SQLJ connection pool. How to implement/etc?

Hi,
Im currently using DefaultContext ctx = DefaultContext.getDefaultContext( ); I get one connection per session. But i would like to implement connection pooling. Can oracle handle this already. I have tried oracle.jdbc.pool and it doesnt seem to work. Do I have to write my own class for this? I have been researching for the last week and so many articles on how to do it, but no examples for sqlj.
Thanks

Thankyou for the feedback.
(I was reading sections 86 to 104 in the Developers Guide for Oracle Toplink to find out about internal connection pools)
postAcquireConnection
PUBLIC: This event is raised on when using the server/client sessions. This event is raised after a connection is acquired from a connection pool.
My understanding of this event is that it will execute every time some java code needs to obtain a connection from the pool to do a query on the database. Is there someway I can just apply the session settings on each database connection once, when the connection is created, rather than every time a connection is retrieved from the pool by a java thread. The overhead on the database will be very large, if every time a java thread performs a single database query and obtains a connection from the pool it has to execute my "alter session" command and then perform its, query before returning the connection back to the pool so that other java threads can use it.
Thanks

Similar Messages

Connection Pooling - how to close properly?? PLEASE HELP!

I am very confused about something and cannot find a clear answer within the Oracle docs or Sun's Java docs. When using connection pooling, you create an OracleConnectionPoolDataSource, obtain a PooledConnection, then a JDBC Connection (this is fine). But when finished, you close the JDBC Connection, but what about the PooledConnection? Do you close this too? Or do you just leave it and let it get garbage collected? Is the implementation smart enough to release/reuse PooledConnections when the underlying JDBC Connection is closed? It seems to me that if you close the PooledConnection too, then how/when does the pooling come into place? How/where are the pool parameters (max num of connections) controlled? Please Help! Thank you!

I have implemented a pooled JDBC connection & driver in a large project. So I know something about this topic.
java.sql.* is only interface, that means this package needs implementation. Under an ordinary circumstance, a Connection is an Client/Server communication, when you call close() method, the communication is really closed, so next time, you should reinstantiate the client/server connection, passing authentication and authorization testing, this is a lot of load to the system; connection pool uses a different mechanism to try to save this load, that is, when you call close() method of Connection, the connection is not really closed, but just set a flag to tell the system this connection is available to others, so next time when you call DriverManager.getConnection(...), the available connection will be returned to the caller, and because this connection has already passed authentication and aothorization, the socker between is available, so the extra load could be saved, in other words, the application could use it immediately.
Check java.sql.Driver interface, you will see boolean acceptsURL(String url) and java.sql.Connection connect(String url, java.util.Properties info) methods, first method is used by DriverManager to judge which JDBC driver should be used according to your connection's url format, then that driver's connect(...) method will be called to return an connection. For pooled connection, when you call connection.close() method, the Connection object is not garbage collected but return a pooled connection manager or repository, so it could be reused by other application.
Hope this helps.
//jdev

RENL & ENS Connection Pooling - How to ?

1.How do i acknowledge a notification from a subscriber ?
2.How do i enable ENS Connection Pooling [http://docs.sun.com/app/docs/doc/819-2655/6n4u96geu?a=view] ?
It would be better if somebody could give me code snippet also for the same.

I have implemented a pooled JDBC connection & driver in a large project. So I know something about this topic.
java.sql.* is only interface, that means this package needs implementation. Under an ordinary circumstance, a Connection is an Client/Server communication, when you call close() method, the communication is really closed, so next time, you should reinstantiate the client/server connection, passing authentication and authorization testing, this is a lot of load to the system; connection pool uses a different mechanism to try to save this load, that is, when you call close() method of Connection, the connection is not really closed, but just set a flag to tell the system this connection is available to others, so next time when you call DriverManager.getConnection(...), the available connection will be returned to the caller, and because this connection has already passed authentication and aothorization, the socker between is available, so the extra load could be saved, in other words, the application could use it immediately.
Check java.sql.Driver interface, you will see boolean acceptsURL(String url) and java.sql.Connection connect(String url, java.util.Properties info) methods, first method is used by DriverManager to judge which JDBC driver should be used according to your connection's url format, then that driver's connect(...) method will be called to return an connection. For pooled connection, when you call connection.close() method, the Connection object is not garbage collected but return a pooled connection manager or repository, so it could be reused by other application.
Hope this helps.
//jdev

Connection Pooling How To ?

