KEMBAR78
Application Continuity with Oracle DB 12c | ODP
LG
Uploaded byLéopold Gault
ODP, PPTX484 views

Application Continuity with Oracle DB 12c

The document consists of personal notes and drawings by Leopold Gault on application continuity, specifically regarding JDBC drivers, connection pools, and how applications manage connections, particularly in WebLogic. It discusses the mechanics of application continuity, the role of JDBC replay drivers, and how they allow for uninterrupted database transactions even after connection disruptions. The practical implementation of these concepts is illustrated with code examples and configurations for maintaining connection pools and request boundaries.

Download as ODP, PPTX
Understanding Application Continuity ● Those slides are Leopold Gault's personal notes and drawings, about application continuity. ● You can see my sources, and comments in the slides' notes. ● You may ask me for the original file at galeopold@gmail.com ● Views expressed here are my own and do not necessarily reflect the views of Oracle.
Understand Application Continuity I. javax.sql.DataSource implementations 1) Understanding JDBC drivers 2) Connection pools and JDBC drivers 3) Weblogic Data Sources and JDBC drivers 4) How an app gets a connection a from Welogic Data Source II. Application Continuity 1) How AC works ● What happens at run-time ● What happens when the connection is interrupted 2) Demo ● AC while fetching data from the cursor of a result set ● AC while performing a local transation 3) Notes About AC ● Notes about Global Transactions
javax.sql.DataSource Understanding Data Sources and JDBC drivers
JVM Used by the app javax.sql.DataSource interface
JVM connection factory class a.k.a. JDBC driver Implements javax.sql.DataSource interface Used by the app « Connection facrory » or « JDBC driver » ? Features of the JDBC driver Connection Factory feature
JVM JDBC connection JDBC driver Implements javax.sql.DataSource interface .getConnection()
JVM Connection pool JDBC driver Implements DataSource interface Often includes Read the notes to access my sources
Connection Pool Its interface Extends javax.sql.DataSource JVM Standalone JDBC app JDBC driver Implements DataSource interface Actually, I think that the JDBC driver can be used both ● in a standalone manner (the app is then called « standalone JDBC application ») ● or configured as the connection factory of a connection pool, such as UCP or Weblogic Server connection pool. source Directly uses Jdbc driver Implements javax.sql.DataSource JVM
Middle-tier JVM The UCP oracle.ucp.jdbc.PoolDataSourceImpl is used to manipulate the JDBC driver (a.k.a. connection factory). The app doesn't directly asks the JDBC driver for a connection. It asks the connection pool. PoolDataSource pds = PoolDataSourceFactory.getPoolDataSource(); pds.setConnectionFactoryClassName("oracle.jdbc.pool.OracleDataSource"); System.out.println("connection factory set"); String URL = "jdbc:oracle:thin:@(DESCRIPTION = (TRANSPORT_CONNECT_TIMEOUT=3) (RETRY_COUNT=20)(FAILOVER=ON) " + (ADDRESS = (PROTOCOL = TCP)(HOST = rac12scan.gns.example.com)(PORT = 1521)) (CONNECT_DATA = " + (SERVER = DEDICATED) (SERVICE_NAME = ACTEST)))"; System.out.println("Using URLn" + URL); pds.setURL(URL); pds.setUser("martin"); pds.setPassword("***"); pds.setInitialPoolSize(10); pds.setMinPoolSize(10); pds.setMaxPoolSize(20); // RAC Features pds.setConnectionPoolName("Application Continuity Pool"); pds.setFastConnectionFailoverEnabled(true); // use srvctl config nodeapps to get the ONS ports on the cluster pds.setONSConfiguration("nodes=192.168.100.30:6200,192.168.100.32:6200"); System.out.println("pool configured, trying to get a connection"); Connection conn = pds.getConnection(); if (conn == null || !((ValidConnection) conn).isValid()) { System.out.println("connection is not valid"); throw new Exception ("invalid connection obtained from the pool"); } source JDBC driver class name Universal Connection Pool Universal Connection Pool Its interface Extends javax.sql.DataSource Jdbc driver Implements javax.sql.DataSource
Weblogic Weblogic Data Source Connection Pool Its interface extends DataSource interface[2] I assume that the concept is the same for Weblogic Data Source Connection Pool: it is a class that manipulates the JDBC driver in order to maintain a connection pool. Looked-up by the app thanks to its JNDI name Weblogic Data Source calling it « Weblogic Server Connection Pool » would make more sense! [1] Jdbc driver Implements javax.sql.DataSource
Weblogic Weblogic Data Source Its interface extends DataSource interface Java app The app looks up the Data Source thanks to its JNDI name 1) import javax.sql.DataSource; public class dbUtilities { public dbUtilities() { super(); } public static final Long getDatabaseId (String seqName) { DataSource dataSource = null; Context ic = null; PreparedStatement stmt = null; ResultSet rs = null; Long dbId = null; String selectString; selectString = "SELECT " + seqName + ".nextval FROM DUAL"; try { ic = new InitialContext(); dataSource = (DataSource) ic.lookup("jdbc/AlphaOfficeAccessDS"); } catch (NamingException e) { } System.out.println("lookup dataSource returned " + dataSource); try { Connection connection = dataSource.getConnection(); ,,,,, Jdbc driver Implements javax.sql.DataSource
Weblogic Java app The app asks the Data Source for a new connection 2) Weblogic Data Source Its interface extends javax.sql.DataSource interface Jdbc replay driver Implements javax.sql.DataSource
Weblogic Java app 3) Weblogic Data Source Its interface extends javax.sql.DataSource interface Jdbc replay driver Implements javax.sql.DataSource A connection is given from the Connection pool. This is called a check-out
Weblogic Java app The app can now access the DB 4) Weblogic Data Source Its interface extends javax.sql.DataSource interface Jdbc replay driver Implements javax.sql.DataSource
Weblogic Java app Everything (all the database calls) that happens on the connection object between the check-out and the check-in is called a request [1] Weblogic Data Source Its interface extends javax.sql.DataSource interface Jdbc replay driver Implements javax.sql.DataSource Check-out Check-in
Understand Application Continuity I. javax.sql.DataSource implementations 1) Understanding JDBC drivers 2) Connection pools and JDBC drivers 3) Weblogic Data Sources and JDBC drivers 4) How an app gets a connection a from Welogic Data Source II. Application Continuity 1) How AC works ● What happens at run-time ● What happens when the connection is interrupted 2) Demo ● AC while fetching data from the cursor of a result set ● AC while performing a local transation 3) Notes About AC ● Notes about Global Transactions
2] Application Continuity How to use the JDBC replay driver
In order to benefit from Application Continuity, you have to get a connection from the JDBC replay driver, whether it is usued by a connection pool (Weblogic JDBC Data Source, UCP, or 3rd party connection pool), or directly used by the application (JDBC standalone app) Connection Pool Its interface Extends javax.sql.DataSource JVM Standalone JDBC app JDBC driver Implements DataSource interface Directly uses Jdbc driver Implements javax.sql.DataSource JVM
Weblogic In order to benefit from Application Continuity, you have to get a connection from the JDBC replay driver, whether it is usued by a connection pool (Weblogic JDBC Data Source, UCP, or 3rd party connection pool), or directly used by the application (JDBC standalone app) Weblogic Data Source Its interface extends DataSource interface Jdbc driver Implements javax.sql.DataSource
JVM PoolDataSource pds = PoolDataSourceFactory.getPoolDataSource(); pds.setConnectionFactoryClassName("oracle.jdbc.replay.OracleDataSourceImpl"); System.out.println("connection factory set"); String URL = "jdbc:oracle:thin:@(DESCRIPTION = (TRANSPORT_CONNECT_TIMEOUT=3) (RETRY_COUNT=20) (FAILOVER=ON) " + (ADDRESS = (PROTOCOL = TCP)(HOST = rac12scan.gns.example.com)(PORT = 1521)) (CONNECT_DATA = " + (SERVER = DEDICATED) (SERVICE_NAME = ACTEST)))"; System.out.println("Using URLn" + URL); pds.setURL(URL); pds.setUser("martin"); pds.setPassword("***"); pds.setInitialPoolSize(10); pds.setMinPoolSize(10); pds.setMaxPoolSize(20); // RAC Features pds.setConnectionPoolName("Application Continuity Pool"); pds.setFastConnectionFailoverEnabled(true); // use srvctl config nodeapps to get the ONS ports on the cluster pds.setONSConfiguration("nodes=192.168.100.30:6200,192.168.100.32:6200"); System.out.println("pool configured, trying to get a connection"); Connection conn = pds.getConnection(); if (conn == null || !((ValidConnection) conn).isValid()) { System.out.println("connection is not valid"); throw new Exception ("invalid connection obtained from the pool"); } source JDBC replay driver class name In order to benefit from Application Continuity, you have to get a connection from the JDBC replay driver, whether it is usued by a connection pool (Weblogic JDBC Data Source, UCP, or 3rd party connection pool), or directly used by the application (JDBC standalone app) Universal Connection Pool Jdbc driver Implements javax.sql.DataSource
Note that if you use connections that do not come from an Oracle connection pool (Weblogic Data Source or UCP, etc.), but instead come from a third party connection pool, you have to add beginRequest and endRequest APIs to the application's own connection pool to identify request boundaries. [1] import java.sql.Connection; import javax.sql.PooledConnection; import oracle.jdbc.OracleConnection; import oracle.jdbc.replay.OracleDataSourceFactory; import oracle.jdbc.replay.OracleDataSource; import oracle.jdbc.replay.OracleConnectionPoolDataSource; ... { ...... OracleDataSource rds = OracleDataSourceFactory.getOracleDataSource(); rds.setUser(user); rds.setPassword(passwd); rds.setURL(url); ...... // Other data source configuration like callback, timeouts, etc. Connection conn = rds.getConnection(); ((OracleConnection) conn).beginRequest(); // Explicit request begin ...... // JDBC calls protected by Application Continuity ((OracleConnection) conn).endRequest(); // Explicit request end conn.close(); OracleConnectionPoolDataSource rcpds = OracleDataSourceFactory.getOracleConnectionPoolDataSource(); rcpds.setUser(user); rcpds.setPassword(passwd); rcpds.setURL(url); ...... // other data source configuration like callback, timeouts, and so on PooledConnection pc = rcpds.getPooledConnection(); Connection conn2 = pc.getConnection(); // Implicit request begin ...... // JDBC calls protected by Application Continuity conn2.close(); // Implicit request end source JDBC replay driver interface name Here we use the replay-driver directly so we have to add beginRequest() Here we use an Oracle connection pool, so no need to add beginRequest()
