Mount Point and Multi-Disk Monitoring

Most Linux servers have multiple partitions, mount points, or network shares that require individual monitoring. A full root partition might not tell the whole story if /var or /home is mounted separately and approaching capacity. Server Scout provides comprehensive multi-disk monitoring through specialised metrics that track each mount point individually.

Enabling Multi-Mount Monitoring

By default, Server Scout monitors overall disk usage, but you can enable detailed per-mount-point tracking using the multi_mount metric. This feature uses the df command internally to collect usage statistics for each mounted filesystem.

To enable multi-mount monitoring, add the metric to your Server Scout configuration:

# In your Server Scout config
metrics:
  - multi_mount

Once enabled, Server Scout will collect usage data for all mounted filesystems, including:

• Standard partitions (/, /home, /var, /tmp)
• Network mounts (NFS, SMB/CIFS shares)
• External storage devices
• Docker volumes and bind mounts

Detecting Read-Only Filesystem Issues

Hardware failures, filesystem corruption, or I/O errors can cause filesystems to automatically remount as read-only to prevent data loss. The fs_readonly metric detects these critical situations:

# Enable read-only filesystem detection
metrics:
  - fs_readonly

This metric monitors /proc/mounts and alerts when filesystems unexpectedly switch to read-only mode. Such events often indicate:

• Failing storage hardware
• Filesystem corruption
• I/O subsystem problems
• Network issues with remote mounts

Early detection of read-only remounts allows you to investigate and resolve underlying issues before they cause service disruptions.

Storage and Data Structure

Mount point usage data is stored in the metrics_mounts table, separate from standard system metrics. This dedicated table structure allows for efficient querying and historical analysis of individual mount points.

Each record includes:

• Mount point path
• Total space available
• Used space and percentage
• Filesystem type
• Mount options and status

You can query this data directly for custom reporting or integration with other monitoring tools.

Viewing Multi-Mount Data

The server detail page displays a comprehensive breakdown of all monitored mount points when multi-mount monitoring is enabled. Each mount point shows:

1. Current usage percentage with visual indicators
2. Available space in human-readable format
3. Filesystem type and mount options
4. Historical usage trends over your selected time period
5. Alert status for any mount points approaching capacity

Mount points are sorted by usage percentage, with the fullest filesystems displayed first for quick identification of potential issues.

Best Practices for Multi-Disk Monitoring

Partition-Specific Thresholds

Different mount points may require different alert thresholds. Consider these guidelines:

• Root partition (/): 85-90% threshold
• Log directories (/var/log): 80-85% threshold
• User data (/home): 90-95% threshold
• Temporary storage (/tmp): 70-80% threshold

Network Mount Considerations

For NFS and other network mounts:

# Ensure proper mount options for monitoring
/nfs/share nfs defaults,soft,intr 0 0

Use soft mounts where appropriate to prevent Server Scout from hanging on unresponsive network storage.

Critical System Partitions

Pay special attention to:

• /var - Contains logs, databases, and application data
• /tmp - Can fill rapidly with temporary files
• /opt - Often contains third-party applications
• /usr/local - Custom software installations

Automated Cleanup

Consider implementing automated cleanup for mount points that frequently approach capacity:

# Example: Clean old logs when /var/log reaches 80%
if [ $(df /var/log | tail -1 | awk '{print $5}' | sed 's/%//') -gt 80 ]; then
    find /var/log -name "*.log.gz" -mtime +30 -delete
fi

Troubleshooting Common Issues

If mount points aren't appearing in Server Scout, verify:

1. The filesystem is properly mounted and accessible
2. Server Scout has read permissions for all mount points
3. Network mounts are stable and responsive
4. The multi_mount metric is correctly configured

Multi-disk monitoring provides essential visibility into storage usage across complex server configurations, helping prevent service disruptions and identify hardware issues before they become critical problems.