Hi All ,
i am developing a struts based web-application with JBoss 4.0.5 as Application Server and Oracle 10g Express edition .
I am configuring my application for connection pooling .
I have tried following methods :
1. Oracle Connection Pooling following thius link http://www.lc.leidenuniv.nl/awcourse/oracle/java.920/a96654/connpoca.htm
2. Configuring Jboss's Internal Container Level Connection pooling simply creating the datasources
3. JDBC level Connection Pooling
but unluckily did not succeed in any of these method .
The application works OK with Second method but When it is being hit with 200 request/second The pool gets exhaulsted .
i am unable to decide which approach i should be using as my application can receive large no. of requests around 1000/sec ?
Please help me out ,
Tell some useful link if possible ..
Regards Jasdeep

The pool can't do magic. If your database operations take ten seconds each, you get 1000 requests a second, and you have a pool size of 200, the pool will effectively empty instantly.
You need to get some measurements of how long the typical transactions last and figure out an appropriate pool size for that. If it's more than the database can handle then you're going to have to upgrade.
It's that simple.

Connection Pooling: how many active connections?

Hi, everybody.
I have a very simple question about connection pooling...
How many active connections should a database see once a connection pool has been opened?
I mean, it should see only one connection (the pool itself) at any time, or the number of "logical" active connections in that moment?
Thanks for any answer.

Sorry...
This is the wrong forum, I posted again my question in the JDBC Forum...

How can i use connection pool within the sqlj?

hello
i am a beginner to sqlj,i find all of the sample code from the oracle DON'T use connectionpool,they only write the database url to "connection.properties" file,but i think in the production environment we should use the connectionpool to optimise the querying,i usually use the weblogic connection pool.
how can i use the connection pool within the sqlj?
thank you very much!

Im not sure if I understand. :)
English is not my best language...
Looking up the connection again? Do you mean I have to create new DataSource object and bind it again to ServletContext?

[Fwd: Re: rdbms realm and connection pool]

Hi,
One reason why I would like to use the connection pool for the RDBMS
realm is because there is the retry machanism built into the connection
pool. With this retry, I don't need to re-start WebLogic if the DB
server is somehow re-started. With the current implementation, all the
connections maintained by the realm will become invalid if the DB server
has been restarted independently.
-------- Original Message --------
Subject: Re: rdbms realm and connection pool
Date: Wed, 27 Sep 2000 09:32:47 +0100
From: "Terry" <[email protected]>
Reply-To: "Terry" <[email protected]>
Organization: BEA SYSTEMS Inc
Newsgroups: weblogic.developer.interest.security
References: <[email protected]>
I believe not- the realm restricts access to connection pools to those
who
are allowed it, so if the realm needs the connection pool to start up,
and
you can't open the connection pool without the realm then you have a bit
of
a no-chicken and no-egg situation, which is I believe one of the reasons
why
there is no use of connection pools, ejbs, jndi, servlets etc. in the
realm
(along with other reasons, like why would it be provided with a servlet)
The delegate pool acts somewhat similarly to a connection pool, and can
even
use the same database, so I'm not sure what the advantage would be
Terry
Nirmala devi <[email protected]> wrote in message
news:[email protected]..
>
I think the rdbms realm uses different connection as it need to be setbefore
the connection pool for Database.Is there any that i can point my rdbmsrealm to use
the connection pool for Database instead
Thanks in advance
Nirmala

