You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@calcite.apache.org by Julian Hyde <jh...@apache.org> on 2019/02/08 18:53:57 UTC

Broken Travis

The Travis build has been broken for 3 cycles: https://travis-ci.org/apache/calcite/builds

It occurs on the JDK11 JDOC=Y build, and started happening with https://gitbox.apache.org/repos/asf?p=calcite.git;a=commit;h=21fe666e9634ab9a752c7cc328924b68f6143afe. 

Vladimir, can you please take a look?

Re: Broken Travis

Posted by Julian Hyde <jh...@apache.org>.

Thanks for taking this on, Kevin, and solving it so efficiently. I really appreciate it.

https://issues.apache.org/jira/browse/CALCITE-2339 was blocked by this issue (its PR was giving CI errors) and is now unblocked.

Julian


> On Feb 8, 2019, at 3:30 PM, Michael Mior <mm...@apache.org> wrote:
> 
> Thanks Kevin!
> --
> Michael Mior
> mmior@apache.org
> 
> Le ven. 8 févr. 2019 à 16:48, Kevin Risden <kr...@apache.org> a écrit :
>> 
>> Master is backed to fixed after committing CALCITE-2836
>> 
>> https://travis-ci.org/apache/calcite/builds/490737420
>> 
>> Kevin Risden
>> 
>> 
>> On Fri, Feb 8, 2019 at 3:47 PM Kevin Risden <kr...@apache.org> wrote:
>> 
>>> https://issues.apache.org/jira/browse/CALCITE-2836
>>> 
>>> Kevin Risden
>>> 
>>> 
>>> On Fri, Feb 8, 2019 at 3:41 PM Kevin Risden <kr...@apache.org> wrote:
>>> 
>>>> Looks like this might just be a leftover compiler plugin in calcite-plus
>>>> module.
>>>> 
>>>> https://github.com/apache/calcite/pull/1038
>>>> 
>>>> The PR above passes with a minor change of removing the compiler-plugin
>>>> to use the top level pom defined one. The comment is out of date as well.
>>>> 
>>>> I'll open a ticket for this and cleanup the PR title/commit message.
>>>> 
>>>> Kevin Risden
>>>> 
>>>> 
>>>> On Fri, Feb 8, 2019 at 2:42 PM Kevin Risden <kr...@apache.org> wrote:
>>>> 
>>>>> From the two builds:
>>>>> 
>>>>>   - Working - Java version: 11.0.1, vendor: Oracle Corporation,
>>>>>   runtime: /usr/lib/jvm/java-11-openjdk-amd64
>>>>>   - Failed - Java version: 11.0.2, vendor: Oracle Corporation,
>>>>>   runtime: /usr/lib/jvm/java-11-openjdk-amd64
>>>>> 
>>>>> Testing with this locally to see how this reproduces and can look into
>>>>> fixing.
>>>>> 
>>>>> Kevin Risden
>>>>> 
>>>>> On Fri, Feb 8, 2019 at 2:34 PM Kevin Risden <kr...@apache.org> wrote:
>>>>> 
>>>>>> I can take a look. I poked around a bit yesterday and saw the same
>>>>>> thing about the commit not changing javadocs. Not sure if JDK 11 got
>>>>>> updated to have this issue. I'll see what it takes to fix.
>>>>>> 
>>>>>> Kevin Risden
>>>>>> 
>>>>>> 
>>>>>> On Fri, Feb 8, 2019 at 2:30 PM Julian Hyde <jh...@apache.org> wrote:
>>>>>> 
>>>>>>> Does someone else have time to take a look?
>>>>>>> 
>>>>>>> 
>>>>>>>> On Feb 8, 2019, at 11:17 AM, Vladimir Sitnikov <
>>>>>>> sitnikov.vladimir@gmail.com> wrote:
>>>>>>>> 
>>>>>>>> Julian> Vladimir, can you please take a look?
>>>>>>>> 
>>>>>>>> As you see, the commit did not touch "Calcite Plus", so I don't
>>>>>>> really
>>>>>>>> think the commit is to blame here. It looks more like JDK-8212233 to
>>>>>>>> me.
>>>>>>>> On top of that, Calcite Plus does not build on my machine (see
>>>>>>>> CALCITE-2816 PsTableFunction fails in Russian locale), so I've a
>>>>>>>> half-baked excuse there as well.
>>>>>>>> 
>>>>>>>> So I'm inlined to tentatively decline your offer.
>>>>>>>> 
>>>>>>>> Vladimir
>>>>>>> 
>>>>>>>