RE: maven build failure for hive with spark

Thanks Xuefu, I am able to build hive successfully now. But while running hive with spark as execution engine I am getting following error. I am setting up set spark.home=location to /spark-assembly-1.2.0-SNAPSHOT-hadoop2.3.0-cdh5.1.2.jar and same spark assembly I am using for spark stand

Detailing on how UPDATE is performed in Hive

Hi friends Where can I find details on how update is performed in Hive. ​1. When an update is performed,whether HDFS will write that block elsewhere with the new value. 2. whether the old block is unallocated and is allowed for further writes. 3. Whether this process create fragmentation ? 4.

Re: Detailing on how UPDATE is performed in Hive

https://cwiki.apache.org/confluence/display/Hive/Hive+Transactions entire implementation is under jira here https://issues.apache.org/jira/browse/HIVE-5317 On Thu, Nov 27, 2014 at 4:11 PM, unmesha sreeveni unmeshab...@gmail.com wrote: Hi friends Where can I find details on how update is

Re: can't get smallint field from hive on spark

Could you provide a test case to demonstrate the problem? Also the whole exception trace should be helpful. Thanks, Xuefu On Wed, Nov 26, 2014 at 11:20 PM, 诺铁 noty...@gmail.com wrote: we are currently using hive on spark, when reading a small int field, it reports error: Cannot get field

Re: maven build failure for hive with spark

To simply the solution, I'd do the following: 1. get Spark 1.2 source and build it. 2. cp spark-assembly jar to hive/lib folder 3. In hive CLI, exectute set spark.home=/patch/to/spark/srcfolder; spark.home should contain a bin folder, which has alll the spark scripts. This should resolve the

Re: Can't integrate Elasticsearch with Hive

https://docs.oracle.com/javase/7/docs/api/java/lang/IllegalAccessError.html Likley this means the storage handler was built against a different version of hive or one of the dependencies of the two has changed in an incompatible way. E.G. (ES) antlr 2.5 vs (hive) antlr 2.6 or some other

Re: hive server2 error upgrade hive from 0.10.0 to 0.13.1

Hi That might not work, as it might need a later version of hadoop/yarn, if not the one that comes with cdh5.2. Hive 0.13 was built with Hadoop 2.3.0 Yarn. Thanks Szehon On Wed, Nov 26, 2014 at 11:25 PM, Jander g jande...@gmail.com wrote: Hi, everybody We use Hadoop CDH4.6.0， now i try to

cwiki.apache.org - need adjustment to email address

Hi, I need to change my email address in my profile 'mladenkov' held on the confluence wiki for cwiki.apache.org. It says that modifying profile preferences is disabled and to contact the infra team. I'm not sure if I'm reaching the infra team via this mailing list - please let me know who to

RE: Can't integrate Elasticsearch with Hive

I was able to create the External table. But now I am getting an issue in loading the data in the table. I did specify the “es.resources” in the external table so I am not sure why this error is coming. External Table: create external table ex_address (name String, st_no INT, st_name string,

Re: Can't integrate Elasticsearch with Hive

Because the elastic search storage handler is not part of hive you might be able to find more help asking the author of the storage handler. From the errors you are reporting it likely does not work and not tested against this version of hive.

Are there start and stop scripts for various hive services?

Hi, Currently we can only user hive --service metastore to start the metastore service, and the hiveserver2 command to start hiveserver2, but both services will keep the terminal open, althrough we can use the following to run them as daemons, nohup hive --service metastore

Re: cwiki.apache.org - need adjustment to email address

Open a jira in INFRA: https://issues.apache.org/jira/browse/INFRA/ (you need to be logged in). Examples: - https://issues.apache.org/jira/browse/INFRA-4591 -- request was denied, but now it's possible at least in jira - https://issues.apache.org/jira/browse/INFRA-7061 -- jira name

Re: Can't integrate Elasticsearch with Hive

Hi Atul Check your typo tblproperties('es.resources' = 'employee/address') need to be es.resource as it rightly indicated in the Error. *RegardsMuthupandi.K* Think before you print. On Fri, Nov 28, 2014 at 3:43 AM, Edward Capriolo edlinuxg...@gmail.com wrote: Because the elastic

RE: Can't integrate Elasticsearch with Hive

Thanks Muthu for pointing out the issue. However unfortunately I am back to the same old issue while accessing the external table. I am getting the following exception again, there wasn’t any change in the environment. I tried all the cleanup but still no luck :-( Here are the steps I tried

Re: Can't integrate Elasticsearch with Hive

How did you get elasticsearch jar?? Try to build the jar against your version and use that jar. *RegardsMuthupandi.K* Think before you print. On Fri, Nov 28, 2014 at 12:23 PM, Atul Paldhikar (apaldhik) apald...@cisco.com wrote: Thanks Muthu for pointing out the issue. However

RE: Can't integrate Elasticsearch with Hive

Actually even that issue is resolved ☺ There is spelling difference in the sample available on the web, all of them have the storage class as “EsStorageHandler” however only Costin Lieu’s post says it is “ESStorageHandler” which is right ! Now the next problem, the MapReduce job is failing for