openjpa/openjpa-project/src/doc/manual/supported_databases.xml

1511 lines
57 KiB
XML

<?xml version="1.0" encoding="UTF-8"?>
<!--
Licensed to the Apache Software Foundation (ASF) under one
or more contributor license agreements. See the NOTICE file
distributed with this work for additional information
regarding copyright ownership. The ASF licenses this file
to you under the Apache License, Version 2.0 (the
"License"); you may not use this file except in compliance
with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing,
software distributed under the License is distributed on an
"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
KIND, either express or implied. See the License for the
specific language governing permissions and limitations
under the License.
-->
<appendix id="supported_databases" role="non-normative">
<title>
Supported Databases
</title>
<section id="dbappendix">
<title>
Overview
</title>
<para>
The following appendix covers the database and JDBC driver versions that are
known to work with OpenJPA, along with any database specific configuration
parameters, known issues or limitations.
The <link linkend="dbsupport">Verified Database Matrix</link>, contains the
list of databases and drivers that were tested extensively against this
release of OpenJPA, while the <link linkend="dbcompatible">Compatible Database
Matrix</link> contains the list of databases and drivers that were tested
against prior releases or by OpenJPA users and may not support every feature
of this release.
The <link linked="dbunverified">Unverified Database Matrix</link> contains a
list of databases which have been reported to work, but have not been tested by
the development team.
</para>
</section>
<section id="dbsupport">
<title>
Verified Database Matrix
</title>
<para>
Following is a table of the supported database and JDBC driver versions that
have been verified against this version of OpenJPA. For the list of other
databases and drivers that were tested in prior releases or by other OpenJPA
users, but may not support every feature of this release, please refer to the
<link linkend="dbcompatible">Compatible Database Matrix</link> section.
</para>
<table tocentry="1">
<title>
Supported Databases and JDBC Drivers
</title>
<tgroup rowsep="1" colsep="1" align="left" cols="4">
<colspec colname="dbname"/>
<colspec colname="dbversion"/>
<colspec colname="drivname"/>
<colspec colname="drivversion"/>
<thead>
<row>
<entry colname="dbname">
Database Name
</entry>
<entry colname="dbversion">
Database Version
</entry>
<entry colname="drivname">
JDBC Driver Name
</entry>
<entry colname="drivversion">
JDBC Driver Version
</entry>
</row>
</thead>
<tbody>
<row>
<entry colname="dbname">
<link linkend="dbsupport_derby">Apache Derby</link>
</entry>
<entry colname="dbversion">
10.2.2.0, 10.3.3.0, 10.4.2.0, 10.5.3.0, 10.8.2.2
</entry>
<entry colname="drivname">
Apache Derby Embedded JDBC Driver
</entry>
<entry colname="drivversion">
Same as Database Version
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_db2">IBM DB2</link>
</entry>
<entry colname="dbversion">
8.1, 8.2, 9.1, 9.5, 9.7
</entry>
<entry colname="drivname">
IBM DB2 JDBC Universal Driver
</entry>
<entry colname="drivversion">
3.50.152
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_informix">IBM Informix Dynamic Server</link>
</entry>
<entry colname="dbversion">
10.00xC6, 11.10xC1, 11.5xC1
</entry>
<entry colname="drivname">
Informix JDBC driver
</entry>
<entry colname="drivversion">
3.00 JC3, 3.10 JC1, 3.50 JC1
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_sqlserver">Microsoft SQL Server</link>
</entry>
<entry colname="dbversion">
2005 (9.00), 2008 (10.00)
</entry>
<entry colname="drivname">
Microsoft SQL Server JDBC Driver
</entry>
<entry colname="drivversion">
1.2 or 2.0
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_mysql">MySQL</link>
</entry>
<entry colname="dbversion">
5.0.26, 5.1.6
</entry>
<entry colname="drivname">
MySQL Driver
</entry>
<entry colname="drivversion">
5.1.6
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_oracle">Oracle</link>
</entry>
<entry colname="dbversion">
10g (10.1, 10.2), 11g (11.1, 11.2)
</entry>
<entry colname="drivname">
Oracle JDBC driver
</entry>
<entry colname="drivversion">
11.2.0.1
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_postgresql">PostgreSQL</link>
</entry>
<entry colname="dbversion">
8.3.5, 8.4
</entry>
<entry colname="drivname">
PostgreSQL Native Driver
</entry>
<entry colname="drivversion">
8.3 JDBC3 (build 603), 8.4 JDBC3 (build 701)
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_soliddb">IBM solidDB</link>
</entry>
<entry colname="dbversion">
6.5.0.0
</entry>
<entry colname="drivname">
solidDB JDBC Driver
</entry>
<entry colname="drivversion">
2.0
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_sybase">Sybase Adaptive Server Enterprise</link>
</entry>
<entry colname="dbversion">
12.5, 15.0
</entry>
<entry colname="drivname">
jConnect
</entry>
<entry colname="drivversion">
5.5, 6.0
</entry>
</row>
</tbody>
</tgroup>
</table>
</section>
<section id="dbcompatible">
<title>
Compatible Database Matrix
</title>
<para>
Following is a table of the database and JDBC driver versions that are
compatible with OpenJPA. Some of these databases have been tested against this
version of OpenJPA, while others were added or tested in prior releases and
may not support all of the new features of this release. For the list of
databases that have been fully tested against this release, please refer to the
<link linkend="dbsupport">Verified Database Matrix</link> section.
