Thanks for the vote JB.
The fixes for the Licenses are merged. I'm checking if additional fixes are
needed before we create a new RC.

On Wed, Sep 23, 2026 at 8:53 PM Jean-Baptiste Onofré <[email protected]>
wrote:

> -1 (non binding)
>
> I checked:
> - source distribution
> -- checksum and signature are good
> -- LICENSE and NOTICE are correct
> -- ASF header is present in all expected files
> -- No binary/compiled file in the source distribution
> - GCP Bundle
> -- LICENSE is incomplete: caffeine is not documented, jackson
> (FastDoubleParser) needs to be updated
> - AWS Bundle, Azure Bundle, Spark Runtime, Flink Runtime
> -- jackson (FastDoubleParser) needs to be updated in the LICENSE
>
> Regards
> JB
>
> On Wed, Sep 23, 2026 at 9:45 PM Kevin Liu <[email protected]> wrote:
>
>> Caffeine is fixed by #18218 which is now merged.
>> OpenTelemetry / Apache HttpComponents 5 are red herring, they are
>> already covered by existing entries. And we should skip fixing for
>> kafka-connect since we dont bundle that artifact (its disabled for now)
>>
>> TODOs:
>> - CEL Java is a nit (not blocking release), opened #18226 [1]
>> - Found a new gap with Jackson, opened #18227 [2]
>>
>> #18227 alone should unblock us for the next RC.
>>
>> [1] https://github.com/apache/iceberg/pull/18226
>> [2] https://github.com/apache/iceberg/pull/18227
>>
>> On Wed, Sep 23, 2026 at 11:35 AM Kevin Liu <[email protected]> wrote:
>>
>>> Thanks for catching the LICENSE issue. I found a couple more we should
>>> take a look at.
>>>
>>>
>>> Dependency
>>>
>>> LICENSE
>>>
>>> NOTICE
>>>
>>> Caffeine
>>>
>>> GCP: Add missing attribution (PR #18218
>>> <https://github.com/apache/iceberg/pull/18218>)
>>>
>>> No change needed
>>>
>>> CEL Java
>>>
>>> GCP, Kafka Connect main and Hive: Add attribution
>>>
>>> Verify requirements
>>>
>>> OpenTelemetry
>>>
>>> Kafka Connect main and Hive: Review missing attribution
>>>
>>> Verify requirements
>>>
>>> Apache HttpComponents 5
>>>
>>> AWS: Confirm existing attribution covers v5
>>>
>>> Verify upstream NOTICE requirements
>>>
>>> As a follow up, I think it's a good idea to include a check in
>>> the release process. The Caffeine addition to the GCP bundle was flagged in
>>> the original PR by the `runtime-deps.txt` file.
>>>
>>>
>>>
>>> On Wed, Sep 23, 2026 at 8:55 AM Neelesh Salian <[email protected]>
>>> wrote:
>>>
>>>> Thanks for voting and raising the issues. Addressing them one by one
>>>> here:
>>>> 1. Gianluca, thank you for fixing issue #18164
>>>> <https://github.com/apache/iceberg/issues/18164> (reading variant
>>>> columns failed for commons-lang3 < 3.13) with the resolution here:
>>>> #18208 <https://github.com/apache/iceberg/pull/18208>. This should be
>>>> included in the next RC.
>>>> 2. Szehon, the fix https://github.com/apache/iceberg/pull/18069 was in
>>>> this current RC already, so it should be included in a future RC as well.
>>>> Thanks for making the fix.
>>>> 3. Kurtis, thanks for finding and fixing the caffeine License entry.
>>>> It's worth getting that merged and would need a new RC. I'll take another
>>>> pass to see if there are additional license issues (did one before the RCs
>>>> but want to make sure nothing else was missed).
>>>> The RCK issue seems worth fixing as well. I will take a look if this
>>>> was existing or something added after 1.11.0 to assess if we need to get it
>>>> in.
>>>>
>>>> All in all, we will need to cut a new Release Candidate to accommodate
>>>> these changes.
>>>> I'm closing this thread and will work with Amogh to build the new RC
>>>> once we are in the clear and open a new thread.
>>>> Thanks again for verifying this RC, everyone.
>>>>
>>>>
>>>>
>>>>
>>>> On Tue, Sep 22, 2026 at 10:25 PM <[email protected]> wrote:
>>>>
>>>>>
>>>>> -1 (non-binding)
>>>>>
>>>>>
>>>>> Disclaimer this was found after prompting my Claude to validate all
>>>>> the licenses for this RC.
