Go to main content

Developer's Guide for Migrating to Oracle Solaris 11

Exit Print View

Updated: April 2020
 
 

High Availability in Oracle Solaris

High availability (HA) ensures that a system, application, or a database is available, continuously accessible, and running for the desired period of time without any loss of service.

Oracle Solaris offers application and database availability. High availability for applications is provided by Oracle Solaris Cluster, which provides load balancing, fault detection, and disaster recovery to keep the application or the system highly available. Oracle Solaris Cluster provides an extensive high-availability framework. Oracle Solaris provides a full range of single and multisystem high-availability in traditional and virtualized environments.

Oracle Solaris Cluster software's high-availability framework detects a node failure and migrates the resources to another node in the network that can take up the action.

Oracle Solaris Cluster can also track, maintain, and monitor zones. It also offers a failover zone feature to protect a zone when it fails. The failover zone is monitored by HA agents for zones. HA Oracle Agent software controls activities in Oracle Database activities on Oracle Solaris Cluster nodes. The agent performs fault checking by starting, stopping, and probing the nodes. If an HA agent is configured on a database that has no active sessions, HA Oracle Agent will open a test transaction. The return error codes from HA Oracle Agent will be validated against a special action file on location.

Oracle Application Server provides local and enterprise high availability using the technologies like process death detection and automatic restart, clustering, server load balancing, rolling and patching, and backup and recovery. The Oracle Clusterware software provides infrastructure to manage any application.

Oracle Solaris Cluster includes an Agent Builder tool for application high-availability that automates the creation of data services. Developers supply Agent Builder with information about the application and data service to be created, such as whether a scalable or failover agent is desired, whether the service is network-aware, and the commands to use to start and stop the application. Agent Builder generates the data service, including source and executable code (C or Korn shell), a customized Resource Type Registration (RTR) file, and an Oracle Solaris package for distribution.

For more information about application server high availability, see https://docs.oracle.com/cd/E12839_01/core.1111/e10106.pdfhttps://docs.oracle.com/cd/E12839_01/core.1111/e10106.pdf.

Oracle Database provides high availability by minimizing both planned and unplanned downtime. Oracle Database achieves high-availability using Oracle Real Application Clusters (RAC) and Oracle Clusterware software. Together, they bind the servers to operate as one server. Some of the benefits of Oracle RAC and Oracle Clusterware software are:

  • Ability to tolerate and quickly recover from computer and instance failures

  • Ability to apply Oracle Clusterware upgrades, patch sets, and interim patches in a rolling fashion

  • Fast, automatic detection of connection failures and removal of terminated connections for any Java application using Oracle Universal Connection Pool (UCP), Fast Connection Failover, and FAN events

  • Comprehensive manageability integrating database and cluster features

  • Automatically restarts failed Oracle processes

  • Automatically manages and fails over Oracle Virtual IP (VIP) on another node in the cluster on node failures

For more information about Oracle Database high availability, see https://www.oracle.com/database/high-availability/index.htmlDatabase High Availability.

For more information about Oracle Solaris Cluster, see Oracle Solaris Cluster Product Documentation.