JVM JDBC replay driver Implements DataSource interface Standalone jdbc app directly uses Note that if you use connections that do not come from an Oracle connection pool (Weblogic Data Source or UCP, etc.), but instead come from a third party connection pool, you have to add beginRequest and endRequest APIs to the application's own connection pool to identify request boundaries. [1] import java.sql.Connection; import javax.sql.PooledConnection; import oracle.jdbc.OracleConnection; import oracle.jdbc.replay.OracleDataSourceFactory; import oracle.jdbc.replay.OracleDataSource; import oracle.jdbc.replay.OracleConnectionPoolDataSource; ... { ...... OracleDataSource rds = OracleDataSourceFactory.getOracleDataSource(); rds.setUser(user); rds.setPassword(passwd); rds.setURL(url); ...... // Other data source configuration like callback, timeouts, etc. Connection conn = rds.getConnection(); ((OracleConnection) conn).beginRequest(); // Explicit request begin ...... // JDBC calls protected by Application Continuity ((OracleConnection) conn).endRequest(); // Explicit request end conn.close(); OracleConnectionPoolDataSource rcpds = OracleDataSourceFactory.getOracleConnectionPoolDataSource(); rcpds.setUser(user); rcpds.setPassword(passwd); rcpds.setURL(url); ...... // other data source configuration like callback, timeouts, and so on PooledConnection pc = rcpds.getPooledConnection(); Connection conn2 = pc.getConnection(); // Implicit request begin ...... // JDBC calls protected by Application Continuity conn2.close(); // Implicit request end source JDBC replay driver interface name Note that if you use the jdbc replay driver in a stand alone manner, you also have to add beginRequest/endRequest [2] Here we use the replay-driver directly so we have to add beginRequest() Here we use an Oracle connection pool, so no need to add beginRequest()
What happens with the replay driver and DB 12c at runtime [source] 2] Application Continuity
Weblogic Weblogic Data Source Java app JDBC driver Implements DataSource interface Used to create connections Connection pool The replay-driver monitors and gets information from ● the connection object ● the continuity director (that sits inside the database). It stores any relevant information for a replay, until the end of the request (when the connection is sent back to the connection pool). Continuity director
Weblogic Weblogic Data Source Java app JDBC driver Implements DataSource interface Used to create connections Connection pool Specifically, the replay- driver saves all the PL/SQL statements that the continuity director has instructed it to store. It saves them into the replay- context. For each call made to the DB, the replay driver retains the context required for a subsequent replay. [1] Replay context PL/SQL statements Continuity director
Weblogic Weblogic Data Source Java app JDBC driver Implements DataSource interface Used to create connections Connection pool The continuity director also stores protocol and validation information. It stores this information locally, on what seems to be called the « replay context » too. However, it is probably quite different from the replay context of the JDBC- replay-driver.Protocol and validation information ? Replay context ? Continuity director
Tx Guard Weblogic Weblogic Data Source Java app JDBC driver Implements DataSource interface Used to create connections Connection pool Moreover, I think that both the JDBC replay driver and the Continuity director store the logical transaction IDs (LTXID) of recent transactions. The LTXID is provided by the Transaction Guard feature of the 12c DB. It allows to identify logically a transaction that has been committed, and to know its outcome (committed and completed or rolledback). Transaction Guard therefore stores a history of past transactions, with their known outcome. Continuity director LTXID SQL statements XXXX INSERT .. DELETE .. COMMIT YYYY CREATE
We interrupt the connection by performing a switchover Application Continuity
Data Guard Weblogic Weblogic Data Source Java app JDBC driver Implements DataSource interface Used to create connections Standby DB Connection pool Primary DB We switchover A switchover is seen from an external point of view as if all the DB sessions were killed
Data Guard Weblogic Weblogic Data Source Java app JDBC driver Implements DataSource interface Used to create connections Standby DB Connection pool Primary DB The app can now access the DB The replay-driver intercepts execution errors. When the error is recoverable, the replay- driver replays the calls that were performed since the beginning of the request. It therefore restores : ● the cursors as they were (same position in the result set), ● and the latest transaction that was being build up, as it was.
Data Guard Weblogic Weblogic Data Source Java app JDBC driver Implements DataSource interface Used to create connections Standby DB Connection pool Primary DB The app can now access the DB Thanks to the replay of the replay-driver, new DB sessions have been created on the new-primary DB. Those re-created DB- sessions don't necessarily have the same SID and serial number (cf next slide). But the application keeps on using the same connection object (it does not check-out a new one), so to me, this means that the replay driver makes the replacement of the old connection object by a new one invisible, by registering the new connection object at the same reference variable. I guess the replay driver does the same for the connections in the connection pool. By the way, I think that the mechanism is the same for all the objects got out of the connection object (such as result-set cursors) : they are replaced by the replay driver and registered at the same reference variable ; so that the application beleive it's the same object.
Before we kill all the sessions, here is the (SID, Serial number) couple of all the existing database sessions for our application (SwingBench) After we killed the sessions, new sessions are created to replace the killed ones. They don't necessarily have the same (SID, Serial number) couple Application continuity when killing all the db sessions on a single database (no RAC nor Data Guard)
Weblogic Weblogic Data Source Java app JDBC driver Implements DataSource interface Used to create connections Connection pool Replay context PL/SQL statements Continuity director Protocol and validation information ? Replay context ? The request is replayed by the driver, using the replay context in order to have the same context as when the request was originally played. The continuity director decides whether or not it accepts a replay ; both before and after the replay (I think). If the replay is successful, the expected response will be given to the app. It will look as if there had been a small time delay.
Tx Guard Weblogic Weblogic Data Source Java app JDBC driver Implements DataSource interface Used to create connections Connection pool Continuity director LTXID SQL statements XXXX INSERT .. DELETE .. COMMIT YYYY CREATE « Has transaction XXXX been successfully committed ? » [About transactions] In order to replay the most recent transaction that was asserted during the request, the replay-driver asks Transaction Guard if the transaction has been successfully committed, using its logical transaction ID. If it was not, the replay driver can replay the transaction. This process prevents from committing two times the same transaction, and thus result with data inconsistency.
Understand Application Continuity I. javax.sql.DataSource implementations 1) Understanding JDBC drivers 2) Connection pools and JDBC drivers 3) Weblogic Data Sources and JDBC drivers 4) How an app gets a connection a from Welogic Data Source II. Application Continuity 1) How AC works ● What happens at run-time ● What happens when the connection is interrupted 2) Demo ● AC while fetching data from the cursor of a result set ● AC while performing a local transation 3) Notes About AC ● Notes about Global Transactions
Demo
Scenario 1 : Restoring an active cursor Demo
$response = $db->query('SELECT * FROM video_games'); while ($rowData = $response->fetch()) { echo $rowData['name']; } $response->closeCursor(); We switchover during the execution of the loop Although AC is only for Java applications, to help you understand what a cursor does, here is how to use a cursor in PHP :
1)Perform a SELECT query 2)Switchover while the cursor of the result-set is still active 3)Try to fetch the next row of results from the cursor 4)Outcome: Data Source with AC Data Source without AC No error is displayed An ORA error is displayed We can keep on fetching data from the cursor The result set is lost
Scenario 2: Restore an in-flight transaction Demo
BEGIN INSERT INTO CUSTOMERS (ID,NAME,AGE,ADDRESS,SALARY) VALUES (1, 'Ramesh', 32, 'Ahmedabad', 2000.00 ); INSERT INTO CUSTOMERS (ID,NAME,AGE,ADDRESS,SALARY) VALUES (2, 'Khilan', 25, 'Delhi', 1500.00 ); COMMIT Example of a local transaction
1)Begin a local transaction, and insert some rows 2)Switchover, without previously committing the transaction 3)Try to insert a new row 4)Outcome: Data Source with AC Data Source without AC No error is displayed An ORA error is displayed We can keep on inserting new values, and finally commit. The transaction is lost
Understand Application Continuity I. javax.sql.DataSource implementations 1) Understanding JDBC drivers 2) Connection pools and JDBC drivers 3) Weblogic Data Sources and JDBC drivers 4) How an app gets a connection a from Welogic Data Source II. Application Continuity 1) How AC works ● What happens at run-time ● What happens when the connection is interrupted 2) Demo ● AC while fetching data from the cursor of a result set ● AC while performing a local transation 3) Notes About AC ● Notes about Global Transactions
Notes about Global Transactions (JDBC driver)
Notes about Global Transactions for Application Continuity ● We can only use the non-XA driver ● However you can still enable the JDBC driver to participate in Global Transactions, by enabling last logging resource, or single-phase-commit, or simulate two-phases-commit.

