Best,
Hang
On Wed, Feb 19, 2025 at 9:43 AM Yanquan Lv wrote:
>
> Hi,
37 matches
Mail list logo