>>>>>
>>>>>
>>>>> Reason: iceberg-gcp-bundle-1.12.0.jar bundles Caffeine with no entry
>>>>> in its LICENSE. Details below. Everything else checked out.
>>>>>
>>>>>
>>>>> Verified:
>>>>>
>>>>>    - sha512 and GPG signature (Amogh Jahagirdar's key from KEYS)
>>>>>    - source tarball matches the apache-iceberg-1.12.0-rc1 tag apart
>>>>>    from the generated iceberg-build.properties and version.txt, and 
>>>>> contains
>>>>>    no compiled artifacts
>>>>>    - LICENSE and NOTICE, dev/check-license (RAT)
>>>>>    - staged binaries in orgapacheiceberg-1285: signatures on the 12
>>>>>    runtime and bundle jars, each jar's LICENSE and NOTICE against the 
>>>>> source
>>>>>    module, and the third-party packages inside each jar against its 
>>>>> LICENSE
>>>>>    entries
>>>>>    - ./gradlew build with tests on three Linux environments:
>>>>>       - Amazon Linux 2023 x86_64, Corretto 21.0.12.1
>>>>>
>>>>> Finding:
>>>>>
>>>>> iceberg-gcp-bundle-1.12.0.jar bundles Caffeine (718 classes under
>>>>> com/github/benmanes/caffeine) and gcp-bundle/LICENSE has no entry for it.
>>>>> Caffeine arrives through com.google.cloud.gcs.analytics:common 1.5.0,
>>>>> pulled in by the gcs-analytics-core bump in #17687. The 1.2.3 version that
>>>>> 1.11.0 used did not depend on Caffeine, and gcp-bundle/LICENSE is 
>>>>> unchanged
>>>>> since 1.11.0, so this is new in 1.12.0. Caffeine is Apache 2.0, so nothing
>>>>> is incompatible, but the bundle's LICENSE has to list what the jar ships.
>>>>> aws-bundle/LICENSE already carries the entry to copy, so the fix is a few
>>>>> lines in gcp-bundle/LICENSE.
>>>>>
>>>>>
>>>>> Issue: https://github.com/apache/iceberg/issues/18217
>>>>>
>>>>> PR: https://github.com/apache/iceberg/pull/18218
>>>>>
>>>>>
>>>>> Additional Non-Blocking Findings:
>>>>>
>>>>> RCK test failures with org.gradle.parallel=true and docker compose
>>>>> available, :iceberg-open-api:test can start while the kafka-connect
>>>>> integration stack holds host port 8181, and the REST Compatibility Kit
>>>>> server then fails to bind. Both use the fixed port 8181.
>>>>>
>>>>>
>>>>> Issue: https://github.com/apache/iceberg/issues/18215
>>>>>
>>>>> PR: https://github.com/apache/iceberg/pull/18216
>>>>>
>>>>>
>>>>> - Kurtis
>>>>>
>>>>> On Sep 22, 2026, at 16:52, Szehon Ho <[email protected]> wrote:
>>>>>
>>>>> 
>>>>> +1
>>>>>
>>>>>
>>>>>    - *Passed:* GPG signature, SHA-512, announced commit/tag, and
>>>>>    license/RAT checks.
>>>>>    - *Core tests:* 8,400 passed, 550 skipped, zero failures.
>>>>>    - *Spark test:* passed using a Spark 4.2 built from verified RC0
>>>>>    source.
>>>>>    -
>>>>>    - I'm unable to access the staged binaries at the moment, so built
>>>>>    from source.
>>>>>    -
>>>>>    - Thanks
>>>>>    - Szehon
>>>>>
>>>>>
>>>>> On Tue, Sep 22, 2026 at 4:08 PM Szehon Ho <[email protected]>
>>>>> wrote:
>>>>>
>>>>>> Hi
>>>>>>
>>>>>> If we make another RC, it'd be nice to get this one in:
>>>>>> https://github.com/apache/iceberg/pull/18069.
>>>>>>
>>>>>> Like the other one, its a pre-existing issue and maybe not worth to
>>>>>> stop this RC, but just another one to add to Gianluca's list.
