So, good news.. based on LEGAL input it's fine to keep redpanda.

On Mon, Dec 16, 2024 at 2:28 PM Alex Porcelli <porce...@apache.org> wrote:
>
> Thank you, Gonzalo!
>
> Please also reach out to gene...@incubator.apache.org mailing list.
>
> -
> Alex
>
>
> On Mon, Dec 16, 2024 at 2:18 PM Gonzalo Muñoz <gonzalo51...@gmail.com> wrote:
>>
>> Thanks Alex, I've created a LEGAL jira as you have proposed:
>> https://issues.apache.org/jira/browse/LEGAL-693
>>
>>
>> El lun, 16 dic 2024 a las 13:25, Alex Porcelli (<a...@porcelli.me>)
>> escribió:
>>
>> > Francisco and Gonzalo,
>> >
>> > I recommend you open a LEGAL ticket in Apache JIRA and send an email to
>> > IPMC to confirm your perspectives.
>> >
>> > I have no personal preference, as long as IPMC agree with it.
>> >
>> > Alex
>> >
>> >
>> > On Mon, Dec 16, 2024 at 7:05 AM Francisco Javier Tirado Sarti <
>> > ftira...@redhat.com> wrote:
>> >
>> > > Alex,
>> > > In the case of Oracle, we were including a dependency to the driver in
>> > the
>> > > pom.xml file (as test, but a dependency in any case, so it might end up
>> > in
>> > > the distributable and it was used by our software directly if the test
>> > was
>> > > run).
>> > > In the case of RedPanda, there is not any dependency in the pom.xml file,
>> > > but a docker container using redpanda is run if you execute some IT. But
>> > in
>> > > the client side (our software) there is not any forbidden dependency (as
>> > it
>> > > happened with the oracle jdbc driver) and RedPanda is not really used
>> > (what
>> > > is used is the kafka client software, which is valid)
>> > > I'm not sure if you realize that  If we follow your rationale, we would
>> > not
>> > > be not able to invoke any REST server that contains software with an
>> > > invalid license using our REST client.
>> > >
>> > >
>> > > On Mon, Dec 16, 2024 at 12:55 PM Alex Porcelli <a...@porcelli.me> wrote:
>> > >
>> > > > Gonzalo,
>> > > >
>> > > > I understand that, and there was a similar discussion with Oracle JDBC
>> > > > driver dependency.
>> > > >
>> > > > The IPMC seems to have a different interpretation from Camel PMC. But
>> > as
>> > > > Apache KIE is under Incubation, we have to cope with IPMC
>> > interpretation.
>> > > >
>> > > >
>> > > > On Mon, Dec 16, 2024 at 6:50 AM Gonzalo Muñoz <gonzalo51...@gmail.com>
>> > > > wrote:
>> > > >
>> > > > > Hi Alex,
>> > > > >
>> > > > > The controversy pertains to the interpretation of "include" in the
>> > > Apache
>> > > > > requirements regarding Category X software. It is important to
>> > clarify
>> > > > that
>> > > > > we are not including any software from Redpanda; it is only
>> > downloaded
>> > > if
>> > > > > the user explicitly chooses to run those specific tests.
>> > > > >
>> > > > > It is worth noting that Apache Camel follows a similar approach [1].
>> > In
>> > > > > fact, Redpanda is their default option, though they also support
>> > > Strimzi.
>> > > > > [1]
>> > > > >
>> > > > >
>> > > >
>> > >
>> > https://github.com/apache/camel/blob/3c0dbfafab8c61a753dc013ca0a49a9ecf3e7217/test-infra/camel-test-infra-kafka/src/test/resources/org/apache/camel/test/infra/kafka/services/container.properties#L18
>> > > > >
>> > > > > El lun, 16 dic 2024 a las 12:01, Alex Porcelli (<a...@porcelli.me>)
>> > > > > escribió:
>> > > > >
>> > > > > > Toshiya,
>> > > > > >
>> > > > > > What I mean by temporary is that it’s not a longer term solution,
>> > it
>> > > > can
>> > > > > be
>> > > > > > effective for a few releases (similar to what we had for 10.0.0),
>> > but
>> > > > we
>> > > > > > shouldn’t consider it a permanent solution.
