+1
Thank you very much for preparing RC2 for the 1.18.1 release, Abhishek!
I built the C++ part of the project on RHEL8.8 on aarch64 (graviton3)
machine.
The build succeeded in DEBUG configuration. After that, I ran the tests.
All tests have passed except for log_block_manager-test, where several
scenarios failed due to the mismatch on expected and actual number of bytes
in FS containers and files, which I attribute to various issues related to
64k block alignment in containers. There were a few attempts to address
the issue in the past, e.g. in the context of KUDU-3527, and I guess it
might be
something similar. So far those failures look like rest-only issues,
where the expectation on the size wasn't properly adjusted when using 64k
block
alignment or similar "non-standard" (i.e. non-4k) alignment:
src/kudu/fs/log_block_manager-test.cc:1658: Failure
Expected equality of these values:
file_size_on_disk
Which is: 57479168
fcs.excess_bytes + initial_file_size_on_disk
Which is: 57417728
57479168 - 57417728 = 61440
With that, I don't think it's something that should block the release.
Kind regards,
Alexey
On Fri, Dec 12, 2025 at 3:09 AM Marton Greber <[email protected]> wrote:
> Hi!
>
> Thanks for preparing the RC Abhishek!
>
> Here are the tests I performed on Ubuntu22:
> - built the Cpp part in release mode
> - all the Cpp tests passed except:
> 350 - kudu-admin-test.5 (Failed)
> 464 - minidump-test (Failed)
> 469 - curl_util-test (Failed)
> These look like test only issues on Ubuntu22.
> - built the Java part
> - all the Java tests passed except:
>
>
> org.apache.kudu.client.TestSecurity.testExternallyProvidedSubjectRefreshedExternally
> This looks like a test issue.
> - built the Python part
> - all the Python tests passed with Python 2.7 and Python 3.10
> - built the Python client, installed it locally and ran the Python
> examples
>
> No major issues found, only minor, test only issues.
> My vote is +1.
>
> Thanks for your efforts Abhishek!
>
> Marton
>
> Abhishek Chennaka <[email protected]> ezt írta (időpont: 2025. dec. 9.,
> K, 0:54):
>
> > Hello all,
> >
> > The Apache Kudu team is happy to announce the second release candidate
> for
> > Apache Kudu 1.18.1.
> >
> > Apache Kudu 1.18.1 is a maintenance release that offers critical bug
> fixes.
> >
> > This is a source-only release. The artifacts have been staged here:
> > https://dist.apache.org/repos/dist/dev/kudu/1.18.1-RC2/
> >
> > Java convenience binaries in the form of a Maven repository are staged
> > here:
> > https://repository.apache.org/content/repositories/orgapachekudu-1131
> >
> > Linux (x86_64 and aarch64 platforms built on RHEL 8) and macOS (built on
> > Sequoia 15.7.2) test-only Kudu binary JAR artifacts are staged here:
> > https://repository.apache.org/content/repositories/orgapachekudu-1132
> >
> > It is tagged in Git as 1.18.1-RC2 and the corresponding hash is the
> > following:
> >
> >
> https://gitbox.apache.org/repos/asf?p=kudu.git;a=commit;h=3a9f8c319ddbfbff2a9c546e71ff7bd248caebfd
> >
> > The release notes can be found here:
> > https://github.com/apache/kudu/blob/1.18.1-RC2/docs/release_notes.adoc
> > <https://github.com/apache/kudu/blob/1.18.1-RC1/docs/release_notes.adoc>
> >
> > The KEYS file to verify the artifact signatures can be found here:
> > https://dist.apache.org/repos/dist/release/kudu/KEYS
> >
> > I'd suggest going through the release notes, building Kudu, and running
> the
> > unit tests. Testing out the Maven repo would also be appreciated. Also,
> > it's worth running Kudu Java tests against kudu-binary JAR artifact as
> > described in the commit message here:
> >
> >
> https://github.com/apache/kudu/commit/8a6faaa93f3e206ac75e8087731daccaf7ab646a
> >
> > The vote will run until a majority[1] is achieved, but at least until
> > Thursday, December 11 16:00:00 PST 2025.
> >
> > Thank you,
> > Abhishek
> >
> > [1] https://www.apache.org/foundation/voting.html#ReleaseVotes
> >
>