>>>>>>
>>>>>> Thanks,
>>>>>> Szehon
>>>>>>
>>>>>> On Tue, Sep 22, 2026 at 1:03 PM Neelesh Salian <
>>>>>> [email protected]> wrote:
>>>>>>
>>>>>>> Thanks for raising that Gianluca.
>>>>>>> Since the issue isn't exactly a regression and constrained to the
>>>>>>> version of the commons-lang3, I think we can make a patch into the
>>>>>>> respective branches.
>>>>>>> I tried a local fix but want to make sure it can be tested well
>>>>>>> before making it into a fix. I'll follow up after the release to see 
>>>>>>> how to
>>>>>>> get a proper fix in.
>>>>>>>
>>>>>>> On Tue, Sep 22, 2026 at 11:18 AM Gianluca Graziadei <
>>>>>>> [email protected]> wrote:
>>>>>>>
>>>>>>>> +1 (non-binding)
>>>>>>>>
>>>>>>>> Verified checksums, signature, source archive vs. tag, RAT, and ran
>>>>>>>> smoke tests with the staged Spark 4.0 runtime.
>>>>>>>>
>>>>>>>> Already mentioned for RC0: I hit a pre-existing NoSuchMethodError
>>>>>>>> when reading variant columns with commons-lang3 < 3.13 on the 
>>>>>>>> classpath.
>>>>>>>> iceberg-parquet uses Streams.of(Iterable) without declaring the 
>>>>>>>> dependency.
>>>>>>>> This is also present in 1.11.0, so it’s not a regression. The issue
>>>>>>>> is tracked here: https://github.com/apache/iceberg/issues/18164
>>>>>>>> (PR ready).
>>>>>>>> If there’s a new RC, please consider including the fix. It’s not
>>>>>>>> impactful enough to block the release process.
>>>>>>>>
>>>>>>>> Cheers,
>>>>>>>> Gianluca
>>>>>>>>
>>>>>>>> On 2026/09/22 03:35:41 Neelesh Salian wrote:
>>>>>>>> > Hi Everyone,
>>>>>>>> >
>>>>>>>> > I propose that we release the following RC as the official Apache
>>>>>>>> Iceberg
>>>>>>>> > 1.12.0 release.
>>>>>>>> >
>>>>>>>> > The commit ID is e8d0f0f3004a608e45986cbc2f8e5baf1a8a7db9
>>>>>>>> > * This corresponds to the tag: apache-iceberg-1.12.0-rc1
>>>>>>>> > *
>>>>>>>> https://github.com/apache/iceberg/commits/apache-iceberg-1.12.0-rc1
>>>>>>>> > *
>>>>>>>> >
>>>>>>>> https://github.com/apache/iceberg/tree/e8d0f0f3004a608e45986cbc2f8e5baf1a8a7db9
>>>>>>>> >
>>>>>>>> > The release tarball, signature, and checksums are here:
>>>>>>>> > *
>>>>>>>> https://dist.apache.org/repos/dist/dev/iceberg/apache-iceberg-1.12.0-rc1
>>>>>>>> >
>>>>>>>> > You can find the KEYS file here:
>>>>>>>> > * https://downloads.apache.org/iceberg/KEYS
>>>>>>>> >
>>>>>>>> > Convenience binary artifacts are staged on Nexus. The Maven
>>>>>>>> repository URL
>>>>>>>> > is:
>>>>>>>> > *
>>>>>>>> https://repository.apache.org/content/repositories/orgapacheiceberg-1285/
>>>>>>>> >
>>>>>>>> > Please download, verify, and test.
>>>>>>>> >
>>>>>>>> > Instructions for verifying a release can be found here:
>>>>>>>> > *
>>>>>>>> https://iceberg.apache.org/how-to-release/#how-to-verify-a-release
>>>>>>>> >
>>>>>>>> > Please vote in the next 72 hours.
>>>>>>>> >
>>>>>>>> > [ ] +1 Release this as Apache Iceberg 1.12.0
>>>>>>>> > [ ] +0
>>>>>>>> > [ ] -1 Do not release this because...
>>>>>>>> >
>>>>>>>> > Only PMC members have binding votes, but other community members
>>>>>>>> are
>>>>>>>> > encouraged to cast
>>>>>>>> > non-binding votes. This vote will pass if there are 3 binding +1
>>>>>>>> votes and
>>>>>>>> > more binding
>>>>>>>> > +1 votes than -1 votes.
>>>>>>>> >
>>>>>>>>
>>>>>>>

Reply via email to