Jason,

I believe Sqoop2 is restricted to CSV formatted text files I believe. The
delimiter should be a comma.

To export to Hive, for now, Sqoop1 is the way to go.

I call the new version of Sqoop, Sqoop2, and the original, Sqoop.

I can't give any estimates on time frames for Sqoop2. What aspects of
Sqoop2 are most interesting to you? I definitely want to see Sqoop2 reach
feature parity as well.

-Abe


On Sun, Feb 16, 2014 at 8:21 PM, Jason Rosenberg <[email protected]> wrote:

> Abe,
>
> Thanks for the info (I guess it wasn't obvious looking at the sqoop client
> doc here:  http://sqoop.apache.org/docs/1.99.3/ClientAPI.html, although
> now looking back at it, it doesn't actually mention hive anywhere :)).
>
> It does look like it can export regular hdfs file data to mysql (is there
> info there on what format the data should be in in hdfs for the export?).
>  I didn't see any docs on delimiting or formatting (just an inputDirectory
> setting, etc.).
>
> It looks like for now, if we want to export data from hive, we should use
> sqoop1 (do you call it 'sqoop1')?
>
> Do you have a rough estimate for the time frame for sqoop2 having feature
> parity with sqoop1?
>
> Thanks,
>
> Jason
>
>
>
>
>
> On Sun, Feb 16, 2014 at 5:53 PM, Abraham Elmahrek <[email protected]>wrote:
>
>> Jason,
>>
>> Sqoop2 unfortunately doesn't have support for Hive yet. There are a few
>> options that the Sqoop2 client has, but not that many. There is the notion
>> of "batch processing" commands which was adding in 
>> SQOOP-773<https://issues.apache.org/jira/browse/SQOOP-773>. This
>> just provides "argument formatted" versions of the interactive menu and
>> executes them in batch.
>>
>> With that being said... the Sqoop community is working hard on bringing
>> Sqoop2 up to feature parity with Sqoop1. It is an open source project and
>> is always seeking new contributors. Feel free to assist in the efforts.
>>
>> -Abe
>>
>>
>> On Sat, Feb 15, 2014 at 5:05 PM, Jason Rosenberg <[email protected]>wrote:
>>
>>> Hi,
>>>
>>> I'm trying to start using sqoop2.  I'm so far not seeing the complete
>>> documentation for describing the arguments for exporting a table from hive,
>>> etc.
>>>
>>> E.g. in sqoop1, you have things like:
>>>
>>> --map-column-java
>>> --fields-terminated-by
>>>
>>> etc.
>>>
>>> Is there a simple mapping for providing these from the sqoop2
>>> thin-client api (as compared to the args we would use directly with
>>> sqoop1)?.  I haven't poked too far down in the source code (was hoping
>>> there's some early documentation on this available?).
>>>
>>> Currently, I'm getting errors like this (and I assume I need to provide
>>> info about types, delimiting, etc.?).
>>>
>>> org.apache.sqoop.common.SqoopException: MAPRED_EXEC_0018:Error occurs 
>>> during loader run
>>>     at 
>>> org.apache.sqoop.job.mr.SqoopOutputFormatLoadExecutor$ConsumerThread.run(SqoopOutputFormatLoadExecutor.java:229)
>>>     at 
>>> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
>>>     at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:334)
>>>     at java.util.concurrent.FutureTask.run(FutureTask.java:166)
>>>     at 
>>> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
>>>     at 
>>> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
>>>     at java.lang.Thread.run(Thread.java:724)
>>> Caused by: java.lang.NumberFormatException: For input string: 
>>> "RilcQmvw9ielrQUw0GB"
>>>     at 
>>> java.lang.NumberFormatException.forInputString(NumberFormatException.java:65)
>>>     at java.lang.Long.parseLong(Long.java:441)
>>>     at java.lang.Long.parseLong(Long.java:483)
>>>     at org.apache.sqoop.job.io.Data.parseField(Data.java:449)
>>>     at org.apache.sqoop.job.io.Data.parse(Data.java:374)
>>>     at org.apache.sqoop.job.io.Data.g
>>> java.lang.Throwable: Child Error
>>>     at org.apache.hadoop.mapred.TaskRunner.run(TaskRunner.java:250)
>>> Caused by: java.io.IOException: Task process exit with nonzero status of 65.
>>>     at org.apache.hadoop.mapred.TaskRunner.run(TaskRunner.java:237)
>>>
>>>
>>> Thanks,
>>>
>>>
>>>
>>> Jason
>>>
>>>
>>>
>>
>

Reply via email to