[HIVE-22530] Connection pool timeout in TxnHandler.java is hardcoded to 30 secs

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Invalid
• Component/s: Locking
• Assignee: Unassigned
• Reporter: Jason Dere
• Created: 22/Nov/19 19:00
• Updated: 22/Nov/19 19:31
• Resolved: 22/Nov/19 19:31

Description

If the time to acquire locks gets long enough, we can end up running into the time limit for acquiring DB connections in TxnHandler:

2019-07-23 11:49:54,285 ERROR [HiveServer2-Background-Pool: Thread-3881156]: operation.Operation (SQLOperation.java:run(258)) - Error running hive query:
org.apache.hive.service.cli.HiveSQLException: Error while processing statement: FAILED: Error in acquiring locks: Error communicating with the metastore

Caused by: org.apache.hadoop.hive.ql.lockmgr.LockException: Error communicating with the metastore
Caused by: MetaException(message:Unable to update transaction database org.apache.commons.dbcp.SQLNestedException: Cannot get a connection, pool error Timeout waiting for idle object

This appears to be hard-coded to 30 seconds here: https://github.com/apache/hive/blob/master/standalone-metastore/metastore-server/src/main/java/org/apache/hadoop/hive/metastore/txn/TxnHandler.java#L2359
It may sense to either make this configurable or eliminate the timeout altogether.