</para>
<table tocentry="1">
<title>
Compatible Databases and JDBC Drivers
</title>
<tgroup rowsep="1" colsep="1" align="left" cols="4">
<colspec colname="dbname"/>
<colspec colname="dbversion"/>
<colspec colname="drivname"/>
<colspec colname="drivversion"/>
<thead>
<row>
<entry colname="dbname">
Database Name
</entry>
<entry colname="dbversion">
Database Version
</entry>
<entry colname="drivname">
JDBC Driver Name
</entry>
<entry colname="drivversion">
JDBC Driver Version
</entry>
</row>
</thead>
<tbody>
<row>
<entry colname="dbname">
<link linkend="dbsupport_derby">Apache Derby</link>
</entry>
<entry colname="dbversion">
10.1.2.1
</entry>
<entry colname="drivname">
Apache Derby Embedded JDBC Driver
</entry>
<entry colname="drivversion">
10.1.2.1
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_interbase">Borland Interbase</link>
</entry>
<entry colname="dbversion">
7.1.0.202
</entry>
<entry colname="drivname">
Interclient
</entry>
<entry colname="drivversion">
4.5.1
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_jdatastore">Borland JDataStore</link>
</entry>
<entry colname="dbversion">
6.0
</entry>
<entry colname="drivname">
Borland JDataStore
</entry>
<entry colname="drivversion">
6.0
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_db2">DB2</link>
</entry>
<entry colname="dbversion">
8.1
</entry>
<entry colname="drivname">
IBM DB2 JDBC Universal Driver
</entry>
<entry colname="drivversion">
1.0.581, 2.10.72
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_empress">Empress</link>
</entry>
<entry colname="dbversion">
8.62
</entry>
<entry colname="drivname">
Empress Category 2 JDBC Driver
</entry>
<entry colname="drivversion">
8.62
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_firebird">Firebird</link>
</entry>
<entry colname="dbversion">
1.5, 2.0, 2.1
</entry>
<entry colname="drivname">
JayBird JCA/JDBC driver
</entry>
<entry colname="drivversion">
2.1.6
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_h2">H2 Database Engine</link>
</entry>
<entry colname="dbversion">
1.1.118
</entry>
<entry colname="drivname">
H2
</entry>
<entry colname="drivversion">
Same as Database Version
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_hypersonic">Hypersonic Database Engine</link>
</entry>
<entry colname="dbversion">
1.8.0, 2.0.1 RC2
</entry>
<entry colname="drivname">
Hypersonic
</entry>
<entry colname="drivversion">
Same as Database Version
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_informix">Informix Dynamic Server</link>
</entry>
<entry colname="dbversion">
9.30.UC10, 9.4xC7
</entry>
<entry colname="drivname">
Informix JDBC driver
</entry>
<entry colname="drivversion">
2.21.JC2, 3.00 JC3, 3.10 JC1
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_ingres">Ingres Database</link>
</entry>
<entry colname="dbversion">
9.2
</entry>
<entry colname="drivname">
Ingres JDBC Driver
</entry>
<entry colname="drivversion">
9.2-3.4.8
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_intersystems_cache">InterSystems Cache</link>
</entry>
<entry colname="dbversion">
5.0
</entry>
<entry colname="drivname">
Cache JDBC Driver
</entry>
<entry colname="drivversion">
5.0
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_access">Microsoft Access</link>
</entry>
<entry colname="dbversion">
9.0 (a.k.a. "2000")
</entry>
<entry colname="drivname">
DataDirect SequeLink
</entry>
<entry colname="drivversion">
5.4.0038
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_sqlserver">Microsoft SQL Server</link>
</entry>
<entry colname="dbversion">
2000 (8.00)
</entry>
<entry colname="drivname">
Microsoft SQL Server JDBC Driver
</entry>
<entry colname="drivversion">
1.2
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_foxpro">Microsoft Visual FoxPro</link>
</entry>
<entry colname="dbversion">
7.0
</entry>
<entry colname="drivname">
DataDirect SequeLink
</entry>
<entry colname="drivversion">
5.4.0038
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_mysql">MySQL</link>
</entry>
<entry colname="dbversion">
3.23.43-log
</entry>
<entry colname="drivname">
MySQL Driver
</entry>
<entry colname="drivversion">
3.0.14
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_oracle">Oracle</link>
</entry>
<entry colname="dbversion">
8.1, 9.2
</entry>
<entry colname="drivname">
Oracle JDBC driver
</entry>
<entry colname="drivversion">
10.2.0.1
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_pointbase">Pointbase</link>
</entry>
<entry colname="dbversion">
4.4
</entry>
<entry colname="drivname">
Pointbase JDBC driver
</entry>
<entry colname="drivversion">
4.4 (4.4)
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_postgresql">PostgreSQL</link>
</entry>
<entry colname="dbversion">
7.2.1, 8.1.5
</entry>
<entry colname="drivname">
PostgreSQL Native Driver
</entry>
<entry colname="drivversion">
8.1
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_soliddb">IBM solidDB</link>
</entry>
<entry colname="dbversion">
6.5.0.0
</entry>
<entry colname="drivname">
solidDB JDBC Driver
</entry>
<entry colname="drivversion">
2.0
</entry>
</row>
<row>
<entry colname="dbname">
<link linkend="dbsupport_sybase">Sybase Adaptive Server Enterprise</link>
</entry>
<entry colname="dbversion">
12.5
</entry>
<entry colname="drivname">
jConnect
</entry>
<entry colname="drivversion">
5.5 (5.5)
</entry>
</row>
</tbody>
</tgroup>
</table>
</section>
<section id="dbunverified">
<title>
Unverified Database Matrix
</title>
<para>
Following is a table of the database and JDBC driver versions that have been reported
to work with OpenJPA by the community but have not been verified by the development
team. In some cases this is a question of availability since the developers may not
be able to obtain a license to test, or have experience configuring these databases.
