Author: Danilo Chiacchio

Lustre filesystem commands can seem intimidating when an administrator first starts managing an HPC cluster. A single filesystem may involve metadata servers, object storage servers, multiple targets, specialized networking, and hundreds of clients accessing data at the same time. Fortunately, daily Lustre administration does not always require advanced knowledge of its internal architecture. A relatively small set of commands can reveal whether the filesystem is mounted, how much storage remains, where files are located, and which components may be experiencing problems. This practical guide covers the basic Lustre filesystem commands administrators can use for routine checks and first-level troubleshooting. The…

Slurm administration commands are part of my daily routine when managing HPC clusters. Whether I need to investigate scheduling issues, monitor running jobs, inspect compute nodes, or respond to unexpected incidents, having the right commands readily available makes cluster administration faster and more reliable. Although Slurm provides hundreds of commands and options, only a core set is required for most administrative tasks. In my experience, mastering these commands significantly reduces troubleshooting time and helps maintain cluster stability while supporting users more efficiently. This guide presents the essential Slurm administration commands I regularly use to manage HPC environments. The examples assume…

This article shows a practical workflow for NSX troubleshooting communication failures between two virtual machines connected to the same NSX overlay segment but running on different ESXi Host Transport Nodes. In this scenario, VM A and VM B are connected to the NSX Web Segment. Because the VMs reside on different ESXi hosts, their traffic must traverse the NSX overlay between the source and destination Transport Endpoint (TEP) interfaces. The troubleshooting workflow starts at the guest operating system and gradually moves through the NSX segment, Transport Nodes, TEP connectivity, Traceflow, and packet capture. Step 1: Are the virtual machines configured…

Packet loss troubleshooting can become difficult because the symptoms often look similar regardless of where the problem actually occurs. A slow application, unstable SSH session, failed API request, or poor storage performance may all eventually lead an administrator to suspect that packets are being dropped somewhere along the communication path. The real challenge in packet loss troubleshooting is not simply proving that packets are being lost, but determining where the loss happens. A Linux server can lose packets because of local interface errors, overloaded queues, kernel processing limits, switch problems, upstream congestion, firewall behavior, or issues much farther away from…

Linux process resource usage analysis is one of the first steps to take when a server becomes slow, unstable, or unresponsive. Instead of immediately restarting services or adding hardware, administrators should identify which resource is under pressure and which process is responsible for it. A high load average does not always mean that the CPUs are overloaded, just as low free memory does not automatically indicate a memory problem. Linux uses available memory for cache and may report high system load when processes are waiting for disk operations rather than CPU time. In this small guide, we will investigate CPU,…

Troubleshooting vCenter and ESXi heartbeat communication is an important step when an ESXi host intermittently appears as Not Responding, loses communication with vCenter Server, or experiences management failures while its virtual machines continue running normally. In a healthy vSphere environment, the vpxa agent on each ESXi host participates in the management communication with vCenter Server and sends heartbeat traffic over UDP port 902. These heartbeats are monitored by the vpxd service on vCenter and provide a lightweight way to confirm that the host remains reachable through the management network. By default, heartbeat traffic is sent approximately every 10 seconds. If…