Application Continuity for the Oracle Database

Application Continuity For The Oracle Database-PDF Download

  • Date:31 Jul 2020
  • Views:1
  • Downloads:0
  • Pages:35
  • Size:734.16 KB

Share Pdf : Application Continuity For The Oracle Database

Download and Preview : Application Continuity For The Oracle Database

Report CopyRight/DMCA Form For : Application Continuity For The Oracle Database


TABLE OF CONTENTS,Introduction 3, Feature Set for Keeping Your Applications Continuously Available 4. Instructions for Application Configuration 5, Building Blocks for Continuous Availability of Your Application 6. Fast Application Notification FAN 8, Maintenance Without Impacting Your Applications 12. Transparent Application Continuity 15,Application Continuity 17. Transparent Application Failover 18, Knowing Your Protection Level when using TAC or AC 20.
Configuring your Clients 20,Conclusion 23,Appendix Configuring your Services 24. Appendix Using acchk to Check for Concrete Classes 26. Appendix Using acchk for Application Continuity Coverage 28. Appendix SQL to Report Protection By PDB Service and History 31. Additional Whitepapers 34,2 W HITE PAPER Continuous Availability. INTRODUCTION, Applications achieve continuous service when planned maintenance unplanned outages and load. imbalances of the database tier are hidden A combination of application best practice simple. configuration changes and an Oracle Database deployed using MAA best practices ensures that. your applications are continuously available, The following checklist is useful for preparing your applications and databases even if you are not. yet using Application Continuity The points discussed here provide significant value for preparing. your systems to support continuous availability reducing possible downtime during planned. maintenance activities and during unplanned outages should they occur. 3 W HITE PAPER Continuous Availability, FEATURE SET FOR KEEP ING YOUR APPLICATION S CONTINUOUSLY AVAILABLE.
Applications achieve continuous availability when planned maintenance unplanned outages and load imbalances of the database tier. are hidden Oracle provides a set of features that you can choose from to keep your application available during planned events. unplanned outages and load imbalances You can think of these features as an insurance policy protecting your applications from. service interruptions The best features are those that are fully transparent to your application so your application developers can focus. on building functionality not infrastructure and that continue to protect the application when it changes in the future W e call this. future proofing,Start with the feature set, Draining and Rebalancing Sessions for Planned Maintenance. When planned maintenance starts sessions that need to be drained from an instance PDB or database are marked to be drained Idle. sessions are released gradually Active sessions are drained when the work executing in that session completes Draining of sessions. is in wide use with Oracle connection pools and mid tiers configured for Fast Application Notification FAN Starting with Oracle. Database 18c the database itself drains sessions when PDBs and instances are stopped or relocated Draining is always the best. solution for hiding planned maintenance Failover solutions such as Application Continuity are the fallback when work will not drain in. the time allocated,Transparent Application Failover TAF. TAF is a feature dating back to Oracle8i Following an instance failure TAF creates a new session and when using SELECT mode on. demand replays queries back to where they were before the failure occurred Starting with Oracle Database 12 2 TAF offers. FAILOVER RESTORE matching Application Continuity to restore the initial session state before queries are replayed Cursors are. replayed using the state re established initially Applications using TAF must not change session state later in the session for example. PLSQL temp tables temp lobs sys context as this session state is not restored. Application Continuity AC, Application Continuity hides outages starting with Oracle database 12 1 for thin Java based applications and Oracle Database 12 2 0 1. for OCI and ODP NET based applications Application Continuity rebuilds the session by recovering the session from a known point. which includes session states and transactional states Application Continuity rebuilds all in flight work The application continues as it. was seeing a slightly delayed execution time when a failover occurs The standard mode for Application Continuity is for OLTP style. pooled applications,Transparent Application Continuity TAC. Starting with Oracle Database18c Transparent Application Continuity TAC transparently tracks and records session and transactional. state so the database session can be recovered following recoverable outages This is done with no reliance on application knowledge. or application code changes allowing Transparent Application Continuity to be enabled for your applications Application transparency. and failover are achieved by consuming the state tracking information that captures and categorizes the session state usage as the. application issues user calls,4 W HITE PAPER Continuous Availability.