For the list of databases that have been fully tested against this release, please
refer to the <link linkend="dbsupport">Verified Database Matrix</link> section.
</para>
<table tocentry="1">
<title>
Unverified Databases and JDBC Drivers
</title>
<tgroup rowsep="1" colsep="1" align="left" cols="4">
<colspec colname="dbname"/>
<colspec colname="dbversion"/>
<colspec colname="drivname"/>
<colspec colname="drivversion"/>
<thead>
<row>
<entry colname="dbname">
Database Name
</entry>
<entry colname="dbversion">
Database Version
</entry>
<entry colname="drivname">
JDBC Driver Name
</entry>
<entry colname="drivversion">
JDBC Driver Version
</entry>
</row>
</thead>
<tbody>
<row>
<entry colname="dbname">
SAP MaxDB
</entry>
<entry colname="dbversion">
</entry>
<entry colname="drivname">
</entry>
<entry colname="drivversion">
</entry>
</row>
</tbody>
</tgroup>
</table>
</section>
<section id="dbsupport_derby">
<title>
Apache Derby
</title>
<example id="example_props_derby">
<title>
Example properties for Derby
</title>
<programlisting>
openjpa.ConnectionDriverName: org.apache.derby.jdbc.EmbeddedDriver
openjpa.ConnectionURL: jdbc:derby:DB_NAME;create=true
</programlisting>
</example>
</section>
<section id="dbsupport_interbase">
<title>
Borland Interbase
</title>
<example id="example_props_interbase">
<title>
Example properties for Interbase
</title>
<programlisting>
openjpa.ConnectionDriverName: interbase.interclient.Driver
openjpa.ConnectionURL: jdbc:interbase://SERVER_NAME:SERVER_PORT/DB_PATH
</programlisting>
</example>
<section id="dbsupport_interbase_issues">
<title>
Known issues with Interbase
</title>
<itemizedlist>
<listitem>
<para>
Interbase does not support record locking, so
datastore transactions cannot use the pessimistic lock manager.
</para>
</listitem>
<listitem>
<para>
Interbase does not support the <literal>LOWER</literal>,
<literal>SUBSTRING</literal>, or <literal>INSTR</literal> SQL functions.
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_jdatastore">
<title>
JDataStore
</title>
<example id="example_props_jdatastore">
<title>
Example properties for JDataStore
</title>
<programlisting>
openjpa.ConnectionDriverName: com.borland.datastore.jdbc.DataStoreDriver
openjpa.ConnectionURL: jdbc:borland:dslocal:db-jdatastore.jds;create=true
</programlisting>
</example>
</section>
<section id="dbsupport_db2">
<title>
IBM DB2
</title>
<example id="example_props_db2">
<title>
Example properties for IBM DB2
</title>
<programlisting>
openjpa.ConnectionDriverName: com.ibm.db2.jcc.DB2Driver
openjpa.ConnectionURL: jdbc:db2://SERVER_NAME:SERVER_PORT/DB_NAME
</programlisting>
</example>
<section id="dbsupport_db2_issues">
<title>
Known issues with DB2
</title>
<itemizedlist>
<listitem>
<para>
Floats and doubles may lose precision when stored.
</para>
</listitem>
<listitem>
<para>
Empty char values are stored as NULL.
</para>
</listitem>
<listitem>
<para>
Fields of type BLOB and CLOB are limited to 1M. This number can be increased by
extending <classname>DB2Dictionary</classname>.
