I'm happy to announce that we have unanimously approved this release.
There are 3 approving votes, 3 of which are binding:
* Danny
* Martijn
* Chesnay
There are no disapproving votes.
Thanks everyone!
On 08/11/2022 11:09, Martijn Visser wrote:
+1 (binding)
- Downloaded artifacts
- Checked ha
+1 (binding)
- Downloaded artifacts
- Checked hash and signature
- No binaries in source archive found
- Verified NOTICE files
- Build from source code
- Spinned up a Flink 1.16 cluster with Kafka and Elasticsearch7 and
used flink-sql-connector-elasticsearch7-3.0.0.jar to successfully sent data
fr
+1 (non-binding)
- checked signatures
- checked tag
- checked LICENSE/NOTICE files
- no binaries in source
- checked versions
On Mon, Nov 7, 2022 at 2:32 PM Danny Cranmer
wrote:
> +1 (binding)
>
> - Source release does not contain binaries
> - Source release builds and tests pass (when using
+1 (binding)
- Source release does not contain binaries
- Source release builds and tests pass (when using CI log4j config)
- Verified signature of source archive
- Source code tag exists
- pom versions look correct in the source release
- README is ok
- The staging repository [1] contains expecte
+1 (binding)
- correct Flink version in source release
- no -SNAPSHOT version in source release
- source release compiles and tests are passing
- (mvn verify
-Dlog4j.configurationFile=file://$(pwd)/tools/ci/log4j.properties)
- clean source release
- tag exists and points to correct commit
-
I remembered now; the test failure you ran into is because this
particular fails if you don't have logging enabled.
See FLINK-292789.
On 03/11/2022 21:25, Danny Cranmer wrote:
Hello Chesnay,
-1 (binding)
I do not approve the release due to the source release [1] containing
binaries, the full
I updated the source release on dist.apache.org.
On 03/11/2022 22:03, Chesnay Schepler wrote:
Good findings!
I thought the new scripts ensured a pristine source release, will look
into that tomorrow.
CI passed for the RC so I'd currently assume it's a local issue. But I
remember seeing that
Good findings!
I thought the new scripts ensured a pristine source release, will look
into that tomorrow.
CI passed for the RC so I'd currently assume it's a local issue. But I
remember seeing that particular error message on CI once in a while.
The archunit stored.rules files should not ha
Hello Chesnay,
-1 (binding)
I do not approve the release due to the source release [1] containing
binaries, the full /target folder is present for each module. Assuming we
missed a "mvn clean"
Other concerns/questions:
- I performed a fresh clone of the source and checked out the tag. The
build
Note that this _shouldn't_ be a problem since connector-base is part of
the distribution and should usually be loaded first than the version
bundled by the connector.
On 03/11/2022 11:28, Chesnay Schepler wrote:
* rely on binary compatibility between Flink /*patch */versions
On 03/11/2022 11:
* rely on binary compatibility between Flink /*patch */versions
On 03/11/2022 11:25, Chesnay Schepler wrote:
I've stumbled in an issue regarding flink-connector-base.
Since the externalized connectors rely on binary compatibility between
Flink minor versions, connector-base should never be bun
I've stumbled in an issue regarding flink-connector-base.
Since the externalized connectors rely on binary compatibility between
Flink minor versions, connector-base should never be bundled by an
externalized connector since connector-base may rely on internal APIs
(which ti is allowed to use
Hi everyone,
Please review and vote on the release candidate #2 for the version
3.0.0, 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 [1],
13 matches
Mail list logo