Which Solution Should I use,TAC AC TAF Draining,I don t know how the Yes No No Yes. application is,implemented,My application does Yes Yes No for unplanned Yes. transactions,Transactional,disconnect only,My application uses Yes Yes No Yes. Oracle state temp lobs,No for static mode,PL SQL temp tables. My application does not Yes No Yes Yes,use connection pools.
My application has side Yes Customizable No Yes,effects such as file. Side effects are not, My app needs Initial Yes and custom Yes and custom Yes and custom Yes. State Restored,Future proofed for Yes No No Yes,application changes. Transparent Application Continuity with Draining is the recommended solution for Continuous Availability If you are using a 12c driver or. customization is required for initial states or side effects then you should use Application Continuity TAF continues to be available and is fully. INSTRUCTIONS FOR APPLICATION CONFIGURATION, Follow these instructions when implementing your solution. 1 Use an Oracle Clusterware managed service that is not the default database service the default service has the same name as the. database or PDB The services that you create provide location transparency and HA features. 2 Use the recommended Connection String explained later in this paper with built in timeouts retries and delay so incoming connections do. not see errors during outages, 3 Fast Application Notification FAN is a mandatory component to initiate draining to break out of failures and to rebalance sessions when.
services resume and when load imbalances occur For outages such as node and network failures fast failover of the application does not. happen if the client is not interrupted by FAN FAN applies to all failover solutions When configuring FAN use Auto configuration of ONS. Use the recommended TNS format strictly Do not alter this format Exception if your client is pre 12c you will manually configure FAN. 5 W HITE PAPER Continuous Availability, 4 Before maintenance starts drain your work from the instances or nodes targeted for maintenance Enable FAN with Oracle Connection. Pools or Connection tests or both Oracle connection pools with FAN are the best solution as pools provides full life cycle of session. movement That is draining and rebalancing of work as maintenance progresses When using FAN return your connections to the pool If. you are using server draining the alternate plan explained later in this paper and your test is not a standard test add your test to the server. using DBMS APP CONT ADMIN Sessions that do not drain within the DRAIN TIMEOUT will be failed over. 5 The standard solution for failing over sessions is Transparent Application Continuity TAC. Use Application Continuity AC if you are using Oracle Database 12c Release 2 or you want to customize with side effects or. callbacks or have an application that uses state such as temporary tables and never cleans up. Use Transparent Application Failover TAF if your application is read only and does not change Oracle session state in the session. after the initial setup, BUILDING BLOCKS FOR CONTINUOUS AVAILABILITY OF YOUR APPLICATION. Use Services, Service is a logical abstraction for managing work Services allow applications to benefit from the reliability of the redundant parts of the MAA. system The services hide the complexity of the underlying system from the client by providing a single system image for managing work. The service is, a unit for management a handful of services are manageable many nodes instances listeners and network interfaces are not. manageable The service provides location transparency for sites and databases. a unit for availability resources are recovered quickly independently and in parallel for each service and without the need to start entire. software stacks and, a unit for performance work is routed transparently across the MAA system according to service quality and priority Services are.