</para>
</listitem>
<listitem>
<para>
Explicit creation of indexes specified by the OpenJPA @Index annotation will
fail on DB2 for iSeries if the default schema used by the JDBC driver does not
exist. If a default schema is not specified on the connection, the iSeries
will default to the user profile name. If a schema of that name does not
exist, DB2 on iSeries will not create the schema, resulting in a failure when
creating the index. The failure message will look similar to: "[SQL0204]
USERNAME in QSYS type *LIB not found." To work around this issue, specify a
default schema on the JDBC URL or data source property and make sure that
schema exists or create a schema which matches the user profile of the
connection.
</para>
</listitem>
<listitem>
<para>
Use of DB2 on z/OS with the IBM JCC driver requires the DESCSTAT subsystem
parameter value to be set to 'YES'. If this parameter is set to 'NO', the
mapping tool will fail with a persistence exception containing this text:
"Invalid parameter: Unknown column name TABLE_SCHEM". After changing the value
of DESCSTAT, DB2 metadata tables must be recreated by running the DSNTIJMS job.
See DB2 for z/OS documentation for additional information.
</para>
</listitem>
<listitem>
<para>
When using LOBs with persistent attributes of a streaming data type (e.g.
<literal>java.io.InputStream</literal>) in the case of very large LOB, DB2 JCC
driver will automatically use progressive streaming to retrieve the LOB data.
With progressiveStreaming, the inputStream retrieved must be fully materialized
before the next iteration of call to rs.next(). By default
this will result in a LobClosedException when OpenJPA processes the InputStream.
</para>
<para>
To work around this condition you may force fullyMaterializedLobData to true in
the connection URL as shown below :
<programlisting>
openjpa.ConnectionURL: jdbc:db2://localhost:50000/demodb:fullyMaterializeLobData=true;progressiveStreaming=NO
</programlisting>
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_empress">
<title>
Empress
</title>
<example id="example_props_empress">
<title>
Example properties for Empress
</title>
<programlisting>
openjpa.ConnectionDriverName: empress.jdbc.empressDriver
openjpa.ConnectionURL: jdbc:empress://SERVER=yourserver;PORT=6322;DATABASE=yourdb
</programlisting>
</example>
<section id="dbsupport_empress_issues">
<title>
Known issues with Empress
</title>
<itemizedlist>
<listitem>
<para>
Empress enforces pessimistic semantics (lock on
read) when not using <literal>AllowConcurrentRead</literal> property (which
bypasses row locking) for <classname>EmpressDictionary</classname>.
</para>
</listitem>
<listitem>
<para>
Only the category 2 non-local driver is supported.
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_h2">
<title>
H2 Database Engine
</title>
<example id="example_props_h2">
<title>
Example properties for H2 Database Engine
</title>
<programlisting>
openjpa.ConnectionDriverName: org.h2.Driver
openjpa.ConnectionURL: jdbc:h2:DB_NAME
</programlisting>
</example>
<section id="dbsupport_h2_issues">
<title>
Known issues with H2 Database Engine
</title>
<itemizedlist>
<listitem>
<para>
None
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_hypersonic">
<title>
Hypersonic
</title>
<example id="example_props_hypersonic">
<title>
Example properties for Hypersonic
</title>
<programlisting>
openjpa.ConnectionDriverName: org.hsqldb.jdbcDriver
openjpa.ConnectionURL: jdbc:hsqldb:DB_NAME
</programlisting>
</example>
<section id="dbsupport_hypersonic_issues">
<title>
Known issues with Hypersonic
</title>
<itemizedlist>
<listitem>
<para>
Hypersonic does not support pessimistic locking, so non-optimistic transactions
will fail unless the <literal>SimulateLocking</literal> property is set for the
<link linkend="openjpa.jdbc.DBDictionary"> openjpa.jdbc.DBDictionary</link>
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_firebird">
<title>
Firebird
</title>
<example id="example_props_firebird">
<title>
Example properties for Firebird
</title>
<programlisting>
openjpa.ConnectionDriverName: org.firebirdsql.jdbc.FBDriver
openjpa.ConnectionURL: jdbc:firebirdsql:SERVER_NAME/3050:DB_PATH_OR_ALIAS
</programlisting>
</example>
<section id="dbsupport_firebird_issues">
<title>
Known issues with Firebird
</title>
<itemizedlist>
<listitem>
<para>
Firebird does not support auto-increment columns.
</para>
</listitem>
<listitem>
<para>
In order to use many of JPQL functions with Firebird 1.5, Interbase UDFs
have to be available in the database.
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_informix">
<title>
Informix
</title>
<example id="example_props_informix">
<title>
Example properties for Informix Dynamic Server
</title>
<programlisting>
openjpa.ConnectionDriverName: com.informix.jdbc.IfxDriver
openjpa.ConnectionURL: \
jdbc:informix-sqli://SERVER_NAME:SERVER_PORT/DB_NAME:INFORMIXSERVER=SERVER_ID
</programlisting>
</example>
<section id="dbsupport_informix_issues">
<title>
Known issues with Informix
</title>
<itemizedlist>
<listitem>
<para>
None
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_ingres">
<title>
Ingres Database
</title>
<example id="example_props_ingres">
<title>
Example properties for Ingres
</title>
<programlisting>
openjpa.ConnectionDriverName: com.ingres.jdbc.IngresDriver
openjpa.ConnectionURL: \
jdbc:ingres://SERVER_NAME:DAS_SERVER_PORT/DB_NAME
</programlisting>
</example>
<section id="dbsupport_ingres_issues">
<title>
Known issues with Ingres
</title>
<itemizedlist>
<listitem>
<para>
Databases must be created with the -n flag for Unicode compatibility in
order to use LOBs.