I believe not- the realm restricts access to connection pools to those who
are allowed it, so if the realm needs the connection pool to start up, and
you can't open the connection pool without the realm then you have a bit of
a no-chicken and no-egg situation, which is I believe one of the reasons why
there is no use of connection pools, ejbs, jndi, servlets etc. in the realm
(along with other reasons, like why would it be provided with a servlet)
The delegate pool acts somewhat similarly to a connection pool, and can even
use the same database, so I'm not sure what the advantage would be
Terry
Nirmala devi <[email protected]> wrote in message
news:[email protected]..
>
I think the rdbms realm uses different connection as it need to be setbefore
the connection pool for Database.Is there any that i can point my rdbmsrealm to use
the connection pool for Database instead
Thanks in advance
Nirmala

What's the difference between using a connection pool and a datasource

Howdy. I figure this is a newbie question, but I can't seem to find an
answer.
In the docs at bea, the datasource docs say
"DataSource objects provide a way for JDBC clients to obtain a DBMS
connection. A DataSource is an interface between the client program and the
connection pool. Each data source requires a separate DataSource object,
which may be implemented as a DataSource class that supports either
connection pooling or distributed transactions."
In there it says the datasource uses the connection pool, but other than
that, what is the difference between a connection pool and a datasource?

Thanks for the info. I think it makes some sense. But it's a bit greek.
I'm sure it'll make more sense the more I work with it. Thanks.
"Chuck Nelson" <[email protected]> wrote in message
news:3dcac1f5$[email protected]..
>
Peter,
Here is a more formal definition of a DataSource from the Sun site
"A factory for connections to the physical data source that thisDataSource object
represents. An alternative to the DriverManager facility, a DataSourceobject
is the preferred means of getting a connection. An object that implementsthe
DataSource interface will typically be registered with a naming servicebased
on the JavaTM Naming and Directory (JNDI) API.
The DataSource interface is implemented by a driver vendor. There arethree types
of implementations:
Basic implementation -- produces a standard Connection object
Connection pooling implementation -- produces a Connection object thatwill automatically
participate in connection pooling. This implementation works with amiddle-tier
connection pooling manager.
Distributed transaction implementation -- produces a Connection objectthat may
be used for distributed transactions and almost always participates inconnection
pooling. This implementation works with a middle-tier transaction managerand
almost always with a connection pooling manager.
Does that help clarify the distinction?
Chuck Nelson
DRE
BEA Technical Support

Connection Pool with Variables

Hi i am a newbie to Obiee and i want to know how to import tables into a connection pool with variables in place of Data Source Name: valueof(DSN_DM) username:value of(DSN_DM_USER)
like that i want to import my new tables into this connection pool how i can do that. i tried importing but it is coming as a new subject area.
Can any Body help me
Thanks

Hi copter,
Firstly you imported some tables and again as adding tables your are using variable in connection pool and importing tables then it takes as a new subject area only.....Instead use this variable connection pool to import all tables from scratch and then work on it.This would do.
Either if it is a seperate subject area pull the columns needed from both the subject areas and work on it.Before doing that you need to establish joins between the subject area tables.
http://forums.oracle.com/forums/thread.jspa?threadID=1123247&tstart=-1
hope helps you.
Cheers,
KK

How to implement a connection pooling in servlet

hi all,
how to implement a connection pooling in servlet.i want to know how to implement it in tomcat in a struts based environment.
any input to the topic is appreciated.
Thanks
Sudha

Take a look at JNDI
http://jakarta.apache.org/tomcat/tomcat-4.1-doc/jndi-datasource-examples-howto.html

How can I implement the connection pool in my java stored procedure

my java stored procedures (in database 'B') have to connect to another oracle database ,let's say 'A'. And how can I implement the behavior like the so-called connection pool in my java stored procedure in 'B', as below.
1. database B, has 2 java stored procedures sp1 and sp2
2. both sp1 and sp2 connects to databse 'A'
whatever I call the sp1 and sp2 and the database 'A' always only one connected session from sp1 and sp2 in database 'B'.
THANKS A LOTS...

