Drill-Down Metrics and Expanded Panels

Server Scout's panel system provides a clean, uncluttered view of your server's vital statistics whilst offering comprehensive detail when you need it. Each monitoring panel displays its most critical metric prominently in the header, with additional drill-down metrics available through panel expansion.

Understanding Panel Structure

Every Server Scout panel follows a consistent design pattern. The header shows the primary metric—the most important indicator for that system component. When you expand a panel by clicking on it, you'll reveal additional sub-metrics that provide deeper insight into system behaviour.

All drill-down metrics include their own individual graphs with the same time range selectors as the main dashboard, allowing you to correlate events across different timeframes.

CPU Panel Drill-Downs

The CPU panel's header displays overall CPU utilisation percentage. Expanding the panel reveals:

• User %: CPU time spent executing user processes
• System %: CPU time consumed by kernel operations
• IO Wait %: CPU idle time whilst waiting for disk I/O operations
• Steal %: CPU time "stolen" by the hypervisor (relevant in virtualised environments)
• Temperature: Current CPU temperature readings (when sensors are available)
• Context Switches: Number of context switches per second, indicating system scheduling activity

These metrics help identify whether high CPU usage stems from user applications, system overhead, or I/O bottlenecks.

Memory Panel Drill-Downs

Beyond the primary memory usage percentage, the memory panel provides:

• Available Memory: Total memory available to applications
• Cached Memory: Memory used for file system caching
• Swap Usage: Current swap space utilisation
• Dirty Pages: Modified memory pages awaiting write to storage
• Buffers: Memory allocated for I/O buffers

These metrics distinguish between different types of memory usage, helping you understand whether apparent high memory usage is actually efficient caching or genuine memory pressure.

Network Panel Drill-Downs

The network panel header shows combined network throughput. Drill-down metrics include:

• RX/TX Error Counts: Network errors on incoming and outgoing traffic
• RX/TX Dropped Packets: Packets dropped due to buffer overflow or other issues
• Interface Name: Network interface identifier
• Private IP: Internal IP address assignment
• MAC Address: Hardware address of the network interface

These details help diagnose network connectivity issues and identify problematic interfaces.

Disk Panel Drill-Downs

The disk panel's primary metric shows overall disk usage. Expanded view provides:

• Per-Mount Breakdowns: Individual usage statistics for each mounted filesystem
• Inodes Usage: Available and used inodes for each filesystem
• Read-Only Filesystem Detection: Identification of filesystems mounted in read-only mode

This granular view helps identify which specific filesystems are approaching capacity and whether issues relate to storage space or inode exhaustion.

Load Panel Drill-Downs

The load panel offers the most comprehensive drill-down capabilities, expanding beyond the header's load average to show:

• Running Processes: Currently executing processes count
• Blocked Processes: Processes waiting for I/O or other resources
• Zombies: Defunct processes awaiting cleanup
• Open File Descriptors: Total system file descriptor usage
• TCP Connections: Current TCP connection count
• Top Processes Table: Real-time view of highest resource-consuming processes
• Services List: Status of monitored system services

Utilising Time Range Selectors

Each drill-down metric includes interactive graphs with time range selectors matching the main dashboard. You can view data across different periods:

• Last hour for immediate troubleshooting
• Last 24 hours for daily pattern analysis
• Last week for trend identification
• Custom ranges for specific incident investigation

Best Practices

When investigating performance issues, start with header metrics to identify affected systems, then expand relevant panels to drill down into specific causes. The correlation between different metrics often reveals root causes—for example, high IO wait alongside specific disk mount point saturation.

Use the consistent time range selectors to correlate events across different system components, helping you understand how CPU spikes relate to memory pressure or network activity.