Null pointer error with UNION ALL on partitioned tables

["Dragga, Christopher" <Ch...@netapp.com>]

Hi-

I'm using Hive 0.13.1 on Tez and running into an odd null pointer error.  If I run a query of the form

SELECT foo, bar FROM a
UNION ALL
SELECT foo, bar FROM b;

where either a or b is an empty table partitioned on a single key (neither foo nor bar), Hive exits with the message "FAILED: NullPointerException null" before it can runs any tasks.  If the tables in question are not partitioned, or if I use Map Reduce as the execution engine rather than Tez, the query completes successfully.  Has anyone run into this before?  The immediate solution seems to be to temporarily switch the execution engine for this query; are there any better workarounds?   Finally, is this likely to be a bug, and, if so, should I submit a JIRA?  The problem seems related to JIRA 2747, but  I don't think it's the same.

Thanks,
Chris Dragga

Re: Null pointer error with UNION ALL on partitioned tables

[Gunther Hagleitner <gh...@hortonworks.com>]

Suma,

UNION ALL wasn't supported on Tez at the time of the first merge, but it is
there in 0.13.0 (https://issues.apache.org/jira/browse/HIVE-6360).

Chris - sounds like a bug. I tried to repro it on trunk, but it worked for
me. Maybe I'm not hitting the right combination of things. Can you open a
jira? More details like full stack trace, schema and Tez version would help
too.

Thanks,
Gunther.




On Fri, Aug 15, 2014 at 9:37 PM, Suma Shivaprasad <
sumasai.shivaprasad@gmail.com> wrote:

> Hi Chris,
>
> According to this - https://issues.apache.org/jira/browse/HIVE-6098,
>  UNION ALL is still not supported on Tez , and so are SMB  joins
>
> Thanks
> Suma
>
>
> On Sat, Aug 16, 2014 at 4:09 AM, Dragga, Christopher <
> Chris.Dragga@netapp.com> wrote:
>
>>  Hi-
>>
>>
>>
>> I’m using Hive 0.13.1 on Tez and running into an odd null pointer error.
>> If I run a query of the form
>>
>>
>>
>> SELECT foo, bar FROM a
>>
>> UNION ALL
>>
>> SELECT foo, bar FROM b;
>>
>>
>>
>> where either a or b is an empty table partitioned on a single key
>> (neither foo nor bar), Hive exits with the message “FAILED:
>> NullPointerException null” before it can runs any tasks.  If the tables in
>> question are not partitioned, or if I use Map Reduce as the execution
>> engine rather than Tez, the query completes successfully.  Has anyone run
>> into this before?  The immediate solution seems to be to temporarily switch
>> the execution engine for this query; are there any better workarounds?
>> Finally, is this likely to be a bug, and, if so, should I submit a JIRA?
>> The problem seems related to JIRA 2747, but  I don’t think it’s the same.
>>
>>
>>
>> Thanks,
>>
>> Chris Dragga
>>
>
>

-- 
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to 
which it is addressed and may contain information that is confidential, 
privileged and exempt from disclosure under applicable law. If the reader 
of this message is not the intended recipient, you are hereby notified that 
any printing, copying, dissemination, distribution, disclosure or 
forwarding of this communication is strictly prohibited. If you have 
received this communication in error, please contact the sender immediately 
and delete it from your system. Thank You.

Re: Null pointer error with UNION ALL on partitioned tables

[Suma Shivaprasad <su...@gmail.com>]

Hi Chris,

According to this - https://issues.apache.org/jira/browse/HIVE-6098,  UNION
ALL is still not supported on Tez , and so are SMB  joins

Thanks
Suma


On Sat, Aug 16, 2014 at 4:09 AM, Dragga, Christopher <
Chris.Dragga@netapp.com> wrote:

>  Hi-
>
>
>
> I’m using Hive 0.13.1 on Tez and running into an odd null pointer error.
> If I run a query of the form
>
>
>
> SELECT foo, bar FROM a
>
> UNION ALL
>
> SELECT foo, bar FROM b;
>
>
>
> where either a or b is an empty table partitioned on a single key (neither
> foo nor bar), Hive exits with the message “FAILED: NullPointerException
> null” before it can runs any tasks.  If the tables in question are not
> partitioned, or if I use Map Reduce as the execution engine rather than
> Tez, the query completes successfully.  Has anyone run into this before?
> The immediate solution seems to be to temporarily switch the execution
> engine for this query; are there any better workarounds?   Finally, is this
> likely to be a bug, and, if so, should I submit a JIRA?  The problem seems
> related to JIRA 2747, but  I don’t think it’s the same.
>
>
>
> Thanks,
>
> Chris Dragga
>