>> > > > > >
>> > > > > > Francisco,
>> > > > > >
>> > > > > > The issue is on dependency, not comments… as of now there’s a
>> > > > dependency
>> > > > > > with a Categoxy X software and IPMC made clear its position.
>> > > > > >
>> > > > > >
>> > > > > > On Mon, Dec 16, 2024 at 5:36 AM Francisco Javier Tirado Sarti <
>> > > > > > ftira...@redhat.com> wrote:
>> > > > > >
>> > > > > > > Also, its not even clear to me if we are even referencing
>> > RedPanda
>> > > > > > beyond a
>> > > > > > > string in the setup of the IT code.
>> > > > > > > I do not think that the rule forbids mentioning "Oracle" and
>> > > > "RedPanda"
>> > > > > > in
>> > > > > > > the source code (that will be as absurd as not being able to
>> > have a
>> > > > > > comment
>> > > > > > > mentioning red pandas).  What the rule forbids is including, in
>> > our
>> > > > > > > distribution, that software (which we are not doing)
>> > > > > > >
>> > > > > > > On Mon, Dec 16, 2024 at 7:13 AM Toshiya Kobayashi <
>> > > > > > > toshiyakobaya...@gmail.com> wrote:
>> > > > > > >
>> > > > > > > > Hi Alex,
>> > > > > > > >
>> > > > > > > > > Removing the test code from source release is a temporary
>> > > > solution,
>> > > > > > > > > and it's also a manual (error prone!) process.
>> > > > > > > >
>> > > > > > > > Does "temporary" mean "until graduating incubator"?
>> > > > > > > >
>> > > > > > > > If yes, then removing the test code from source release
>> > wouldn't
>> > > > > bring
>> > > > > > > much
>> > > > > > > > value, because "Has Category X dependencies" is a Minor Issue
>> > > which
>> > > > > > also
>> > > > > > > > gives us time until graduation (
>> > > > > > > > https://issues.apache.org/jira/browse/LEGAL-469), so our
>> > option
>> > > > > seems
>> > > > > > to
>> > > > > > > > be
>> > > > > > > > only one which Ricardo suggested.
>> > > > > > > >
>> > > > > > > > Toshiya
>> > > > > > > >
>> > > > > > > > On Sat, Dec 14, 2024 at 5:02 AM Alex Porcelli <
>> > a...@porcelli.me>
>> > > > > > wrote:
>> > > > > > > >
>> > > > > > > > > Removing the test code from source release is a temporary
>> > > > solution,
>> > > > > > > > > and it's also a manual (error prone!) process.
>> > > > > > > > >
>> > > > > > > > > IPMC has been clear about legal implications in a similar
>> > case
>> > > we
>> > > > > had
>> > > > > > > > > with the Oracle JDBC driver. The mentioned license is
>> > CategoryX
>> > > > and
>> > > > > > > > > can't be referenced.
>> > > > > > > > >
>> > > > > > > > > Yes, different TLP might have different interpretations, but
>> > > IPMC
>> > > > > has
>> > > > > > > > > been clear about theirs for us (again, using the Oracle JDBC
>> > as
>> > > > the
>> > > > > > > > > reference).
>> > > > > > > > >
>> > > > > > > > > Alex
>> > > > > > > > >
>> > > > > > > > > On Fri, Dec 13, 2024 at 11:50 AM Gonzalo Muñoz <
>> > > > > > gonzalo51...@gmail.com
>> > > > > > > >
>> > > > > > > > > wrote:
>> > > > > > > > > >
>> > > > > > > > > > Hi,
>> > > > > > > > > >
>> > > > > > > > > > Adding more context to this, Quarkus Kafka Dev services
>> > uses
>> > > > > > Redpanda
>> > > > > > > > as
>> > > > > > > > > default option (they also have Strimzi and native Kafka, but
>> > > have
>> > > > > to
>> > > > > > be
>> > > > > > > > > opted-in).
