Disk Space Health Check

As part of its suite of health checks, StorNext Storage Manager periodically runs the "Disk Space" health check to find all file systems in use by StorNext that are running out of space. This check includes all file systems that are accessible from /usr/adic, including those reached by symbolic link. This check does not include any SNFS file systems configured by the user.

The Disk Space health check and the CRITICAL_FILL_LEVEL parameter for /usr/adic/util/sn_log_update.cfg can be used together to determine desired notification behavior. The system will send an admin alert warning if the Disk Space health check exceeds the health check's warning threshold. When the sn_log_update CRITICAL_FILL_LEVEL triggers, it will cause Storage Manager logs to start being automatically removed from the system until the percentage of the file system in use falls below the CRITICAL_FILL_LEVEL. The Disk Space health check generates a RAS event if the health check's fail threshold is exceeded. If users continually run with low disk space conditions, they may want to consider dropping the CRITICAL_FILL_LEVEL to match the Disk Space warning threshold.

By default, the Disk Space health check will report warnings if a file system is above 95% disk usage, and it reports failures if a file system is above 99% disk usage. If desired, the user may override these defaults by editing the /usr/adic/TSM/config/filelist configuration file. Find the health_check entry with the checkDiskSpaceTsm command. Append the -w option to the command to override the percentage threshold for warning messages. Append the -f option to the command to override the percentage threshold for failure messages. For example:

In the example above, a user has configured the Disk Space health check to issue file system space usage warning messages at 96% usage and failure messages at 98% usage.