GitHub user zhengsg added a comment to the discussion: How to configure SparkSQL to access the Hive catalog in Gravitino？
I believe the version of my Hive JAR files is correct. I copied the necessary JARs one by one from the Hive metastore service's runtime directory to a separate directory for SparkSQL to operate on Hive tables. Then, I assigned this directory path to spark.sql.hive.metastore.jars. So where exactly does the problem lie? I suggest that the Gravitino Spark plugin should focus on ensuring proper compatibility and adaptation, rather than relying on configuring spark.sql.hive.metastore.jars, as this approach is prone to compatibility conflicts with Spark. GitHub link: https://github.com/apache/gravitino/discussions/9161#discussioncomment-15010333 ---- This is an automatically sent email for [email protected]. To unsubscribe, please send an email to: [email protected]