More Related Content

PDF
Trivadis TechEvent 2016 Oracle Client Failover - Under the Hood by Robert Bialek
byTrivadis
 
PDF
Angular
byLilia Sfaxi
 
PDF
Helm 3
byMatthew Farina
 
PDF
2 ways to get total sum of interactive grid column oracle apex ontoor blogs
bysulimankareem
 
PPTX
An Overview of Web Services: SOAP and REST
byRam Awadh Prasad, PMP
 
PPTX
CQRS: Command/Query Responsibility Segregation
byBrian Ritchie
 
PPT
Angular 8
bySunil OS
 
PDF
REST API and CRUD
byPrem Sanil
 
Trivadis TechEvent 2016 Oracle Client Failover - Under the Hood by Robert Bialek
byTrivadis
 
Angular
byLilia Sfaxi
 
Helm 3
byMatthew Farina
 
2 ways to get total sum of interactive grid column oracle apex ontoor blogs
bysulimankareem
 
An Overview of Web Services: SOAP and REST
byRam Awadh Prasad, PMP
 
CQRS: Command/Query Responsibility Segregation
byBrian Ritchie
 
Angular 8
bySunil OS
 
REST API and CRUD
byPrem Sanil
 

What's hot

PPTX
Web services SOAP
byprinceirfancivil
 
PPT
Spring data presentation
byOleksii Usyk
 
PDF
Modern JavaScript Frameworks: Angular, React & Vue.js
byJonas Bandi
 
PPT
Easy enterprise application integration with RabbitMQ and AMQP
byRabbit MQ
 
PPTX
Introduction to microservices
byAnil Allewar
 
PPTX
Comprehensive Terraform Training
byYevgeniy Brikman
 
PPTX
Introducing Anypoint Exchange 2.0
byMuleSoft
 
PDF
API for Beginners
bySébastien Saunier
 
PPTX
Jetpack Compose.pptx
byGDSCVJTI
 
PDF
TypeScript Introduction
byDmitry Sheiko
 
PPTX
jstl ( jsp standard tag library )
byAdarsh Patel
 
PPT
jQuery Ajax
byAnand Kumar Rajana
 
PDF
Bizweb Microservices Architecture
byKhôi Nguyễn Minh
 
PPT
C#.NET
bygurchet
 
PDF
SOAP-based Web Services
byKatrien Verbert
 
PPTX
Hibernate jpa
byLhouceine OUHAMZA
 
PPT
.Net overview|Introduction Of .net
bypinky singh
 
PPTX
Introduction to Javascript
byambuj pathak
 
PPTX
Entity Framework - Entity Data Model (edm)
byEyal Vardi
 
PPTX
Introduction to Apex for Developers
bySalesforce Developers
 
Web services SOAP
byprinceirfancivil
 
Spring data presentation
byOleksii Usyk
 
Modern JavaScript Frameworks: Angular, React & Vue.js
byJonas Bandi
 
Easy enterprise application integration with RabbitMQ and AMQP
byRabbit MQ
 
Introduction to microservices
byAnil Allewar
 
Comprehensive Terraform Training
byYevgeniy Brikman
 
Introducing Anypoint Exchange 2.0
byMuleSoft
 
API for Beginners
bySébastien Saunier
 
Jetpack Compose.pptx
byGDSCVJTI
 
TypeScript Introduction
byDmitry Sheiko
 
jstl ( jsp standard tag library )
byAdarsh Patel
 
jQuery Ajax
byAnand Kumar Rajana
 
Bizweb Microservices Architecture
byKhôi Nguyễn Minh
 
C#.NET
bygurchet
 
SOAP-based Web Services
byKatrien Verbert
 
Hibernate jpa
byLhouceine OUHAMZA
 
.Net overview|Introduction Of .net
bypinky singh
 
Introduction to Javascript
byambuj pathak
 
Entity Framework - Entity Data Model (edm)
byEyal Vardi
 
Introduction to Apex for Developers
bySalesforce Developers
 

Similar to Application Continuity with Oracle DB 12c

PDF
JDBC in Servlets
byEleonora Ciceri
 
PDF
Database failover from client perspective
byPriit Piipuu
 
PPT
JDBC Connectivity Model
bykunj desai
 
PPTX
Application Continuity
bySean Braymen
 
PDF
SmartDB Office Hours: Connection Pool Sizing Concepts
byKoppelaars
 
PPTX
Java Database Connectivity by shreyash simu dbce.pptx
byash909077
 
PPTX
Project Presentation on Advance Java
byVikas Goyal
 
PPTX
Oracle WebLogic Server 12c: Seamless Oracle Database Integration (with NEC, O...
byjeckels
 
PDF
Achieving Continuous Availability for Your Applications with Oracle MAA
byMarkus Michalewicz
 
PPTX
21CS642 Module 5 JDBC PPT.pptx VI SEM CSE Students
byVENKATESHBHAT25
 
PDF
Jdbc
byharibee2000
 
PDF
JDBC Next: A New Asynchronous API for Connecting to a Database
byYolande Poirier
 
PDF
Bea weblogic job_interview_preparation_guide
byPankaj Singh
 
PDF
20151010 my sq-landjavav2a
byIvan Ma
 
PPTX
Jdbc
byBindhuBhargaviTalasi
 
PPT
SHOW104: Practical Java
byMark Myers
 
PDF
AJ-notes-final-converted.pdfhgfcgfjfjhgf
bykomalbhagat8803
 
PDF
Java Web Programming [3/9] : Servlet Advanced
byIMC Institute
 
PPT
Oracle High Availabiltity for application developers
byAlexander Tokarev
 
PDF
ADBA (Asynchronous Database Access)
byLogico
 
JDBC in Servlets
byEleonora Ciceri
 
Database failover from client perspective
byPriit Piipuu
 
JDBC Connectivity Model
bykunj desai
 
Application Continuity
bySean Braymen
 
SmartDB Office Hours: Connection Pool Sizing Concepts
byKoppelaars
 
Java Database Connectivity by shreyash simu dbce.pptx
byash909077
 
Project Presentation on Advance Java
byVikas Goyal
 
Oracle WebLogic Server 12c: Seamless Oracle Database Integration (with NEC, O...
byjeckels
 
Achieving Continuous Availability for Your Applications with Oracle MAA
byMarkus Michalewicz
 
21CS642 Module 5 JDBC PPT.pptx VI SEM CSE Students
byVENKATESHBHAT25
 
Jdbc
byharibee2000
 
JDBC Next: A New Asynchronous API for Connecting to a Database
byYolande Poirier
 
Bea weblogic job_interview_preparation_guide
byPankaj Singh
 
20151010 my sq-landjavav2a
byIvan Ma
 
Jdbc
byBindhuBhargaviTalasi
 
SHOW104: Practical Java
byMark Myers
 
AJ-notes-final-converted.pdfhgfcgfjfjhgf
bykomalbhagat8803
 
Java Web Programming [3/9] : Servlet Advanced
byIMC Institute
 
Oracle High Availabiltity for application developers
byAlexander Tokarev
 
ADBA (Asynchronous Database Access)
byLogico
 

More from Léopold Gault

PDF
OAuth OpenID Connect
byLéopold Gault
 
PDF
SAML
byLéopold Gault
 
PPTX
Notes leo kafka
byLéopold Gault
 
PPTX
Containers and Kubernetes -Notes Leo
byLéopold Gault
 
PPTX
NoSQL - Leo's notes
byLéopold Gault
 
PPTX
Leo's Notes about Apache Kafka
byLéopold Gault
 
PPTX
Leo's notes - Oracle DBA 2 Days
byLéopold Gault
 
OAuth OpenID Connect
byLéopold Gault
 
SAML
byLéopold Gault
 
Notes leo kafka
byLéopold Gault
 
Containers and Kubernetes -Notes Leo
byLéopold Gault
 
NoSQL - Leo's notes
byLéopold Gault
 
Leo's Notes about Apache Kafka
byLéopold Gault
 
Leo's notes - Oracle DBA 2 Days
byLéopold Gault
 

Recently uploaded

PPT
SAP ERP Training For Senior Managers.ppt
byBalanathanVirupasan
 
PDF
Autowiring all the things! - DrupalCon Vienna 2025 - Luca Lusso, SparkFabrik
bysparkfabrik
 
PPSX
Domain Centered Architecture for complex business domains
byRob Vens
 
PDF
ASTC Conference 2025 - Bridging the Gap!
byGareth Oakes
 
PDF
Austin Marketo User Group: Ground Control 2 MOPS: AI-Ignite your Map
bybbedford2
 
PDF
VADY Smart Data Flow – Automated Data Insights for Scalable Enterprise Decisions
byNewFangledVision
 