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_intersystems_cache">
<title>
InterSystems Cache
</title>
<example id="example_props_intersystems_cache">
<title>
Example properties for InterSystems Cache
</title>
<programlisting>
openjpa.ConnectionDriverName: com.intersys.jdbc.CacheDriver
openjpa.ConnectionURL: jdbc:Cache://SERVER_NAME:SERVER_PORT/DB_NAME
</programlisting>
</example>
<section id="dbsupport_intersystems_cache_issues">
<title>
Known issues with InterSystems Cache
</title>
<itemizedlist>
<listitem>
<para>
Support for Cache is done via SQL access over JDBC, not through their object
database APIs.
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_access">
<title>
Microsoft Access
</title>
<example id="example_props_access">
<title>
Example properties for Microsoft Access
</title>
<programlisting>
openjpa.ConnectionDriverName: com.ddtek.jdbc.sequelink.SequeLinkDriver
openjpa.ConnectionURL: jdbc:sequelink://SERVER_NAME:SERVER_PORT
</programlisting>
</example>
<section id="dbsupport_access_issues">
<title>
Known issues with Microsoft Access
</title>
<itemizedlist>
<listitem>
<para>
Using the Sun JDBC-ODBC bridge to connect is not supported.
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_sqlserver">
<title>
Microsoft SQL Server
</title>
<example id="example_props_sqlserver">
<title>
Example properties for Microsoft SQL Server
</title>
<programlisting>
openjpa.ConnectionDriverName: com.microsoft.sqlserver.jdbc.SQLServerDriver
openjpa.ConnectionURL: \
jdbc:sqlserver://SERVER_NAME:1433;DatabaseName=DB_NAME;selectMethod=cursor;sendStringParametersAsUnicode=false
</programlisting>
</example>
<section id="dbsupport_sqlserver_issues">
<title>
Known issues with SQL Server
</title>
<itemizedlist>
<listitem>
<para>
When using a Microsoft SQL Server JDBC Driver v1.2 or earlier, the
ConnectionURL must always contain the <literal>selectMethod=cursor
</literal> string, which is necessary for the driver to properly
support large result sets.
</para>
</listitem>
<listitem>
<para>
When using a Microsoft SQL Server JDBC Driver v1.2 or earlier, the
JDBC driver has bugs that manifest themselves when prepared statements
are pooled. Please disable prepared statement pooling by including the
<literal>MaxCachedStatements=0</literal> configuration property
in your org.apache.openjpa.ConnectionFactoryProperties.
</para>
</listitem>
<listitem>
<para>
SQL Server date fields are accurate only to the nearest 3 milliseconds,
possibly resulting in precision loss in stored dates.
</para>
</listitem>
<listitem>
<para>
Adding <literal>sendStringParametersAsUnicode=false</literal> to the
ConnectionURL may significantly increase performance.
</para>
</listitem>
<listitem>
<para>
The Microsoft SQL Server driver only emulates batch updates. The DataDirect JDBC
driver has true support for batch updates, and may result in a significant
performance gain.
</para>
</listitem>
<listitem>
<para>
Floats and doubles may lose precision when stored.
</para>
</listitem>
<listitem>
<para>
<literal>TEXT</literal> columns cannot be used in queries.
</para>
</listitem>
<listitem>
<para>
When using a SQL Server instance that has been configured to be case-sensitive
in schema names, you need to set the "schemaCase=preserve" parameter in the
<link linkend="openjpa.jdbc.DBDictionary">openjpa.jdbc.DBDictionary</link>
property.
</para>
</listitem>
<listitem>
<para>
SQL Server 2005 does not support native sequences. If you would like to use
generated values with SQL Server you should use GenerationType.IDENTITY,
GenerationType.TABLE, or GenerationType.AUTO.
</para>
</listitem>
<listitem>
<para>
The use of <link linkend="ref_guide_streamsupport">LOB streaming</link> is limited.
When reading LOB data from the database, the Microsoft SQL Server driver will
actually load all the data into memory at the same time.
</para>
</listitem>
<listitem>
<para>
The SQL Server 2008 DATETIME2 data type supports 7 digits sub-second precision.
When DataDirect JDBC driver is used with SQL Server 2008, setTimestamp method call with
a java.sql.Timestamp argument of more than 3 digits precision in a prepared statement
will result in truncation. This may cause loss of data precision or
optimistic lock exception if an entity uses Timestamp type as version field.
