Independent Java Patching with AHF 24.7

July 30, 2024 | 4 minute read
Gareth Chapman
Senior Principal Development Operations Manager
Text Size 100%:

Oracle Autonomous Health Framework (AHF) is a diagnostic and system performance framework designed to keep database systems healthy and running while minimizing human reaction time.

It's the same technology Oracle uses in the cloud to keep the Autonomous Database running, and it's available for you at no extra cost as part of your existing software support agreement.

Autonomous Health Framework 24.7 is released and available to download

 

New Features in 24.7

 

Upgrade Java Without Upgrading AHF

Java used by AHF is now upgradable without having to upgrade AHF.

Before, the only way to upgrade the Java version used by AHF was to also upgrade AHF. This added extra work to patching new Java CVEs.

Now, the Java used by AHF is upgradable to the latest version without having to upgrade AHF.

On the discovery of new CVEs AHF will release patches containing the latest Java version. Customers can then apply the patch using the existing AHF patching command:

Copied to Clipboard
Error: Could not Copy
Copied to Clipboard
Error: Could not Copy
ahfctl applyupdate -updatefile {patch_zip}

Applying a Java patch does not update the AHF version.

See the user guide for more information on applying patches.

 

 

Improved Platinum Monitoring and Patching

AHF now enables Platinum to query data from dom0, storage servers and switches.

Platinum provides fault monitoring and patching services for Exadata customers. The service relies on AHF to provide Exadata configuration data.

Previous AHF releases were unable to provide Platinum with full details from dom0, storage servers or switches.

AHF now provides the following capabilities on Exadata dom0, storage servers and switches:

  • Auto-upgrade
  • Automatic best practice checking
  • Automatic diagnostic collections
  • Auto-upload diagnostic collections to SRs

This enables the Platinum fault detection and patching service to use component relationships. When it detects a fault from dom0, it can determine which database nodes it impacts. Patch planning for virtualized racks understands component relationships, allowing it to cut downtime.

 

Store Exadata Infrastructure Details for Best Practice Checking

The ahf command can now store the details of Exadata Dom0s, storage servers and switches. This is then later used in best practice checking.

AHF may not discover all Exadata infrastructure when run on Dom0. This means that Best Practice checks miss peer Dom0s, storage servers and switches.

The ahf command now provides the ability to save their details, using the command:

Copied to Clipboard
Error: Could not Copy
Copied to Clipboard
Error: Could not Copy
ahf configuration set --type cell --node {nodename} --password

Best Practice checks will then use this saved configuration for full infrastructure analysis.

AHF discovers peer DomUs from the Oracle Cluster Registry. Merging of Exachk reports from DomU and Dom0 provides s full report for the whole Exadata rack.

Use the following command to merge Exachk reports:

Copied to Clipboard
Error: Could not Copy
Copied to Clipboard
Error: Could not Copy
exachk -merge report_1,report_2

For more information on the ahf unified command line interface see the user guide.

 

New Problem Summaries

AHF can now detect and provide the resolution for more problems.

Since version 24.4 AHF has had the ability to detect problems and show a summary with the resolution. The Problem Summary page is available under the Detected Problems panel in Insights.

The Problem Summary contains:

  • Problem – describing what happened
  • Reason – describing why it happened
  • Cause – describing the root cause.
  • Evidence – proving why this is the cause.
  • Resolution Steps – detailing in simple terms exactly how to resolve the problem.

This release includes the ability to detect the following new problem causes:

Node Eviction due to:

  • Archiver blocked due to insufficient space in the recovery area
  • IO error due to insufficient space in ASM diskgroup
  • Private network performance degradation due to misconfigured MTU size

ASM Instance Eviction due to:

  • Being stuck waiting for a failed network interface card

Read more information see the user guide about Exploring Diagnostic Insights.

 

Recent AHF Features

AHF releases a new version every month, here are a recent new features by release:

AHF 24.6

AHF 24.5

AHF 24.4

AHF 24.3

AHF 24.2

 

Get started with Autonomous Health Framework today

To get started today download Oracle Autonomous Health Framework.

Gareth Chapman

Senior Principal Development Operations Manager

Gareth is Senior Principal Development Operations Manager for Oracle Autonomous Health Framework.


Previous Post

Enabling the MongoDB API for ORDS containers just got a lot easier

Martin Bach | 3 min read

Next Post


Introducing custom alert policies in Oracle Data Safe

Angeline Dhanarani | 10 min read