>> > > > > > > > > > The rationale is "Because starting a Kafka broker can be
>> > > long,
>> > > > > Dev
>> > > > > > > > > Services for Kafka uses Redpanda, a Kafka compatible broker
>> > > which
>> > > > > > > starts
>> > > > > > > > in
>> > > > > > > > > ~1 second." [1]
>> > > > > > > > > >
>> > > > > > > > > > So, for example, in kogito-apps, there was no reference to
>> > > > > > Redpanda,
>> > > > > > > > > just the use of Quarkus Kafka Dev Services which delegates as
>> > > > > default
>> > > > > > > to
>> > > > > > > > > Redpanda.
>> > > > > > > > > >
>> > > > > > > > > > [1] https://quarkus.io/guides/kafka-dev-services
>> > > > > > > > > >
>> > > > > > > > > > El vie, 13 dic 2024 a las 17:31, ricardo zanini fernandes
>> > (<
>> > > > > > > > > ricardozan...@gmail.com>) escribió:
>> > > > > > > > > >>
>> > > > > > > > > >> Folks, replacing Redpanda with Apache Kafka will
>> > deteriorate
>> > > > our
>> > > > > > > test
>> > > > > > > > > cases
>> > > > > > > > > >> since the first is a lightweight version of the Kafka
>> > > > > > implementation
>> > > > > > > > [1]
>> > > > > > > > > >>
>> > > > > > > > > >> Our tests are already a pain to run, consuming too many
>> > > > > resources
>> > > > > > > from
>> > > > > > > > > the
>> > > > > > > > > >> infrastructure. Can't we just NOT distribute the IT
>> > sources?
>> > > > > > > > > >>
>> > > > > > > > > >> Alternatively, we can remove docker containers and have a
>> > > > Kafka
>> > > > > > Mock
>> > > > > > > > [2]
>> > > > > > > > > >> re-engineering our test architecture slightly and use
>> > > > > Apache/Kafka
>> > > > > > > > ONLY
>> > > > > > > > > for
>> > > > > > > > > >> very specific cases (if any).
>> > > > > > > > > >>
>> > > > > > > > > >> Let me know what you think, and I will open a vote thread.
>> > > > > > > > > >>
>> > > > > > > > > >> [1]
>> > > > > > > > >
>> > > > > https://www.redpanda.com/guides/kafka-alternatives-kafka-limitations
>> > > > > > > > > >> [2] https://www.baeldung.com/kafka-mockproducer
>> > > > > > > > > >>
>> > > > > > > > > >> On Fri, Dec 13, 2024 at 6:37 AM Francisco Javier Tirado
>> > > Sarti
>> > > > <
>> > > > > > > > > >> ftira...@redhat.com> wrote:
>> > > > > > > > > >>
>> > > > > > > > > >> > I vote for B), that way we remove the IT test, if any,
>> > > from
>> > > > > the
>> > > > > > > > source
>> > > > > > > > > >> > distribution, which I think is good anyway. because we
>> > if
>> > > w
>> > > > > only
>> > > > > > > > > distribute
>> > > > > > > > > >> > the actual running code and its sources, we are reducing
>> > > the
>> > > > > > size
>> > > > > > > of
>> > > > > > > > > the
>> > > > > > > > > >> > distributable (and in SW, the smaller, if working, the
>> > > > better)
>> > > > > > > > > >> >
>> > > > > > > > > >> >
>> > > > > > > > > >> >
>> > > > > > > > > >> > On Fri, Dec 13, 2024 at 7:16 AM Toshiya Kobayashi <
>> > > > > > > > > tkoba...@redhat.com>
>> > > > > > > > > >> > wrote:
>> > > > > > > > > >> >
>> > > > > > > > > >> > > Thank you, Andrea.