DOCX
Best Tableau Alternatives for Data Analytics and Reporting.docx
byVarsha Nayak
 
PDF
Making sense of AWS Serverless operations- AWS Community Day CEE 2025
byVadym Kazulkin
 
PDF
Ensuring Regulatory Excellence with Insurance Compliance Software
byInsurance Tech Services
 
PPTX
Integrated Billing, Accounting & Inventory Solution for MSMEs
byMoving Cloud
 
PDF
The Rise of AI Agents: Powering the Next Industrial Era
byMaxim Salnikov
 
PPTX
AI-Powered Intelligent Agents for Fraud Detection
byleolucus1995
 
DOCX
What Is the Best BI Software Other Than Tableau.docx
byVarsha Nayak
 
DOCX
Comprehensive Guide to SharePoint Consulting Services – Benefits, Process, an...
bySharepoint Designs
 
PPTX
Oracle_AWS_Azure_GCP_Comparison_ModernTech.pptx
byVITS184D1
 
PDF
Practical Performance Tuning for Serverless Java on AWS- InfoQ Dev Summit
byVadym Kazulkin
 
PDF
The 50+ First-Timer: It’s never too late for Open Source!
byImma Valls Bernaus
 
PDF
Scraping Amazon Prime Watchlist Data for Viewing Patterns.pdf
byyashpatric
 
PDF
Driving Stress Detection Using Multimodal CNN with Nonlinear Representation o...
byPranjal Kumar
 
PDF
Safe Confined Space Entry Monitoring_ Singapore Experts.pdf
byprasadexiga0
 
SAP ERP Training For Senior Managers.ppt
byBalanathanVirupasan
 
Autowiring all the things! - DrupalCon Vienna 2025 - Luca Lusso, SparkFabrik
bysparkfabrik
 
Domain Centered Architecture for complex business domains
byRob Vens
 
ASTC Conference 2025 - Bridging the Gap!
byGareth Oakes
 
Austin Marketo User Group: Ground Control 2 MOPS: AI-Ignite your Map
bybbedford2
 
VADY Smart Data Flow – Automated Data Insights for Scalable Enterprise Decisions
byNewFangledVision
 
Best Tableau Alternatives for Data Analytics and Reporting.docx
byVarsha Nayak
 
Making sense of AWS Serverless operations- AWS Community Day CEE 2025
byVadym Kazulkin
 
Ensuring Regulatory Excellence with Insurance Compliance Software
byInsurance Tech Services
 
Integrated Billing, Accounting & Inventory Solution for MSMEs
byMoving Cloud
 
The Rise of AI Agents: Powering the Next Industrial Era
byMaxim Salnikov
 
AI-Powered Intelligent Agents for Fraud Detection
byleolucus1995
 
What Is the Best BI Software Other Than Tableau.docx
byVarsha Nayak
 
Comprehensive Guide to SharePoint Consulting Services – Benefits, Process, an...
bySharepoint Designs
 
Oracle_AWS_Azure_GCP_Comparison_ModernTech.pptx
byVITS184D1
 
Practical Performance Tuning for Serverless Java on AWS- InfoQ Dev Summit
byVadym Kazulkin
 
The 50+ First-Timer: It’s never too late for Open Source!
byImma Valls Bernaus
 
Scraping Amazon Prime Watchlist Data for Viewing Patterns.pdf
byyashpatric
 
Driving Stress Detection Using Multimodal CNN with Nonlinear Representation o...
byPranjal Kumar
 
