Release Notes (version 8.1.2)
XSLT user defined functions are not supported in a partitioned database
environment.
When working with XML in a partitioned database environment the data is
split across multiple physical nodes. The data distribution in this
case is not predictable. When working in this kind of environment, you
should ensure the following:
- Use the XMLVARCHAR or XMLCLOB data types instead of XMLFile in your
UDFs.
- When using the UNIX or Windows operating systems, store your XML files on
a file server, and mount or map that server to each machine so that the file
has the same path regardless of what machine accesses it.
- Create a response file when you install DB2 on the instance owning
computer. Use this response file for the rest of your
installations. This ensures that the same components are installed and
configured the same way on each machine.
- Use the -r option to specify the root id in the enable_column
command, so that a consistent partitioning key is used for all the table
data.
[ Top of Page | Previous Page | Next Page | Table of Contents ]