All Forums Connectivity
BLE 2 posts Joined 06/11
17 Oct 2013
Hibernate, blob and large files

I use Hibernate to read and load data into a database Terdadata.
I have large files (xml) to be stored in a table. For that I use a BLOB field.
I did not encounter any problem to store files via BTEQ, but when I go through hibernate, I get the following error :

Caused by: com.teradata.jdbc.jdbc_4.util.JDBCException: [Teradata JDBC Driver] [TeraJDBC] [Error 1186] [SQLState HY000] Parameter 2 length is 129120 bytes, which is greater than the maximum 64000 bytes that can be set.


My table


ColumnName       Type                      Length                    Format

id                       INTEGER                     4                         -(10)9

binary_lob           BLOB               2 097 088 000            X(64000)
My settings :

<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE hibernate-configuration PUBLIC
"-//Hibernate/Hibernate Configuration DTD 3.0//EN"
" configuration-3.0.dtd">
	<session-factory name="sessionFactoryToa">
		<property name="hibernate.bytecode.use_reflection_optimizer" >false</property>
		<!--  Paramètres de connexion à la base de données -->
		<property name="hibernate.dialect" >org.hibernate.dialect.TeradataDialect</property>
		<property name="hibernate.connection.driver_class" >com.ncr.teradata.TeraDriver</property>
		<property name="hibernate.connection.url">jdbc:teradata:// my_dwh,CLIENT_CHARSET=cp936,TMODE=TERA,CHARSET=utf8,LO B_SUPPORT=on,LOB_TEMP_TABLE=toa.JdbcLobUpdate</ property>
		<property name="hibernate.connection.username">admin</property>
		<property name="hibernate.connection.password">admin</property>
		<property name="hibernate.current_session_context_class" >org.hibernate.context.ThreadLocalSessionContext</ property>
		<property name="hibernate.default_schema">my_dwh</property>
		<!-- Comportement pour la conservation des tables 
			validate: valide le schéma, aucune modification n’est faite sur la structure de la base ;
			update: met à jour le schéma existant ;
			create: crée le schéma en supprimant les données préalablement existantes ;
			create-drop: même comportement que create avec suppression du schéma en fin de session.-->
		<!-- <property name="">create</property> -->
		<!-- Configuration hibernate -->
		<property name="">false</ property>
		<property name="hibernate.show_sql">true</property>
		<property name="current_session_context_class">thread</property> 
		<property name="hibernate.generate_statistics">false</property>
		<property name="hibernate.use_sql_comments">false</property>
		<!-- Pool de connexion C3P0 -->
		<property name="hibernate.c3p0.validate">true</property>
		<property name="connection.provider_class" >org.hibernate.connection.C3P0ConnectionProvider</ property>
		<property name="hibernate.c3p0.min_size">2</property>
		<property name="hibernate.c3p0.max_size">20</property>
		<property name="hibernate.c3p0.timeout">14000</property>
		<property name="hibernate.c3p0.max_statements">50</property>
		<property name="hibernate.c3p0.preferredTestQuery">SELECT 1;</property> 
		<property name="hibernate.c3p0.testConnectionOnCheckout">true</ property>
		<!-- Mapping avec les bases -->
		<mapping resource="fr/trimane/TOA/bean/dwh/Dimension.hbm.xml" />
		<mapping resource="fr/trimane/TOA/bean/dwh/Fact1.hbm.xml" />
		<mapping resource="fr/trimane/TOA/bean/dwh/Fact2.hbm.xml" />
		<mapping resource="fr/trimane/TOA/bean/dwh/Fact3.hbm.xml" />
		<mapping resource="fr/trimane/TOA/bean/dwh/Fact4.hbm.xml" />
		<mapping resource="fr/trimane/TOA/bean/dwh/Fact10000.hbm.xml" />

Is it possible to load large files from Hibernate ?
I forgot something in my setting ?
Thank you

CarlosAL 512 posts Joined 04/08
18 Oct 2013

I don't know much about hibernate (I've learnt enough to hate it though) but, why does your 2MB BLOB column seem to have a 64KB format?

tomnolan 594 posts Joined 01/08
18 Oct 2013

Do not use connection parameter CLIENT_CHARSET=cp936 in conjunction with CHARSET=utf8. You are very likely to corrupt your data by doing that.
Generally speaking, it is a bad idea to use the CLIENT_CHARSET connection parameter. But if you must use it, then it should only be used in conjunction with CHARSET=ASCII or omitted CHARSET (the default is CHARSET=ASCII).

