You might want to check the HiveServer2 log which should have the complete
exception trace to understand why it is not able to create a
MetastoreClient.
On Sun, Nov 26, 2017 at 3:48 PM, Marco wrote:
> Hi Fabricio, did you solve this ?
>
> Looks that your Hive is not accessing the MetaStore datab
Right now we only support Spark 2.0.0, the issue you are facing is probably
due to a version mismatch issue.
You may find this JIRA useful - SPARK-16292 - you want to make sure you are
building the Spark distribution correctly.
On Mon, Nov 27, 2017 at 8:48 AM, wrote:
> Hello all
>
>
>
> I’m try
Hello all
I'm trying to have Hive running on top of a Spark cluster.
- Hive version: 2.3.2, installed with the embedded derby database
(local mode)
- Spark version: 2.2.0, installed in cluster mode, no yarn, no mesos
- Hadoop version: 2.7.4
- OS: Redhat 7
Hi,
If you've got the 1st starvation fixed (with Hadoop 2.8 patch), all these
configs + enable log4j2 async logging, you should definitely see a performance
improvement.
Here's the log patches, which need a corresponding LLAP config (& have to be
disabled in HS2, for the progress bar to work)
