[IMPALA-1217] Error running refesh command after restarting catalogd

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: Impala 2.0
• Fix Version/s: Impala 2.1
• Assignee: Lenni Kuff
• Reporter: Alex Leblang
• Created: 09/Sep/14 23:35
• Updated: 20/Dec/15 00:05
• Resolved: 07/Oct/14 23:57

Description

This was discovered running test_restart_catalogd which can be found in tests/experiments/test_process_failures.py

After the catalogd is killed and restarted, the test calls impalad.service.wait_for_metric_value('catalog.ready', 1, timeout=60). When this returns, impala should function normally.
The test then calls self.execute_query_using_client(client, "refresh lineitem", vector) which results in an error.

The log files show the error: I0909 16:27:56.813926 12652 jni-util.cc:177] com.cloudera.impala.catalog.CatalogException: Detected catalog service ID change. Aborting updateCatalog()

It seems that the updated catalog ID is not being successfully integrated with the connection to impala, probably an error in the connection pool.