my problem is I have a lots of java stored procedures need to cnnect to the remote oracle db, and I hope the remote db can only have a connected session from my java stored procedures in my local db. I try as below
class sp{
static Connection conn=null; //the remote db connection,
public static void sp1(){...}//procedure 1, using conn
public static void sp2(){...}//procedure 2, using conn,too
I can 'see' the 'conn' variable if I invoke the sp1() and sp2() from the same client application(maybe sqlplus). But if I invoke the sp1() from client 'A' and invoke sp2() from client 'B' then the sp1() and sp2() can not see the 'conn' variable for each other. I think it's because the two clients cause oracle to create two instances of the class 'sp' and the sp1() and sp2() located in different instance seperately. Thus the sp1() and sp2() can not see 'conn' for each other. They can only see its own 'conn'.
To connect to the remote db from the java stored procedure is easy but is it possible to connect to the remote db via database link from the java stored procedure at my local db ? If so, then I also archive my goal .
BTW , thanks a lots...
andrew :-)

How to implement a Connection Pool?

Hello.
I need to create a class that implements a connection pool to an Oracle database, so that if from any other class someone ask for a connection then my new class has to supply it.
The the jdbc driver I am trying is ""jdbc:oracle:thin" release "10.2.0.4.0".
Could someone, please show me how I should do it?
Thanks in advance.

fsanchezherrero wrote:
I need to create a small java application that just has a method that calls a stored procedure in the database. Connections must be pooled, that's all I know.
Thanks for answering.This really doesn't help at all.
Is this a school homework assignment? Or something you are doing for the "real" world? If it's the former and the idea is for you to write your connection pool (as odd as that is) then fine. If it's the latter then you are making a very big mistake and should stop now* until you get your real requirements figured out.
So which is it?

How do I expand my simple JSP page to implement Connection Pooling?

Hi everyone,
I have a small SQL database which contains information about Students and a StudenBean which contains their attributes and appropriate get/set methods (which I haven't needed to use yet).
I want to write a simple JSP page which displays all the students details in one big table. I can do this using the basic prototyping method of adding the dataSource and driver details to the deployment descriptor (web.xml - I'm using the latest version of Tomcat by the way) and then accessing the database using JSP standard SQL actions.
Here's a snipet of what I've done (just an example, the final code works):
<sql:query var="temp"
sql="SELECT * FROM Employee ORDER BY UserName"
/>
<c:forEach items="${temp.rows}" var="row">
<td><c:out value ="${row.UserName}"/></td>
<td><c:out value ="${row.FirstName}"/></td>
<td><c:out value ="${row.LastName}"/></td>
Now I want to upgrade! I want to access the dataSource using Connection Pooling and it's causing me a lot of trouble. I already have the required working classes/data such as ConnectionPool.java, ResourceManagerListener.java, the MySQL driver etc.
However, I don't know what the best way to setup the connection for my purposes would be and where I should do it.
Do I need to write a servlet or tag handler/custom action to establish the connection and access my bean through it? I would like to keep a similar sort of business logic as I already have developed i.e. be able to use actions in my JSP page to cycle through the database and print out each detail from the row.
Rather confused and muddled. Thanks for any input!

Good idea.
Set up a JNDI data source for your app, according to these docs:
http://jakarta.apache.org/tomcat/tomcat-4.1-doc/jndi-datasource-examples-howto.html
You'll have to put the right information into your web.xml and a context.xml. Once you've done that, your JSTL JSPs will simply use the connection resource automagically.
MOD