BLE 2 posts Joined 06/11
21 Oct 2013

My BLOB column can contain up to 2GB. 
For the 64K format, if I understand, this is the cutting perform Teradata.
I understood that Teradata stores the lob fields in another table. Store file is divided into packet of 64K.
I removed the parameter CLIENT_CHARSET, unfortunately I do not always manage to save my files :(

tomnolan 594 posts Joined 01/08
21 Oct 2013

The Teradata JDBC Driver Engineering team is investigating your issue. We will reply again after our investigation.

bchalal 1 post Joined 10/13
22 Oct 2013



I've same probleme with my DB on teradata, i wanna save Binary Files (BLOB) wich size exceed 100MB. 


My JDBC connexions are managed by EclipseLink and the config file contains the following :



<persistence-unit name="teradata">





	<property name="eclipselink.logging.level" value="INFO" />
	<property name="eclipselink.jdbc.driver" value="com.ncr.teradata.TeraDriver" />
	<property name="eclipselink.jdbc.url"value="jdbc:teradata://localhost/my_db,CLIENT_CHARSET=cp936,TMODE=TERA,CHARSET=utf8,LOB_SUPPORT=on" />
	<property name="eclipselink.jdbc.user" value="myValue" />
	<property name="eclipselink.jdbc.password" value="myValue" />
	<property name="eclipselink.ddl-generation" value="none" />
	<property name="eclipselink.ddl-generation.output-mode" value="database" />
	<property name="eclipselink.orm.throw.exceptions" value="true" />




tomnolan 594 posts Joined 01/08
23 Oct 2013

Do not use connection parameter CLIENT_CHARSET=cp936 in conjunction with CHARSET=utf8. You are very likely to corrupt your data by doing that.
Generally speaking, it is a bad idea to use the CLIENT_CHARSET connection parameter. But if you must use it, then it should only be used in conjunction with CHARSET=ASCII or omitted CHARSET (the default is CHARSET=ASCII).

drepshas 3 posts Joined 07/11
08 Nov 2013

Teradata is capable of supporting BLOBS over 64000 bytes but there is a 64000 byte limit when:

PreparedStatement.setBytes (int parameterIndex, byte [] x) 

is called and that is the limit you appear to be hitting.
This hibernate code from org.hibernate.type.descriptor.sql.BlobTypeDescriptor is where the decision to use setBytes is made when it sets the descriptor to  PRIMITIVE_ARRAY_BINDING as seen below:

public <X> BasicBinder<X> getBlobBinder(final JavaTypeDescriptor<X> javaTypeDescriptor) {
	return new BasicBinder<X>( javaTypeDescriptor, this ) {
		protected void doBind(PreparedStatement st, X value, int index, WrapperOptions options) throws SQLException {
			BlobTypeDescriptor descriptor = BLOB_BINDING;
			if ( byte[].class.isInstance( value ) ) {
				// performance shortcut for binding BLOB data in byte[] format
			else if ( options.useStreamForLobBinding() ) {
				descriptor = STREAM_BINDING;
			descriptor.getBlobBinder( javaTypeDescriptor ).doBind( st, value, index, options );

You can work around this by using

session.getLobHelper().createBlob(byte[] ba);

to create the Blob.
This will later access the same getBlobBinder code but will set the descriptor to STREAM_BINDING and bypass the 64000 byte restriction.

You must sign in to leave a comment.