Inode usage monitoring and alerting
Some filesystems or volumes in Flex Appliance are provisioned with a fixed number of inodes at creation time. Unlike disk capacity, inode count cannot grow dynamically. A filesystem with exhausted inodes causes critical failures across the instance.
When inode consumption reaches 100%:
No new files or directories can be created on the volume, even if sufficient free disk space remains.
Application and platform services may fail to write logs, metadata, or state files.
Critical NetBackup operations may be interrupted.
Instance level services may malfunction, leading to degraded functionality or downtime.
To prevent service disruption caused by inode exhaustion, Flex Appliance introduces inode usage monitoring and alerting for all application instance volumes.
Collection of inode usage for every application instance volume
Metrics flow into the central telemetry system for visualization.
Administrators can use these insights to take preemptive action before inode exhaustion affects service availability.
Exposure of inode metrics for platform wide monitoring and troubleshooting
Automated alerting when inode usage reaches predefined thresholds.
Email and SNMP alerts are automatically sent to configured recipients.
SHI alerts appear in the UI under tab.
Early detection of inode pressure on Flex volumes.
Prevention of service interruptions caused by inability to write logs or metadata.
Improved troubleshooting with accurate inode consumption metrics.
Better operational reliability for applications running on Flex.
Flex Appliance captures and exposes the following per volume inode metrics:
Metric Name | Description |
|---|---|
storage_filesystem_files | Total number of inodes available on the filesystem |
storage_filesystem_files_used | Number of inodes currently consumed |
These metrics are made available to the platform's monitoring and observability pipeline for dashboards, analysis, and automated alert evaluation.
Inode utilization is continuously monitored. Flex Appliance triggers alerts based on the following thresholds:
Warning Threshold (80%)
Triggered when inode usage reaches 80% of available inodes.
Generates:
Warning level email alert
SHI (System Health Indicator) warning alert
Critical Threshold (90%)
Triggered when inode usage reaches 90% of available inodes.
Generates:
Critical level email alert
SHI critical alert
Both alert types include details such as:
Affected instance and volume
Total inodes
Used inodes
Current utilization percentage
Recommended corrective actions such as cleaning unnecessary files, increasing filesystem size, or reprovisioning