I'm not sure when I checked that the slow test worked. I don’t think it has 
worked recently, and we should not make it a prerequisite of the release.

By the way, I test mongo & splunk every night (using Vladimir’s VM 
https://github.com/vlsi/calcite-test-dataset); I also built & ran the test 
suite on windows yesterday. 

Also, I'm going to be on vacation for 7 days. Limited email. Don’t hold the 
release for me. :)

Julian

> On Aug 23, 2015, at 1:28 PM, Jacques Nadeau <[email protected]> wrote:
> 
> Hey Guys,
> 
> I've resolved my issues.  However, I'm seeing a test appear to run
> forever.  I've tried twice now and it seems to be that one of the test is
> running forever. I've posted the jstack at [1].  Anybody have any thoughts?
> 
> I'm running:
> 
> mvn clean
> mvn install -Dcalcite.test.db={mysql,hsqldb} -Dcalcite.test.slow=true
> -Dcalcite.test.mongodb=true -Dcalcite.test.splunk=true
> 
> @Jesus, I know you made some changes related to an infinite loop.  Did you
> verify that your changes worked with slow tests?
> 
> thanks,
> Jacques
> 
> [1] https://gist.github.com/jacques-n/2607810e775e01d24264
> 
> On Fri, Aug 21, 2015 at 12:33 PM, Jacques Nadeau <[email protected]> wrote:
> 
>> The snapshot should be updated.  Still working on the other issues.
>> Hopefully resolve today or tomorrow.
>> 
>> On Fri, Aug 21, 2015 at 4:14 AM, Jesus Camachorodriguez <
>> [email protected]> wrote:
>> 
>>> I just checked in CALCITE-850. This should be the last issue that was
>>> remaining on the Hive side.
>>> 
>>> @Jacques, could we get a new SNAPSHOT with the last commits so I could run
>>> Hive QA in the meantime?
>>> 
>>> Thanks,
>>> Jesús
>>> 
>>> 
>>> On 8/19/15, 2:12 AM, "Jacques Nadeau" <[email protected]> wrote:
>>> 
>>>> I'm still working on resolving 741/743.  Once I get those figured out,
>>>> I'll
>>>> start a release vote.
>>>> 
>>>> On Tue, Aug 18, 2015 at 4:06 PM, Julian Hyde <[email protected]> wrote:
>>>> 
>>>>> I have checked in Jesus' fix for
>>>>> https://issues.apache.org/jira/browse/CALCITE-834 (thanks Jesus!) and
>>>>> a fix for a cosmetic javadoc issue, so we're good to go.
>>>>> 
>>>>> Julian
>>>>> 
>>> 
>>> 
>> 

Reply via email to