Safe Confined Space Entry Monitoring_ Singapore Experts.pdf
byprasadexiga0
 

Application Continuity with Oracle DB 12c

  • 1.
    Understanding Application Continuity ● Thoseslides are Leopold Gault's personal notes and drawings, about application continuity. ● You can see my sources, and comments in the slides' notes. ● You may ask me for the original file at galeopold@gmail.com ● Views expressed here are my own and do not necessarily reflect the views of Oracle.
  • 2.
    Understand Application Continuity I.javax.sql.DataSource implementations 1) Understanding JDBC drivers 2) Connection pools and JDBC drivers 3) Weblogic Data Sources and JDBC drivers 4) How an app gets a connection a from Welogic Data Source II. Application Continuity 1) How AC works ● What happens at run-time ● What happens when the connection is interrupted 2) Demo ● AC while fetching data from the cursor of a result set ● AC while performing a local transation 3) Notes About AC ● Notes about Global Transactions
  • 3.
  • 4.
    JVM Used by theapp javax.sql.DataSource interface
  • 5.
    JVM connection factory classa.k.a. JDBC driver Implements javax.sql.DataSource interface Used by the app « Connection facrory » or « JDBC driver » ? Features of the JDBC driver Connection Factory feature
  • 6.
    JVM JDBC connection JDBC driver Implementsjavax.sql.DataSource interface .getConnection()
  • 7.
    JVM Connection pool JDBC driver Implements DataSourceinterface Often includes Read the notes to access my sources
  • 8.
    Connection Pool Its interfaceExtends javax.sql.DataSource JVM Standalone JDBC app JDBC driver Implements DataSource interface Actually, I think that the JDBC driver can be used both ● in a standalone manner (the app is then called « standalone JDBC application ») ● or configured as the connection factory of a connection pool, such as UCP or Weblogic Server connection pool. source Directly uses Jdbc driver Implements javax.sql.DataSource JVM
  • 9.
    Middle-tier JVM The UCPoracle.ucp.jdbc.PoolDataSourceImpl is used to manipulate the JDBC driver (a.k.a. connection factory). The app doesn't directly asks the JDBC driver for a connection. It asks the connection pool. PoolDataSource pds = PoolDataSourceFactory.getPoolDataSource(); pds.setConnectionFactoryClassName("oracle.jdbc.pool.OracleDataSource"); System.out.println("connection factory set"); String URL = "jdbc:oracle:thin:@(DESCRIPTION = (TRANSPORT_CONNECT_TIMEOUT=3) (RETRY_COUNT=20)(FAILOVER=ON) " + (ADDRESS = (PROTOCOL = TCP)(HOST = rac12scan.gns.example.com)(PORT = 1521)) (CONNECT_DATA = " + (SERVER = DEDICATED) (SERVICE_NAME = ACTEST)))"; System.out.println("Using URLn" + URL); pds.setURL(URL); pds.setUser("martin"); pds.setPassword("***"); pds.setInitialPoolSize(10); pds.setMinPoolSize(10); pds.setMaxPoolSize(20); // RAC Features pds.setConnectionPoolName("Application Continuity Pool"); pds.setFastConnectionFailoverEnabled(true); // use srvctl config nodeapps to get the ONS ports on the cluster pds.setONSConfiguration("nodes=192.168.100.30:6200,192.168.100.32:6200"); System.out.println("pool configured, trying to get a connection"); Connection conn = pds.getConnection(); if (conn == null || !((ValidConnection) conn).isValid()) { System.out.println("connection is not valid"); throw new Exception ("invalid connection obtained from the pool"); } source JDBC driver class name Universal Connection Pool Universal Connection Pool Its interface Extends javax.sql.DataSource Jdbc driver Implements javax.sql.DataSource
  • 10.
    Weblogic Weblogic Data SourceConnection Pool Its interface extends DataSource interface[2] I assume that the concept is the same for Weblogic Data Source Connection Pool: it is a class that manipulates the JDBC driver in order to maintain a connection pool. Looked-up by the app thanks to its JNDI name Weblogic Data Source calling it « Weblogic Server Connection Pool » would make more sense! [1] Jdbc driver Implements javax.sql.DataSource
  • 11.
    Weblogic Weblogic Data Source Itsinterface extends DataSource interface Java app The app looks up the Data Source thanks to its JNDI name 1) import javax.sql.DataSource; public class dbUtilities { public dbUtilities() { super(); } public static final Long getDatabaseId (String seqName) { DataSource dataSource = null; Context ic = null; PreparedStatement stmt = null; ResultSet rs = null; Long dbId = null; String selectString; selectString = "SELECT " + seqName + ".nextval FROM DUAL"; try { ic = new InitialContext(); dataSource = (DataSource) ic.lookup("jdbc/AlphaOfficeAccessDS"); } catch (NamingException e) { } System.out.println("lookup dataSource returned " + dataSource); try { Connection connection = dataSource.getConnection(); ,,,,, Jdbc driver Implements javax.sql.DataSource
  • 12.
    Weblogic Java app Theapp asks the Data Source for a new connection 2) Weblogic Data Source Its interface extends javax.sql.DataSource interface Jdbc replay driver Implements javax.sql.DataSource
  • 13.
    Weblogic Java app 3) Weblogic DataSource Its interface extends javax.sql.DataSource interface Jdbc replay driver Implements javax.sql.DataSource A connection is given from the Connection pool. This is called a check-out
  • 14.
    Weblogic Java app The appcan now access the DB 4) Weblogic Data Source Its interface extends javax.sql.DataSource interface Jdbc replay driver Implements javax.sql.DataSource
  • 15.
    Weblogic Java app Everything (allthe database calls) that happens on the connection object between the check-out and the check-in is called a request [1] Weblogic Data Source Its interface extends javax.sql.DataSource interface Jdbc replay driver Implements javax.sql.DataSource Check-out Check-in
  • 16.
    Understand Application Continuity I.javax.sql.DataSource implementations 1) Understanding JDBC drivers 2) Connection pools and JDBC drivers 3) Weblogic Data Sources and JDBC drivers 4) How an app gets a connection a from Welogic Data Source II. Application Continuity 1) How AC works ● What happens at run-time ● What happens when the connection is interrupted 2) Demo ● AC while fetching data from the cursor of a result set ● AC while performing a local transation 3) Notes About AC ● Notes about Global Transactions
  • 17.
    2] Application Continuity Howto use the JDBC replay driver
  • 18.
    In order tobenefit from Application Continuity, you have to get a connection from the JDBC replay driver, whether it is usued by a connection pool (Weblogic JDBC Data Source, UCP, or 3rd party connection pool), or directly used by the application (JDBC standalone app) Connection Pool Its interface Extends javax.sql.DataSource JVM Standalone JDBC app JDBC driver Implements DataSource interface Directly uses Jdbc driver Implements javax.sql.DataSource JVM
  • 19.
