IBM Db2 Big SQL

Problems and workarounds: SQL and data movement

Interrupted operations can cause metadata inconsistency
Many operations in Db2 Big SQL are not transactional. HBase and Hive are non-transactional systems. When some operations, such as CREATE INDEX, are interrupted, Db2 Big SQL can be left in an inconsistent state.
Db2 Big SQL memory error during LOAD
Db2 Big SQL generates a memory error during a LOAD operation when the size of the input data exceeds the memory available for the LOAD.
Instance level operations which include topology changes might not synchronize with queries in flight
Online DSM operations, as well as user packages that are running against Db2 Big SQL during online topology changes, might require package rebind.
LOAD using TDCH connector with split column fails due to missing "optional" parameter
If you attempt to run the LOAD HADOOP statement or the sqoop tdimport command from Teradata by using the Teradata Connector for Hadoop (TDCH) 1.5.2 connector, the operation might return an error if you specify the SPLIT COLUMN option.
SQL5197N error during INSERT or SELECT operations on partitioned tables
See the resolution below to resolve scheduler timeout errors which might occur on queries over tables that have many files or partitions.
Error when selecting table data that is stored in the sequence file format
You might encounter SQL5105N while querying table data that is stored in sequence files.
INSERT into a wide table fails with SQL1585N error
CREATE HADOOP TABLE statement fails with SQL0104 invalid keyword after applying patch
Recurring errors on cluster with many worker nodes
When running queries on a cluster with a large number of worker nodes, you encounter recurring SQL errors.
IN list query fails with SQLCODE -901
A large IN list query fails with SQLCODE -901 (continuation is not supported).
Failing delete operation in Big SQL with Isilon OneFS
When Big SQL is integrated with Isilon OneFS, delete operations might fail.

Parent topic: Problems and workarounds

Feedback