[jira] [Assigned] (SPARK-25547) Pluggable jdbc connection factory
[ https://issues.apache.org/jira/browse/SPARK-25547?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Apache Spark reassigned SPARK-25547: Assignee: (was: Apache Spark) > Pluggable jdbc connection factory > - > > Key: SPARK-25547 > URL: https://issues.apache.org/jira/browse/SPARK-25547 > Project: Spark > Issue Type: Improvement > Components: Spark Core >Affects Versions: 2.3.1 >Reporter: Frank Sauer >Priority: Major > > The ability to provide a custom connectionFactoryProvider via JDBCOptions so > that JdbcUtils.createConnectionFactory can produce a custom connection > factory would be very useful. In our case we needed to have the ability to > load balance connections to an AWS Aurora Postgres cluster by round-robining > through the endpoints of the read replicas since their own loan balancing was > insufficient. We got away with it by copying most of the spark jdbc package > and provide this feature there and changing the format from jdbc to our new > package. However it would be nice if this were supported out of the box via > a new option in JDBCOptions providing the classname for a > ConnectionFactoryProvider. I'm creating this Jira in order to submit a PR > which I have ready to go. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Assigned] (SPARK-25547) Pluggable jdbc connection factory
[ https://issues.apache.org/jira/browse/SPARK-25547?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Apache Spark reassigned SPARK-25547: Assignee: Apache Spark > Pluggable jdbc connection factory > - > > Key: SPARK-25547 > URL: https://issues.apache.org/jira/browse/SPARK-25547 > Project: Spark > Issue Type: Improvement > Components: Spark Core >Affects Versions: 2.3.1 >Reporter: Frank Sauer >Assignee: Apache Spark >Priority: Major > > The ability to provide a custom connectionFactoryProvider via JDBCOptions so > that JdbcUtils.createConnectionFactory can produce a custom connection > factory would be very useful. In our case we needed to have the ability to > load balance connections to an AWS Aurora Postgres cluster by round-robining > through the endpoints of the read replicas since their own loan balancing was > insufficient. We got away with it by copying most of the spark jdbc package > and provide this feature there and changing the format from jdbc to our new > package. However it would be nice if this were supported out of the box via > a new option in JDBCOptions providing the classname for a > ConnectionFactoryProvider. I'm creating this Jira in order to submit a PR > which I have ready to go. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
