: If the disk group lacks redundancy or loses a quorum of failure groups, Oracle triggers an involuntary dismount .
When the checker finds a failure, it logs the alert. Many users report seeing "ASM Health Checker found 1 new failures" appear in the alert log file ( log.xml ) or as a metric alert in Oracle Enterprise Manager (OEM).
Here is a deep dive into what this error means, how to diagnose it, and the steps to resolve it. 1. Understanding the ASM Health Checker
If a path to a physical disk is lost (due to HBA failure or cable issues), ASM might mark the disk as "OFFLINE." If the diskgroup is still mounted but missing a member, the Health Checker will flag it.
Your first diagnostic step is to view the Grid Infrastructure alert log. This log is typically located under the Automatic Diagnostic Repository (ADR) home. asm health checker found 1 new failures
# Navigate to the Alert log directory cd $GRID_HOME/log/ /alert .log tail -n 200 alert .log Use code with caution.
an SRE or security engineer I want to receive an alert when the ASM health checker finds 1 or more new failures So that I can investigate regressions before they impact production.
Symptoms: Errors mentioning "mysql" or "table marked as crashed."
Apply the latest Release Updates (RUs). Oracle regularly refines the AMHC framework algorithms to eliminate false-positive alerts and optimize automated disk rebalancing. : If the disk group lacks redundancy or
If this guide helped you resolve the issue, please share your experience or let others know which specific fix worked for you in the comments below.
The checker alerts administrators if read/write operations exceed predefined latency thresholds (typically over 20 milliseconds for sustained periods).
flowchart TD A[ASM Alert Log Message<br>"ASM Health Checker found 1 new failures"] --> B(Step 1:<br>Run Diagnostic CHECK) B --> CCheck<br>V$ASM_OPERATION C --> D[Step 2:<br>Generate Health Check Report] D --> E[Step 3:<br>Analyze Failure] E --> FCorruption Found? F -- Yes --> G(Step 4:<br>Execute CHECK ALL REPAIR) G --> H[Step 5:<br>Monitor Rebalance] H --> I[Step 6:<br>Re-validate & Verify] I --> J[Step 7:<br>Document & Prevent] F -- No --> K[False Positive?<br>Engage ASM Team]
The DBMS_HM.GET_RUN_REPORT function returns a detailed report, including the specific Finding Name (e.g., "System datafile is old"), its Type (e.g., FAILURE), Priority (e.g., CRITICAL or HIGH), and a descriptive message. This report is invaluable for understanding the exact nature of the problem. Here is a deep dive into what this
The alert itself is a summary. To fix the issue, you must find the specific failure code. Log into the BIG-IP Configuration Utility (GUI). Navigate to Security > Overview > Health Reports.
Here’s a visual overview of the complete process:
The V$HM_RUN view displays information about all health monitor checks and their status. The check_name column will identify the specific type of check that was performed, such as DB Structure Integrity Check , which examines file corruptions, data dictionary inconsistencies, and other core components.