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_foxpro">
<title>
Microsoft FoxPro
</title>
<example id="example_props_foxpro">
<title>
Example properties for Microsoft FoxPro
</title>
<programlisting>
openjpa.ConnectionDriverName: com.ddtek.jdbc.sequelink.SequeLinkDriver
openjpa.ConnectionURL: jdbc:sequelink://SERVER_NAME:SERVER_PORT
</programlisting>
</example>
<section id="dbsupport_foxpro_issues">
<title>
Known issues with Microsoft FoxPro
</title>
<itemizedlist>
<listitem>
<para>
Using the Sun JDBC-ODBC bridge to connect is not supported.
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_mysql">
<title>
MySQL
</title>
<example id="example_props_mysql">
<title>
Example properties for MySQL
</title>
<programlisting>
openjpa.ConnectionDriverName: com.mysql.jdbc.Driver
openjpa.ConnectionURL: jdbc:mysql://SERVER_NAME/DB_NAME
</programlisting>
</example>
<section id="dbsupport_mysql_query_hints">
<title>
Using Query Hints with MySQL
</title>
<para>
MySQL has support for "query hints", which are keywords embedded in
SQL that provide some hint for how the query should be executed. These hints are
usually designed to provide suggestions to the MySQL query optimizer for how to
efficiently perform a certain query, and aren't typically needed for any but
the most intensive queries.
OpenJPA supports hints to be placed between SELECT keyword and column list.
</para>
<example id="dbsupport_mysql_query_hints_ex">
<title>
Using MySQL Hints
</title>
<programlisting>
Query query = em.createQuery(...);
query.setHint("openjpa.hint.MySQLSelectHint", "SQL_NO_CACHE");
List results = query.getResultList();
</programlisting>
</example>
</section>
<section id="dbsupport_mysql_issues">
<title>
Known issues with MySQL
</title>
<itemizedlist>
<listitem>
<para>
The default table types that MySQL uses do not support transactions, which will
prevent OpenJPA from being able to roll back transactions. Use the
<literal>InnoDB</literal> table type for any tables that OpenJPA will access.
</para>
</listitem>
<listitem>
<para>
MySQL does not support sub-selects in versions prior to 4.1, so
some operations (such as the <function>isEmpty()</function> method in a
query) will fail due to this.
</para>
</listitem>
<listitem>
<para>
Rollback due to database error or optimistic lock violation is not supported
unless the table type is one of the MySQL transactional types. Explicit calls to
<function>rollback()</function> before a transaction has been committed,
however, are always supported.
</para>
</listitem>
<listitem>
<para>
Floats and doubles may lose precision when stored in some datastores.
</para>
</listitem>
<listitem>
<para>
When storing a field of type <classname>java.math.BigDecimal</classname>, some
datastores will add extraneous trailing 0 characters, causing an equality
mismatch between the field that is stored and the field that is retrieved.
</para>
</listitem>
<listitem>
<para>
When using large result sets with MySQL there are a number of documented limitations.
Please read the section titled "ResultSet" in the "MySQL JDBC API Implementation Notes".
The net effect of these limitations is that you will have to read all of the rows of a
result set (or close the connection) before you can issue any other queries on
the connection, or an exception will be thrown. Setting openjpa.FetchBatchSize
to any value greater than zero will enable streaming result sets.
</para>
</listitem>
<listitem>
<para>
The use of <link linkend="ref_guide_streamsupport">LOB streaming</link> is limited.
When reading LOB data from the database, the MySQL JDBC driver will actually
load all the data into memory at the same time.
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_oracle">
<title>
Oracle
</title>
<example id="example_props_oracle">
<title>
Example properties for Oracle
</title>
<programlisting>
openjpa.ConnectionDriverName: oracle.jdbc.driver.OracleDriver
openjpa.ConnectionURL: jdbc:oracle:thin:@SERVER_NAME:1521:DB_NAME
</programlisting>
</example>
<section id="dbsupport_oracle_query_hints">
<title>
Using Query Hints with Oracle
</title>
<para>
Oracle has support for "query hints", which are formatted comments embedded in
SQL that provide some hint for how the query should be executed. These hints are
usually designed to provide suggestions to the Oracle query optimizer for how to
efficiently perform a certain query, and aren't typically needed for any but
the most intensive queries.
</para>
<example id="dbsupport_oracle_query_hints_ex">
<title>
Using Oracle Hints
</title>
<programlisting>
Query query = em.createQuery(...);
query.setHint("openjpa.hint.OracleSelectHint", "/*+ first_rows(100) */");
List results = query.getResultList();
</programlisting>
</example>
</section>
<section id="dbsupport_oracle_issues">
<title>
Known issues with Oracle
</title>
<itemizedlist>
<listitem>
<para>
The Oracle JDBC driver has significant differences between different versions.
It is important to use the officially supported version of the drivers
(10.2.0.1.0/11.2.0.x.0), which is backward compatible with previous versions of the Oracle
server. It can be downloaded from
<ulink url="http://www.oracle.com/technetwork/database/features/jdbc/index-091264.html">
http://www.oracle.com/technetwork/database/features/jdbc/index-091264.html</ulink>.