>> > > > > > > > > >> > >
>> > > > > > > > > >> > > As Alex suggested, we would:
>> > > > > > > > > >> > >
>> > > > > > > > > >> > > A) Replace redpanda with another container which is
>> > > under
>> > > > an
>> > > > > > > > > acceptable
>> > > > > > > > > >> > > license (e.g. apache/kafka)
>> > > > > > > > > >> > > or
>> > > > > > > > > >> > > B) Exclude the relevant test cases from the source
>> > > > > > distribution
>> > > > > > > > > >> > >
>> > > > > > > > > >> > > It will be tracked by
>> > > > > > > > > >> > >
>> > > > https://github.com/apache/incubator-kie-issues/issues/1682
>> > > > > > > > > >> > >
>> > > > > > > > > >> > > Thanks!
>> > > > > > > > > >> > > Toshiya
>> > > > > > > > > >> > >
>> > > > > > > > > >> > >
>> > > > > > > > > >> > >
>> > > > > > > > > >> > > On Thu, Dec 12, 2024 at 7:37 PM Alex Porcelli <
>> > > > > > a...@porcelli.me
>> > > > > > > >
>> > > > > > > > > wrote:
>> > > > > > > > > >> > >
>> > > > > > > > > >> > > > Based on some inputs we got from IPMC when
>> > discussing
>> > > > > Oracle
>> > > > > > > > JDBC
>> > > > > > > > > >> > > > dependency, the test dependencies on CategoryX are
>> > > also
>> > > > > not
>> > > > > > > > > allowed.
>> > > > > > > > > >> > > >
>> > > > > > > > > >> > > > The workaround could be that the source is not
>> > > released
>> > > > > and
>> > > > > > > the
>> > > > > > > > > >> > binaries
>> > > > > > > > > >> > > > aren’t directly dependent on those.
>> > > > > > > > > >> > > >
>> > > > > > > > > >> > > >
>> > > > > > > > > >> > > > On Thu, Dec 12, 2024 at 5:28 AM Andrea Cosentino <
>> > > > > > > > > anco...@gmail.com>
>> > > > > > > > > >> > > > wrote:
>> > > > > > > > > >> > > >
>> > > > > > > > > >> > > >> In the past they were excluded.
>> > > > > > > > > >> > > >>
>> > > > > > > > > >> > > >> At some point the license changed and it was no
>> > more
>> > > in
>> > > > > the
>> > > > > > > > > category
>> > > > > > > > > >> > X.
>> > > > > > > > > >> > > >>
>> > > > > > > > > >> > > >> I don't remember the dependency anyway.
>> > > > > > > > > >> > > >>
>> > > > > > > > > >> > > >> Il giorno gio 12 dic 2024 alle ore 11:12 Toshiya
>> > > > > Kobayashi
>> > > > > > <
>> > > > > > > > > >> > > >> toshiyakobaya...@gmail.com> ha scritto:
>> > > > > > > > > >> > > >>
>> > > > > > > > > >> > > >> > Thank you for the reply, Andrea.
>> > > > > > > > > >> > > >> >
>> > > > > > > > > >> > > >> > > If the dependency is used just for testing and
>> > > it's
>> > > > > not
>> > > > > > > > part
>> > > > > > > > > of
>> > > > > > > > > >> > the
>> > > > > > > > > >> > > >> > release, there are no impediments in using it.
>> > > > > > > > > >> > > >> > >
>> > > > > > > > > >> > > >> > > We were doing something similar in Apache Camel
>> > > too
>> > > > > for
>> > > > > > > > some
>> > > > > > > > > >> > > >> > dependencies with license non-compliant with ASF
>> > > > rules.
>> > > > > > > > > >> > > >> >
>> > > > > > > > > >> > > >> > Are those dependencies excluded from the source
>> > zip
>> > > > of
>> > > > > > the
>> > > > > > > > > Apache
>> > > > > > > > > >> > > Camel
>> > > > > > > > > >> > > >> > release?
