This is an issue with SparkSQL. When it creates the Spark table, it needs Hive meta store under HDFS and without it, the issue you ran into manifests.
The workaround is to run with hdfs user and it has all the downside of elevated priviledges.
↧
Channel:
Hortonworks » All Replies
X
Are you the publisher?
Claim or
contact us
about this channel.
X
No ratings yet.
Channel Details:
- Title: Hortonworks » All Replies
- Channel Number: 30456327
- Language: English
- Registered On: August 23, 2014, 8:28 pm
- Number of Articles: 3435
- Latest Snapshot: February 11, 2016, 10:08 pm
- RSS URL: http://hortonworks.com/community/forums/feed
- Publisher: http://hortonworks.com/community/forums/feed/
- Description:
- Catalog: //hortonworks69.rssing.com/catalog.php?indx=30456327
© 2025 //www.rssing.com