Autonomous Health Framework (AHF)

Oracle Autonomous Health Framework (AHF) presents the next generation of tools as components, which work together autonomously, 24/7, to keep database systems healthy and functional while minimizing human reaction time. Utilizing machine learning technologies, Oracle AHF provides early warning or automatically solves operational runtime issues faced by Database and System administrators in the areas of availability and performance.

• Proactive problem identification: AHF identifies potential issues before they become critical, allowing for preventative measures.
• Improved performance: AHF helps maintain optimal database performance by identifying bottlenecks and recommending solutions.
• Reduced downtime: AHF proactively addresses issues to minimize database downtime.
• Simplified troubleshooting: AHF automates tasks and provide insights to simplify troubleshooting complex database issues.

Oracle Autonomous Health Framework is a value add, available for no extra cost as part of existing Oracle product software support agreements.

The latest version of Oracle AHF can be downloaded here.

Information on Oracle Autonomous Health Framework is available here.

Oracle Trace File Analyzer (TFA) is a component of Autonomous Health Framework that analyzes, trims, and collects diagnostics from your database system. These diagnostics contain information that can be helpful for troubleshooting issues and errors. The diagnostics captured can help you identify the root cause of problems and integrate with My Oracle Support tooling and services if you need further assistance.

ORAchk and EXAchk are components of the Autonomous Health Framework that provide database health checks by comparing your configuration to best practices. EXAchk is used on Oracle Exadata systems and ORAchk on non-Exadata systems.

Insights is the user interface for viewing Oracle Autonomous Health Framework best practice results and diagnostics. It provides a bird's eye view of your entire system, allowing you to spot problems, drill into the root causes, and understand how to resolve them.

No. Oracle Autonomous Health Framework does not require a Real Application Cluster. It’s useful for all Oracle Database systems.

No. Oracle Autonomous Health Framework is useful for all supported versions of Oracle Database systems. 

Oracle AHF can resolve runtime issues that threaten the availability of the software stack, either through a software issue (DB, GI, OS) that causes the database or one of its instances to become unresponsive, or issues with the underlying hardware resources (CPU, Memory, Network, Storage) that cause a server to be evicted from the cluster and shut down all the database instances running there.

Oracle AHF can resolve the runtime issues that threaten the performance of a system as determined by the database client or application, either due to software issues (bugs, configuration, contention, etc.) or client issues (demand, query types, connection management, etc.).

Oracle AHF components work 24/7 in daemon mode to address availability and performance issues and ensure high availability for the database system. The components work together to provide a framework that continuously monitors the database system, maintains best practice configurations, and issues alerts about vulnerability to known concerns. The framework monitors database system performance autonomously and manages resources to maintain service level agreements. It also preserves resource availability in common cases of memory stress and of resources being blocked by hung sessions. AHF even uses machine learning to discover potential cluster and database problems, sending notifications with corrective actions to prevent the issues altogether. And finally, for the issues that are hard to detect and require Oracle Support, Oracle AHF speeds the issue diagnosis, triage, and resolution process.

Oracle Autonomous Health Framework autonomously detects and resolves hangs. A hang occurs when a chain of one or more sessions is blocked by another session and is not able to make any progress. Usually, these chains of sessions have a root or a final blocker session, called the victim session, that blocks all the other sessions in the chain. Once a victim session is selected, AHF applies hang resolution heuristics on the session. If the chain of sessions progresses or the hang resolves itself automatically, then AHF doesn’t take any action. However, if the hang doesn’t resolve itself, then the framework may resolve it by terminating the victim session or, if that’s unsuccessful, terminating the process.

While Oracle AHF detects and resolves hangs autonomously, it continuously logs all detections and resolutions in DB Alert Logs. So, if a session is terminated, it will be noted in these alert logs. The details of complete hang resolution are also available in dump trace files. 

Oracle AHF, containing ORAchk, EXAchk, and TFA, is probably already installed on your system, as it’s included with Oracle Database and Grid Infrastructure installations and Release Updates.

Separate installation instructions can be found in the user guide here.

When installed with Oracle Database or Grid Infrastructure, Oracle AHF will be located within the respective product’s installation directory. When installed separately, the default installation directory on Linux is /opt/oracle.ahf, but you can also customize the installation location.

Once installed, Oracle AHF will automatically and proactively run weekly best practice health checks, collect diagnostics, and provide notifications when problems are detected. For more information, see the Oracle Autonomous Health Framework user guide.

The command ahfctl statusahf can be used to check the status of Oracle AHF. More information can be found in the ahfctl statusahf section of the user guide.

ORAchk or EXAchk can be run either automatically or on demand with the orachk or exachk command. More information can be found in the user guide under Compliance Checking with Oracle ORAchk and Oracle EXAchk.

TFA collects diagnostics and sends notifications automatically. It can also be run on demand via the tfactl command. More information can be found in the user guide under Collect Diagnostic Data.

The TFA daemon can be stopped by running tfactl stop. More information can be found in the user guide under Managing and Configuring Oracle Trace File Analyzer.

Oracle AHF can be uninstalled by running ahfctl uninstall. More information can be found in the user guide under ahfctl uninstall.

By default, the TFA daemon will run automatically, monitoring your database for problems on an ongoing basis. If it detects any issues, it will collect relevant diagnostics and send notifications.

You can check the status of Trace File Analyzer by running the ahfctl statusahf command.

Oracle AHF can be used to upload TFA diagnostic collections to My Oracle Support to attach to your Service Request. To upload a TFA collection, run the command ahfctl upload -name mos -id -file<file>. More information can be found in the user guide under Running AHFCTL Upload Commands to Upload Diagnostics.

The TFA version can be found by running the command tfactl version.

The easiest way to maintain AHF is to use AutoUpgrade.

• Configure AHF AutoUpgrade: ahfctl setupgrade -swstage /scratch/ahf_stage -autoupg rade on -frequency 7
• When a new version of AHF is available, simply download and stage in the swstage directory
• Based on the checking frequency you’ve specified, AHF will detect the new version and automatically upgrade

More information can be found in the user guide under ahfctl upgrade commands.

Yes. Oracle AHF compliance checks are integrated into Enterprise Manager, and alerts are sent via Enterprise Manager. More information can be found in the Oracle Autonomous Health Framework user guide.