    Weblogic In order tobenefit from Application Continuity, you have to get a connection from the JDBC replay driver, whether it is usued by a connection pool (Weblogic JDBC Data Source, UCP, or 3rd party connection pool), or directly used by the application (JDBC standalone app) Weblogic Data Source Its interface extends DataSource interface Jdbc driver Implements javax.sql.DataSource
  • 20.
    JVM PoolDataSource pds =PoolDataSourceFactory.getPoolDataSource(); pds.setConnectionFactoryClassName("oracle.jdbc.replay.OracleDataSourceImpl"); System.out.println("connection factory set"); String URL = "jdbc:oracle:thin:@(DESCRIPTION = (TRANSPORT_CONNECT_TIMEOUT=3) (RETRY_COUNT=20) (FAILOVER=ON) " + (ADDRESS = (PROTOCOL = TCP)(HOST = rac12scan.gns.example.com)(PORT = 1521)) (CONNECT_DATA = " + (SERVER = DEDICATED) (SERVICE_NAME = ACTEST)))"; System.out.println("Using URLn" + URL); pds.setURL(URL); pds.setUser("martin"); pds.setPassword("***"); pds.setInitialPoolSize(10); pds.setMinPoolSize(10); pds.setMaxPoolSize(20); // RAC Features pds.setConnectionPoolName("Application Continuity Pool"); pds.setFastConnectionFailoverEnabled(true); // use srvctl config nodeapps to get the ONS ports on the cluster pds.setONSConfiguration("nodes=192.168.100.30:6200,192.168.100.32:6200"); System.out.println("pool configured, trying to get a connection"); Connection conn = pds.getConnection(); if (conn == null || !((ValidConnection) conn).isValid()) { System.out.println("connection is not valid"); throw new Exception ("invalid connection obtained from the pool"); } source JDBC replay driver class name In order to benefit from Application Continuity, you have to get a connection from the JDBC replay driver, whether it is usued by a connection pool (Weblogic JDBC Data Source, UCP, or 3rd party connection pool), or directly used by the application (JDBC standalone app) Universal Connection Pool Jdbc driver Implements javax.sql.DataSource
  • 21.
    Note that ifyou use connections that do not come from an Oracle connection pool (Weblogic Data Source or UCP, etc.), but instead come from a third party connection pool, you have to add beginRequest and endRequest APIs to the application's own connection pool to identify request boundaries. [1] import java.sql.Connection; import javax.sql.PooledConnection; import oracle.jdbc.OracleConnection; import oracle.jdbc.replay.OracleDataSourceFactory; import oracle.jdbc.replay.OracleDataSource; import oracle.jdbc.replay.OracleConnectionPoolDataSource; ... { ...... OracleDataSource rds = OracleDataSourceFactory.getOracleDataSource(); rds.setUser(user); rds.setPassword(passwd); rds.setURL(url); ...... // Other data source configuration like callback, timeouts, etc. Connection conn = rds.getConnection(); ((OracleConnection) conn).beginRequest(); // Explicit request begin ...... // JDBC calls protected by Application Continuity ((OracleConnection) conn).endRequest(); // Explicit request end conn.close(); OracleConnectionPoolDataSource rcpds = OracleDataSourceFactory.getOracleConnectionPoolDataSource(); rcpds.setUser(user); rcpds.setPassword(passwd); rcpds.setURL(url); ...... // other data source configuration like callback, timeouts, and so on PooledConnection pc = rcpds.getPooledConnection(); Connection conn2 = pc.getConnection(); // Implicit request begin ...... // JDBC calls protected by Application Continuity conn2.close(); // Implicit request end source JDBC replay driver interface name Here we use the replay-driver directly so we have to add beginRequest() Here we use an Oracle connection pool, so no need to add beginRequest()
  • 22.
    JVM JDBC replay driver Implements DataSourceinterface Standalone jdbc app directly uses Note that if you use connections that do not come from an Oracle connection pool (Weblogic Data Source or UCP, etc.), but instead come from a third party connection pool, you have to add beginRequest and endRequest APIs to the application's own connection pool to identify request boundaries. [1] import java.sql.Connection; import javax.sql.PooledConnection; import oracle.jdbc.OracleConnection; import oracle.jdbc.replay.OracleDataSourceFactory; import oracle.jdbc.replay.OracleDataSource; import oracle.jdbc.replay.OracleConnectionPoolDataSource; ... { ...... OracleDataSource rds = OracleDataSourceFactory.getOracleDataSource(); rds.setUser(user); rds.setPassword(passwd); rds.setURL(url); ...... // Other data source configuration like callback, timeouts, etc. Connection conn = rds.getConnection(); ((OracleConnection) conn).beginRequest(); // Explicit request begin ...... // JDBC calls protected by Application Continuity ((OracleConnection) conn).endRequest(); // Explicit request end conn.close(); OracleConnectionPoolDataSource rcpds = OracleDataSourceFactory.getOracleConnectionPoolDataSource(); rcpds.setUser(user); rcpds.setPassword(passwd); rcpds.setURL(url); ...... // other data source configuration like callback, timeouts, and so on PooledConnection pc = rcpds.getPooledConnection(); Connection conn2 = pc.getConnection(); // Implicit request begin ...... // JDBC calls protected by Application Continuity conn2.close(); // Implicit request end source JDBC replay driver interface name Note that if you use the jdbc replay driver in a stand alone manner, you also have to add beginRequest/endRequest [2] Here we use the replay-driver directly so we have to add beginRequest() Here we use an Oracle connection pool, so no need to add beginRequest()
  • 23.
    What happens withthe replay driver and DB 12c at runtime [source] 2] Application Continuity
  • 24.
    Weblogic Weblogic Data Source Javaapp JDBC driver Implements DataSource interface Used to create connections Connection pool The replay-driver monitors and gets information from ● the connection object ● the continuity director (that sits inside the database). It stores any relevant information for a replay, until the end of the request (when the connection is sent back to the connection pool). Continuity director
  • 25.
    Weblogic Weblogic Data Source Javaapp JDBC driver Implements DataSource interface Used to create connections Connection pool Specifically, the replay- driver saves all the PL/SQL statements that the continuity director has instructed it to store. It saves them into the replay- context. For each call made to the DB, the replay driver retains the context required for a subsequent replay. [1] Replay context PL/SQL statements Continuity director
  • 26.
    Weblogic Weblogic Data Source Javaapp JDBC driver Implements DataSource interface Used to create connections Connection pool The continuity director also stores protocol and validation information. It stores this information locally, on what seems to be called the « replay context » too. However, it is probably quite different from the replay context of the JDBC- replay-driver.Protocol and validation information ? Replay context ? Continuity director
  • 27.