How to create Connection pool in Tomcat 5

present i am developing one web site in that i am using tomcat 5.
i am trying to establish connection pooling in tomcat,i am facing some problems..
tomcat throws the following exception ie.,
cannot create JDBC Driver class for connect URL null.
my code lik this
i configured the web.xml
<?xml version="1.0" encoding="ISO-8859-1"?>
<web-app>
<resource-ref>
<description>Oracle Datasource example</description>
<res-ref-name>sqlserver</res-ref-name>
<res-type>javax.sql.DataSource</res-type>
<res-auth>Container</res-auth>
</resource-ref>
</web-app>
my server.xml
<?xml version='1.0' encoding='utf-8'?>
<Server>
<Listener className="org.apache.catalina.mbeans.ServerLifecycleListener"/>
<Listener className="org.apache.catalina.mbeans.GlobalResourcesLifecycleListener"/>
<GlobalNamingResources>
<Environment name="simpleValue" type="java.lang.Integer" value="30"/>
<Resource auth="Container" description="User database that can be updated and saved" name="UserDatabase" type="org.apache.catalina.UserDatabase"/>
<Resource auth="Container" name="sqlserver" type="javax.sql.DataSource"/>
<ResourceParams name="UserDatabase">
<parameter>
<name>factory</name>
<value>org.apache.catalina.users.MemoryUserDatabaseFactory</value>
</parameter>
<parameter>
<name>password</name>
<value>sudha</value>
</parameter>
<parameter>
<name>pathname</name>
<value>conf/tomcat-users.xml</value>
</parameter>
<parameter>
<name>username</name>
<value>sudha</value>
</parameter>
</ResourceParams>
<ResourceParams name="sqlserver">
<parameter>
<name>url</name>
<value>jdbc:odbc:SQL</value>
</parameter>
<parameter>
<name>validationQuery</name>
<value>select * from employee</value>
</parameter>
<parameter>
<name>maxIdle</name>
<value>4</value>
</parameter>
<parameter>
<name>maxActive</name>
<value>8</value>
</parameter>
<parameter>
<name>driverClassName</name>
<value>sun.jdbc.odbc.JdbcOdbcDriver</value>
</parameter>
<parameter>
<name>maxWait</name>
<value>5000</value>
</parameter>
<parameter>
<name>username</name>
<value>kpi</value>
</parameter>
<parameter>
<name>factory</name>
<value>org.apache.naming.factory.DbcpDataSourceFactory</value>
</parameter>
<parameter>
<name>password</name>
<value>kpi</value>
</parameter>
</ResourceParams>
</GlobalNamingResources>
<Service name="Catalina">
<Connector acceptCount="100" connectionTimeout="20000" disableUploadTimeout="true" port="8080" redirectPort="8443" maxSpareThreads="75" maxThreads="150" minSpareThreads="25">
</Connector>
<Connector port="8009" protocol="AJP/1.3" protocolHandlerClassName="org.apache.jk.server.JkCoyoteHandler" redirectPort="8443">
</Connector>
<Engine defaultHost="localhost" name="Catalina">
<Host appBase="webapps" name="localhost">
<Logger className="org.apache.catalina.logger.FileLogger" prefix="localhost_log." suffix=".txt" timestamp="true"/>
</Host>
<Logger className="org.apache.catalina.logger.FileLogger" prefix="catalina_log." suffix=".txt" timestamp="true"/>
<Realm className="org.apache.catalina.realm.UserDatabaseRealm"/>
</Engine>
</Service>
</Server>
and i studied the the tomcat documentation in that tomcat mentioned
The configuration properties for Tomcat's standard data source resource factory (org.apache.naming.factory.DbcpDataSourceFactory) are as follows:
driverClassName - Fully qualified Java class name of the JDBC driver to be used.
maxActive - The maximum number of active instances that can be allocated from this pool at the same time.
maxIdle - The maximum number of connections that can sit idle in this pool at the same time.
maxWait - The maximum number of milliseconds that the pool will wait (when there are no available connections) for a connection to be returned before throwing an exception.
password - Database password to be passed to our JDBC driver.
url - Connection URL to be passed to our JDBC driver. (For backwards compatibility, the property driverName is also recognized.)
user - Database username to be passed to our JDBC driver.
validationQuery - SQL query that can be used by the pool to validate connections before they are returned to the application. If specified, this query MUST be an SQL SELECT statement that returns at least one row.
how to configure server.xml file
and how to get the org.apache.naming.factory.DbcpDataSourceFactory
please help me...this module is very very urgen to me..