</para>
</listitem>
<listitem>
<para>
Empty string/char values are stored as NULL.
</para>
</listitem>
<listitem>
<para>
Oracle corp's JDBC driver for Oracle has only limited support for batch updates.
The result for OpenJPA is that batching of some statements may fail and in some cases,
the exact object that failed an optimistic lock check cannot be determined. OpenJPA will
throw an <classname>OptimisticException</classname> with more failed objects than actually
failed. This situation may be resolved by disabling statement batching by setting the
batchLimit value to zero or by using a more recent Oracle JDBC Driver (11.2.0.1) with
batch support improvements. Attempting to resolve the issue with a more current driver
is recommended since disabling statement batching can result in a decrease in performance.
<example id="dbsupport_oracle_disable_batch_updates">
<title>
Property to disable statement batching for Oracle
</title>
<programlisting>
openjpa.jdbc.DBDictionary: oracle(batchLimit=0)
</programlisting>
</example>
</para>
</listitem>
<listitem>
<para>
Oracle cannot store numbers with more than 38 digits in numeric columns.
</para>
</listitem>
<listitem>
<para>
Floats and doubles may lose precision when stored.
</para>
</listitem>
<listitem>
<para>
CLOB columns cannot be used in queries.
</para>
</listitem>
<listitem>
<para>
The use of LOBs with persistent attributes of a streaming data type (ex.
<literal>java.io.InputStream</literal> or <literal>java.io.Reader</literal>) may
require the same connection to be used over the life of the transaction or
entity manager. If the same connection is not used for persistent operations
a <literal>java.io.IOException</literal> with message <literal>Closed Connection
</literal> may result. The OpenJPA property <literal>openjpa.ConnectionRetainMode</literal>
can be used to control how OpenJPA uses datastore connections. See
<xref linkend="ref_guide_dbsetup_retain"/> for details.
<example id="dbsupport_oracle_retain_connection">
<title>
Property to retain connection over the lifetime of the entity manager
</title>
<programlisting>
openjpa.ConnectionRetainMode: always
</programlisting>
</example>
</para>
</listitem>
<listitem>
<para>
Mapping persistent attributes to <link linkend="ref_guide_xmlmapping">XML columns</link> requires
a JDBC 4 compliant driver if XML strings are longer than 4000 bytes, as counted in database.
Otherwise an <literal>ORA-01461: can bind a LONG value only for insert into a LONG column</literal>
error may result.
</para>
</listitem>
<listitem>
<para>
If Oracle dictionary property <literal>MaxEmbeddedBlobSize</literal> or
<literal>MaxEmbeddedClobSize</literal> is set to some limit (i.e. not -1) and embedded collection
with BLOB/CLOB attribute is used, a
<literal>"org.apache.openjpa.persistence.ArgumentException:
"x.y.z.EmbedOwner.embedCollection&lt;element:class x.y.z.EmbedValue&gt;"
is mapped as embedded, but embedded field
"x.y.z.EmbedOwner.embedCollection.x.y.z.EmbedValue.blob" is not embeddable.
Embedded element/key/value types are limited to simple fields and direct relations to other
persistent types"</literal> error may result. To overcome this limitation, either use JDBC driver
11.2.0.x.0 (or later version) or set both <literal>MaxEmbeddedBlobSize</literal> and
<literal>MaxEmbeddedClobSize</literal> properties to -1.
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_pointbase">
<title>
Pointbase
</title>
<example id="example_props_pointbase">
<title>
Example properties for Pointbase
</title>
<programlisting>
openjpa.ConnectionDriverName: com.pointbase.jdbc.jdbcUniversalDriver
openjpa.ConnectionURL: \
jdbc:pointbase:DB_NAME,database.home=pointbasedb,create=true,cache.size=10000,database.pagesize=30720
</programlisting>
</example>
<section id="dbsupport_pointbase_issues">
<title>
Known issues with Pointbase
</title>
<itemizedlist>
<listitem>
<para>
Fields of type BLOB and CLOB are limited to 1M. Set the <literal>BlobTypeName
</literal> and/or <literal>ClobTypeName</literal> properties of the
<literal>openjpa.jdbc.DBDictionary</literal> setting to override.
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_postgresql">
<title>
PostgreSQL
</title>
<example id="example_props_postgresql">
<title>
Example properties for PostgreSQL
</title>
<programlisting>
openjpa.ConnectionDriverName: org.postgresql.Driver
openjpa.ConnectionURL: jdbc:postgresql://SERVER_NAME:5432/DB_NAME
</programlisting>
</example>
<section id="dbsupport_postgresql_issues">
<title>
Known issues with PostgreSQL
</title>
<itemizedlist>
<listitem>
<para>
Floats and doubles may lose precision when stored.
</para>
</listitem>
<listitem>
<para>
PostgreSQL cannot store very low and very high dates.