measured against service level thresholds and violations are reported to management with advised solutions in AWR. FAN connection identifier TAC AC switchover consumer groups and many other features and operations are predicated on the use of. services Do not use the default database service as this cannot be disabled relocated or restricted and so has no high availability support The. services you use are associated with a specific primary or standby role in a Data Guard environment Do not use the initialization parameter. service names for application usage,An Example of Services. From ATP D each PDB database is supplied with 5 pre configured services to choose from All provide FAN and draining. Service Name Description Draining FAN TAC,TPURGENT OLTP Highest Priority Yes Yes Yes. TP OLTP General Priority Yes Yes Yes,Recommended to be. used as main service,HIGH Reporting or Batch Yes Yes. Highest Priority,MEDIUM Reporting or Batch Yes Yes.
6 W HITE PAPER Continuous Availability,Medium Priority. LOW Reporting or Batch Yes Yes,Lowest Priority, Configure the Connection String or URL for High Availability. All Oracle supplied connect strings will conform to the following recommendations There is no need to do anything if you use the. Oracle supplied wallet The following TNS URL configuration is recommended for use for connecting at failover switchover fallback. and basic startup, Set RETRY COUNT RETRY DELAY CONNECT TIMEOUT and TRANSPORT CONNECT TIMEOUT parameters in the tnsnames ora file. or the URL to allow connection requests to wait for the service and connect successfully Connection attempts and retries are managed. by Oracle Database Net Services, Set CONNECT TIMEOUT to a high value to prevent login storms such as 90s or 120s Low values can result in frenzied login attempts. due to the application or pool cancelling and retrying connection attempts. Do not set RETRY COUNT 1 RETRY DELAY or CONNECT TIMEOUT larger than your response time SLA The application should. either connect or receive an error within the response time SLA. These are general recommendations for configuring the connections for high availability Do not use Easy Connect Naming on the. client as EZCONNECT has no high availability capabilities. Note that the standby scan specified below refers to the SCAN address available on the STANDBY site specified in your Active Data. Guard configuration Attempt will be made to connect to the PRIMARY site first and if the service is not available attempt to connect to. this service at the standby Once the location of the service is known Oracle drivers 12 2 and later remember the address list with that. service offered and chooses this first until the service next moves. Adding the standby scan to TNS connection descriptor to transparently fail over to the standby scan is optional Failing over to a. standby database within the same region will have acceptable performance in most cases versus failing over to a standby database in. a different region where additional network latency may result in unacceptable response time performance In the latter case a site. failover operation will be required which involves DNS failover to another region containing mid tier resources and standby database. Use this Connection String for ALL Oracle driver version 12 2 or higher. Alias or URL,DESCRIPTION, CONNECT TIMEOUT 120 RETRY COUNT 20 RETRY DELAY 3 TRANSPORT CONNECT TIMEOUT 3.
ADDRESS LIST,LOAD BALANCE on,ADDRESS PROTOCOL TCP HOST primary scan PORT 1521. ADDRESS LIST,LOAD BALANCE on,ADDRESS PROTOCOL TCP HOST standby scan PORT 1521. CONNECT DATA SERVICE NAME YOUR SERVICE, For JDBC connections in 12 1 or earlier the following should be used This Connection String uses a lower CONNECT TIMEOUT. because TRANSPORT CONNECT TIMEOUT is not available for thin Java drivers until Oracle Database 12 2 RETRY DELAY requires a. patch for 11 2 0 4 clients,7 W HITE PAPER Continuous Availability. Alias or URL,DESCRIPTION,CONNECT TIMEOUT 15 RETRY COUNT 20 RETRY DELAY 3.
ADDRESS LIST,LOAD BALANCE on,ADDRESS PROTOCOL TCP HOST primary scan PORT 1521. ADDRESS LIST,LOAD BALANCE on,ADDRESS PROTOCOL TCP HOST standby scan PORT 1521. CONNECT DATA SERVICE NAME YOUR SERVICE,FAST APPLICATION NOTIFICATION FAN. FAN must be used FAN is a required component for interrupting the application to failover Without FAN applications can hang on. TCP IP timeout following hardware and network failures and omit to rebalance when resources resume All Oracle pools and all Oracle. application servers use FAN Third party JAVA application servers can use UCP to enable FAN No application changes are required to. are hidden Oracle provides a set of features that you can choose from to keep your application available during planned events unplanned outages and load imbalances You can think of these features as an insurance policy protecting your applications from service interruptions The best features are those that are fully transparent to your application so your application developers can focus

Related Books