    Tx Guard Weblogic Weblogic DataSource Java app JDBC driver Implements DataSource interface Used to create connections Connection pool Moreover, I think that both the JDBC replay driver and the Continuity director store the logical transaction IDs (LTXID) of recent transactions. The LTXID is provided by the Transaction Guard feature of the 12c DB. It allows to identify logically a transaction that has been committed, and to know its outcome (committed and completed or rolledback). Transaction Guard therefore stores a history of past transactions, with their known outcome. Continuity director LTXID SQL statements XXXX INSERT .. DELETE .. COMMIT YYYY CREATE
  • 28.
    We interrupt theconnection by performing a switchover Application Continuity
  • 29.
    Data Guard Weblogic Weblogic DataSource Java app JDBC driver Implements DataSource interface Used to create connections Standby DB Connection pool Primary DB We switchover A switchover is seen from an external point of view as if all the DB sessions were killed
  • 30.
    Data Guard Weblogic Weblogic DataSource Java app JDBC driver Implements DataSource interface Used to create connections Standby DB Connection pool Primary DB The app can now access the DB The replay-driver intercepts execution errors. When the error is recoverable, the replay- driver replays the calls that were performed since the beginning of the request. It therefore restores : ● the cursors as they were (same position in the result set), ● and the latest transaction that was being build up, as it was.
  • 31.
    Data Guard Weblogic Weblogic DataSource Java app JDBC driver Implements DataSource interface Used to create connections Standby DB Connection pool Primary DB The app can now access the DB Thanks to the replay of the replay-driver, new DB sessions have been created on the new-primary DB. Those re-created DB- sessions don't necessarily have the same SID and serial number (cf next slide). But the application keeps on using the same connection object (it does not check-out a new one), so to me, this means that the replay driver makes the replacement of the old connection object by a new one invisible, by registering the new connection object at the same reference variable. I guess the replay driver does the same for the connections in the connection pool. By the way, I think that the mechanism is the same for all the objects got out of the connection object (such as result-set cursors) : they are replaced by the replay driver and registered at the same reference variable ; so that the application beleive it's the same object.
  • 32.
    Before we killall the sessions, here is the (SID, Serial number) couple of all the existing database sessions for our application (SwingBench) After we killed the sessions, new sessions are created to replace the killed ones. They don't necessarily have the same (SID, Serial number) couple Application continuity when killing all the db sessions on a single database (no RAC nor Data Guard)
  • 33.
    Weblogic Weblogic Data Source Javaapp JDBC driver Implements DataSource interface Used to create connections Connection pool Replay context PL/SQL statements Continuity director Protocol and validation information ? Replay context ? The request is replayed by the driver, using the replay context in order to have the same context as when the request was originally played. The continuity director decides whether or not it accepts a replay ; both before and after the replay (I think). If the replay is successful, the expected response will be given to the app. It will look as if there had been a small time delay.
  • 34.
    Tx Guard Weblogic Weblogic DataSource Java app JDBC driver Implements DataSource interface Used to create connections Connection pool Continuity director LTXID SQL statements XXXX INSERT .. DELETE .. COMMIT YYYY CREATE « Has transaction XXXX been successfully committed ? » [About transactions] In order to replay the most recent transaction that was asserted during the request, the replay-driver asks Transaction Guard if the transaction has been successfully committed, using its logical transaction ID. If it was not, the replay driver can replay the transaction. This process prevents from committing two times the same transaction, and thus result with data inconsistency.
  • 35.
    Understand Application Continuity I.javax.sql.DataSource implementations 1) Understanding JDBC drivers 2) Connection pools and JDBC drivers 3) Weblogic Data Sources and JDBC drivers 4) How an app gets a connection a from Welogic Data Source II. Application Continuity 1) How AC works ● What happens at run-time ● What happens when the connection is interrupted 2) Demo ● AC while fetching data from the cursor of a result set ● AC while performing a local transation 3) Notes About AC ● Notes about Global Transactions
  • 36.
  • 37.
    Scenario 1 : Restoringan active cursor Demo
  • 38.
    $response = $db->query('SELECT* FROM video_games'); while ($rowData = $response->fetch()) { echo $rowData['name']; } $response->closeCursor(); We switchover during the execution of the loop Although AC is only for Java applications, to help you understand what a cursor does, here is how to use a cursor in PHP :
  • 39.
    1)Perform a SELECTquery 2)Switchover while the cursor of the result-set is still active 3)Try to fetch the next row of results from the cursor 4)Outcome: Data Source with AC Data Source without AC No error is displayed An ORA error is displayed We can keep on fetching data from the cursor The result set is lost
  • 40.
    Scenario 2: Restorean in-flight transaction Demo
  • 41.
    BEGIN INSERT INTO CUSTOMERS(ID,NAME,AGE,ADDRESS,SALARY) VALUES (1, 'Ramesh', 32, 'Ahmedabad', 2000.00 ); INSERT INTO CUSTOMERS (ID,NAME,AGE,ADDRESS,SALARY) VALUES (2, 'Khilan', 25, 'Delhi', 1500.00 ); COMMIT Example of a local transaction
  • 42.
    1)Begin a localtransaction, and insert some rows 2)Switchover, without previously committing the transaction 3)Try to insert a new row 4)Outcome: Data Source with AC Data Source without AC No error is displayed An ORA error is displayed We can keep on inserting new values, and finally commit. The transaction is lost
  • 43.
    Understand Application Continuity I.javax.sql.DataSource implementations 1) Understanding JDBC drivers 2) Connection pools and JDBC drivers 3) Weblogic Data Sources and JDBC drivers 4) How an app gets a connection a from Welogic Data Source II. Application Continuity 1) How AC works ● What happens at run-time ● What happens when the connection is interrupted 2) Demo ● AC while fetching data from the cursor of a result set ● AC while performing a local transation 3) Notes About AC ● Notes about Global Transactions
  • 44.
    Notes about GlobalTransactions (JDBC driver)
  • 45.
    Notes about GlobalTransactions for Application Continuity ● We can only use the non-XA driver ● However you can still enable the JDBC driver to participate in Global Transactions, by enabling last logging resource, or single-phase-commit, or simulate two-phases-commit.