I spent many hours when i first implemented connection pooling trying to put it together. The information is scattered and the config has changed slightly from 5.0 to 5.5 etc. This is how I got it working in 5.5.17:
* server.xml seems to be a wrong place for the pool configuration if you consider it to be an application specific thing. It might be a better idea to put it into C:\tomcat55\conf\Catalina\localhost\ConfigFileForYourWeapp.xml as follows:
<Context path="/ClientLink" docBase="C:\eclipseWorkspace\PortfolioUI3\WebRoot">
      <Resource name="jdbc/testportfolios" auth="Container"
                     type="javax.sql.DataSource"
                     factory="org.apache.tomcat.dbcp.dbcp.BasicDataSourceFactory"
                     username="****" password="******"
                     driverClassName="com.mysql.jdbc.Driver"
                     url="jdbc:mysql://192.168.0.101:3306/testportfolios?user=***********&password=***********&useUnicode=true&characterSet=swedish"
                     maxWait="10000" maxActive="30" maxIdle="10" removeAbandoned="true"
                     removeAbandonedTimeout="60" logAbandoned="true" testWhileIdle="true"
                     minEvictableIdleTimeMillis="4000"
                     timeBetweenEvictionRunsMillis="60000" validationQuery="SELECT '1'" />
      </Context>* Then add the following snippet into web.xml in your webapp directory:
<resource-ref>
                <description>Database pool defined in ClientLink.xml</description>
                <res-ref-name>jdbc/testportfolios</res-ref-name>
                <res-type>javax.sql.DataSource</res-type>
                <res-auth>Container</res-auth>
           </resource-ref>* Next put the jdbc-driver jar in a place where tomcat can find it. That is common\lib. Note that the shared/lib does not work.
* Get connections in java as follows:
                try {
                     Context initCtx = new InitialContext();
                     DataSource ds = (DataSource) initCtx.lookup("java:comp/env/jdbc/testportfolios");
                     this.connection = ds.getConnection();
                } catch (NamingException e) {
                     throw new RuntimeException("Java naming exception when getting connection from tomcat pool: " + e.getMessage());
                }I hope that this helps you to save the time I lost figuring this out!
Message was edited by:
Tomppu

How to use JDBC Connection Pools in a standalone application?

Hi, there,
I have a question about how to use JDBC Connection Pools in an application. I know well about connection pool itself, but I am not quite sure how to keep the pool management object alive all the time to avoid being destroyed by garbage collection.
for example, at the website: http://www.developer.com/java/other/article.php/626291, there is a simple connection pool implementation. there are three classes:JDBCConnection, the application's gateway to the database; JDBCConnectionImpl, the real class/object to provide connection; and JDBCPool, the management class to manage connection pool composed by JDBCConnectionImpl. JDBCPool is designed by Singleton pattern to make sure only one instance. supposing there is only one client to use connection for many times, I guess it's ok because this client first needs instantiate JDBCPool and JDBCConnectionImpl and then will hold the reference to JDBCPool all the time. but how about many clients want to use this JDBCPool? supposing client1 finishes using JDBCPool and quits, then JDBCPool will be destroyed by garbage collection since there is no reference to it, also all the connections of JDBCConnectionImpl in this pool will be destroyed too. that means the next client needs recreate pool and connections! so my question is that if there is a way to keep pool management instance alive all the time to provide connection to any client at any time. I guess maybe I can set the pool management class as daemon thread to solve this problem, but I am not quite sure. besides, there is some other problems about daemon thread, for example, how to make sure there is only one daemon instance? how to quit it gracefully? because once the whole application quits, the daemon thread also quits by force. in that case, all the connections in the pool won't get chance to close.
I know there is another solution by JNDI if we develop servlet application. Tomcat provides an easy way to setup JNDI database pooling source that is available to JSP and Servlet. but how about a standalone application? I mean there is no JNDI service provider. it seems a good solution to combine Commons DBCP with JNID or Apache's Naming (http://jakarta.apache.org/commons/dbcp/index.html). but still, I don't know how to keep pool management instance alive all the time. once we create a JNDI enviroment or naming, if it will save in the memory automatically all the time? or we must implement it as a daemon thread?
any hint will be great apprieciated!
Sam

To my knoledge the pool management instance stays alive as long as the pool is alive. What you have to figure out is how to keep a reference to it if you need to later access it.

SQLJ connection pool. How to implement/etc?

Similar Messages

Maybe you are looking for