</para>
</listitem>
<listitem>
<para>
Empty string/char values are stored as NULL.
</para>
</listitem>
<listitem>
<para>
Persistent fields of type <classname>java.io.Reader</classname> are not
supported when using
<link linkend="ref_guide_streamsupport">LOB streaming</link>.
</para>
</listitem>
</itemizedlist>
</section>
</section>
<section id="dbsupport_soliddb">
<title>
IBM solidDB
</title>
<example id="example_props_soliddb">
<title>
Example properties for IBM solidDB
</title>
<programlisting>
openjpa.ConnectionDriverName: solid.jdbc.SolidDriver
openjpa.ConnectionURL: jdbc:solid://localhost:2315/dba/dba
</programlisting>
</example>
<section id="dbsupport_soliddb_table_types">
<title>
M-type tables vs. D-type tables
</title>
<para>
IBM solidDB supports two types of tables: in-memory tables (M-tables) and
on-disk tables (D-tables). Since cursor hold over commit can not apply to M-tables
(which will cause SOLID Table Error 13187: The cursor cannot continue
accessing M-tables after the transaction has committed or aborted.
The statement must be re-executed), the default OpenJPA tables are D-tables.
One can set the whole server to disk-based mode by adding
[General]
DefaultStoreIsMemory=no
in solid.ini. The table types can also be determined by setting OpenJPA property
"openjpa.jdbc.DBDictionary" with value "storeIsMemory=true" or "storeIsMemory=false"
in the persistence.xml. The "STORE MEMORY" and "STORE DISK" will be appended to
the create table DDL, respectively.
</para>
</section>
<section id="dbsupport_soliddb_concurrency_control">
<title>
Concurrency control mechanism
</title>
<para>
The default concurrency control mechanism depends on the table type:
Disk-based tables (D-tables) are by default optimistic.
Main-memory tables (M-tables) are always pessimistic.
Since OpenJPA applications expects lock waits as usually is done with
normal pessimistic databases, the server should be set to the pessimistic mode.
The optimistic mode is about not waiting for the locks at all. That increases
concurrency but requires more programming. The pessimistic mode with the
READ COMMITTED isolation level (default) should get as much concurrency as one
might need. The pessimistic locking mode can be set in solid.ini:
[General]
Pessimistic=yes
One can override the locking mode on the per table base by setting OpenJPA property
"openjpa.jdbc.DBDictionary" to value "lockingMode=PESSIMISTIC" in the persistence.xml.
An extra SQL will be generated along with CREATE TABLE DDL:
ALTER TABLE EX_POBJECT SET PESSIMISTIC.
The possible values for lockingMode is OPTIMISTIC/PESSIMISTIC.
</para>
</section>
</section>
<section id="dbsupport_sybase">
<title>
Sybase Adaptive Server
</title>
<example id="example_props_sybase">
<title>
Example properties for Sybase
</title>
<programlisting>
openjpa.ConnectionDriverName: com.sybase.jdbc2.jdbc.SybDriver
openjpa.ConnectionURL: \
jdbc:sybase:Tds:SERVER_NAME:4100/DB_NAME?ServiceName=DB_NAME&amp;BE_AS_JDBC_COMPLIANT_AS_POSSIBLE=true
</programlisting>
</example>
<section id="dbsupport_sybase_issues">
<title>
Known issues with Sybase
</title>
<itemizedlist>
<listitem>
<para>
The "<literal>DYNAMIC_PREPARE</literal>" parameter of the Sybase JDBC driver
cannot be used with OpenJPA.
</para>
</listitem>
<listitem>
<para>
Datastore locking cannot be used when manipulating many-to-many relations using
the default OpenJPA schema created by the schematool, unless an auto-increment
primary key field is manually added to the table.
</para>
</listitem>
<listitem>
<para>
Persisting a zero-length string results in a string with a single space
character being returned from Sybase, Inc.'s JDBC driver.
</para>
</listitem>
<listitem>
<para>
The <literal>BE_AS_JDBC_COMPLIANT_AS_POSSIBLE</literal> is required in order to
use datastore (pessimistic) locking. Failure to set this property may lead to
obscure errors like " <literal>FOR UPDATE can not be used in a SELECT which is
not part of the declaration of a cursor or which is not inside a stored
procedure.</literal> ".
</para>
</listitem>
<listitem>
<para>
Applications performing update/insert data of the BigDecimal Java type may fail
with OptimisticException if the data exceeds the scale or precision of the
column on Sybase. To avoid this problem, applications can specify the precision
and scale for the numeric type by setting numericTypeName='NUMERIC(p,s)' for
the column type mapped by the BigDecimal Java type. See
<link linkend="openjpa.jdbc.DBDictionary">openjpa.jdbc.DBDictionary</link> for
more detail. Alternatively, application can set the precision and scale using
the standard <classname>Column</classname> annotation, described in
<xref linkend="jpa_overview_mapping_column"/>.
</para>
</listitem>
</itemizedlist>
</section>
</section>
</appendix>