>> > > > > > > > > >> > > >> >
>> > > > > > > > > >> > > >> > Thanks,
>> > > > > > > > > >> > > >> > Toshiya
>> > > > > > > > > >> > > >> >
>> > > > > > > > > >> > > >> >
>> > > > > > > > > >> > > >> > On Wed, Dec 11, 2024 at 5:35 PM Andrea Cosentino
>> > <
>> > > > > > > > > anco...@gmail.com
>> > > > > > > > > >> > >
>> > > > > > > > > >> > > >> > wrote:
>> > > > > > > > > >> > > >> >
>> > > > > > > > > >> > > >> >> Hello Toshiya,
>> > > > > > > > > >> > > >> >>
>> > > > > > > > > >> > > >> >> If the dependency is used just for testing and
>> > > it's
>> > > > > not
>> > > > > > > part
>> > > > > > > > > of the
>> > > > > > > > > >> > > >> >> release, there are no impediments in using it.
>> > > > > > > > > >> > > >> >>
>> > > > > > > > > >> > > >> >> We were doing something similar in Apache Camel
>> > > too
>> > > > > for
>> > > > > > > some
>> > > > > > > > > >> > > >> dependencies
>> > > > > > > > > >> > > >> >> with license non-compliant with ASF rules.
>> > > > > > > > > >> > > >> >>
>> > > > > > > > > >> > > >> >> Cheers.
>> > > > > > > > > >> > > >> >>
>> > > > > > > > > >> > > >> >> Il giorno mer 11 dic 2024 alle ore 09:30 Toshiya
>> > > > > > > Kobayashi <
>> > > > > > > > > >> > > >> >> toshiyakobaya...@gmail.com> ha scritto:
>> > > > > > > > > >> > > >> >>
>> > > > > > > > > >> > > >> >>> Hello,
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>> Currently we have GHA test failures with "pull
>> > > > access
>> > > > > > > > denied
>> > > > > > > > > for
>> > > > > > > > > >> > > >> >>> vectorized/redpanda" in kogito repos.
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>> I filed a GH issue.
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>>
>> > > > > > > https://github.com/apache/incubator-kie-issues/issues/1682
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>> Also I have filed a PR to replace
>> > > > > > > > > "vectorized/redpanda:v21.11.8"
>> > > > > > > > > >> > > with
>> > > > > > > > > >> > > >> >>> "redpandadata/redpanda:v24.3.1".
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > >
>> > > > https://github.com/apache/incubator-kie-kogito-runtimes/pull/3812
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>> and now I'm waiting for CI.
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>> Here I want to raise a point. redpanda is under
>> > > > > > Business
>> > > > > > > > > Source
>> > > > > > > > > >> > > >> License
>> > > > > > > > > >> > > >> >>> 1.1.
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> >
>> > > > > > >
>> > https://github.com/redpanda-data/redpanda/blob/dev/licenses/bsl.md
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>> , which is Category X (
>> > > > > > > > > >> > > >> >>>
>> > > > > https://www.apache.org/legal/resolved.html#category-x
>> > > > > > ).
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>> A question to our mentors; Claus, Andrea and
>> > > Brian:
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>> redpanda is used for test infrastructure. Is it
>> > > > > > > considered
>> > > > > > > > as
>> > > > > > > > > >> > > >> >>> "dependency"?
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>> If yes, using redpanda is considered as "Has
>> > > > > Category X
>> > > > > > > > > >> > > dependencies"
>> > > > > > > > > >> > > >> in
>> > > > > > > > > >> > > >> >>>
>> > https://issues.apache.org/jira/browse/LEGAL-469
>> > > ?
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>> Regards,
>> > > > > > > > > >> > > >> >>> Toshiya
>> > > > > > > > > >> > > >> >>>
>> > > > > > > > > >> > > >> >>
>> > > > > > > > > >> > > >>
>> > > > > > > > > >> > > >
>> > > > > > > > > >> > >
>> > > > > > > > > >> >
>> > > > > > > > >
>> > > > > > > >
>> > > > > > >
>> > > > > >
>> > > > >
>> > > >
>> > >
>> >

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org
For additional commands, e-mail: dev-h...@kie.apache.org

Reply via email to