Editor's Notes

  • #5 https://docs.oracle.com/javase/7/docs/api/javax/sql/DataSource.html
  • #6 https://docs.oracle.com/database/121/JJUAR/oracle/ucp/jdbc/PoolDataSourceImpl.html The connection factory class is a class that implements javax.sql.DataSource and does not implement its own connection pooling. The connection factory's getConnection(...) methods should return a physical JDBC connection to the database. For example, "oracle.jdbc.pool.OracleDataSource" is a valid connection factory class name.
  • #8 Source : https://docs.oracle.com/javase/7/docs/api/javax/sql/DataSource.html « The DataSource interface is implemented by a driver vendor. There are three types of implementations: Basic implementation -- produces a standard Connection object Connection pooling implementation -- produces a Connection object that will automatically participate in connection pooling. This implementation works with a middle-tier connection pooling manager. Distributed transaction implementation -- produces a Connection object that may be used for distributed transactions and almost always participates in connection pooling. This implementation works with a middle-tier transaction manager and almost always with a connection pooling manager.  »  https://docs.oracle.com/javase/tutorial/jdbc/basics/sqldatasources.html « A JDBC driver should include at least a basic Data Source implementation. For example, the Java DB JDBC driver (-Leo : this is the one provided by the JDK, not Weblogic) includes the implementation org.apache.derby.jdbc.ClientData Source and for MySQL, com.mysql.jdbc.jdbc2.optional.MysqlData Source. »
  • #11 [2] that's why we cast the Weblogic Data Source as a javax.sql.DataSource when we look it up from the JNDI import javax.sql.DataSource; ... dataSource = (DataSource) ic.lookup("jdbc/AlphaOfficeAccessDS");
  • #16 1- « In general, a request is demarcated by the calls made between check-out and check-in of a database connection from a connection pool. » https://docs.oracle.com/database/121/CNCPT/transact.htm#CNCPT89323
  • #19 https://docs.oracle.com/database/121/ADFNS/adfns_app_continuity.htm#ADFNS1207
  • #21 https://docs.oracle.com/database/121/ADFNS/adfns_app_continuity.htm#ADFNS1207
  • #22 [1] https://docs.oracle.com/database/121/ADFNS/adfns_app_continuity.htm#ADFNS1207 Determine whether the application borrows and returns connections from the WebLogic Server Pool or Universal Connection Pool for each request, or whether to add beginRequest and endRequest APIs to the application's own connection pool to identify request boundaries. [2]https://docs.oracle.com/database/121/JJDBC/appcontnew.htm#JJDBC29073 You must use either the oracle.jdbc.replay.OracleDataSourceImpl or oracle.jdbc.replay.OracleConnectionPoolDataSourceImpl data source to obtain JDBC connections. They are new Oracle JDBC data sources, and work similarly to the existing non-XA data sources, such as oracle.jdbc.pool.OracleDataSource. You can use both in a standalone manner, or configure them as connection factories for a connection pool, such as Universal Connection Pool (UCP), or Oracle WebLogic Server connection pool.
  • #23 [1] https://docs.oracle.com/database/121/ADFNS/adfns_app_continuity.htm#ADFNS1207 Determine whether the application borrows and returns connections from the WebLogic Server Pool or Universal Connection Pool for each request, or whether to add beginRequest and endRequest APIs to the application's own connection pool to identify request boundaries. [2]https://docs.oracle.com/database/121/JJDBC/appcontnew.htm#JJDBC29073 You must use either the oracle.jdbc.replay.OracleDataSourceImpl or oracle.jdbc.replay.OracleConnectionPoolDataSourceImpl data source to obtain JDBC connections. They are new Oracle JDBC data sources, and work similarly to the existing non-XA data sources, such as oracle.jdbc.pool.OracleDataSource. You can use both in a standalone manner, or configure them as connection factories for a connection pool, such as Universal Connection Pool (UCP), or Oracle WebLogic Server connection pool.
  • #26 [1] « In collaboration with the database, the JDBC replay driver maintains a history of calls during a conversation between a client and the database. For each call made at run time, the driver retains the context required for a subsequent replay. » https://docs.oracle.com/database/121/CNCPT/transact.htm#CNCPT-GUID-2E973242-2B39-42FC-AA16-CD97B460B6D2
  • #32 [1] On the demo app O*Trade : With the replay-driver With the non-replay driver After killing the session, the same session get re-created : with the same ID (cf image below) However in Swingbench, the couple SID and serial # is different after the replay After killing the session, when you close the cursor and clidk Query again, the session ID is different..
  • #34 [1] « In collaboration with the database, the JDBC replay driver maintains a history of calls during a conversation between a client and the database. For each call made at run time, the driver retains the context required for a subsequent replay. » https://docs.oracle.com/database/121/CNCPT/transact.htm#CNCPT-GUID-2E973242-2B39-42FC-AA16-CD97B460B6D2