Author: Danilo Chiacchio

vCenter Server continuously monitors communication with the ESXi hosts that it manages. As part of this communication, the vpxa agent running on an ESXi host sends heartbeat packets to vCenter Server over UDP port 902. Under normal conditions, these heartbeat packets are transmitted approximately every 10 seconds. If vCenter Server does not receive a heartbeat within its configured timeout window, the ESXi host can transition to a Not Responding state in the vSphere Client. Current Broadcom guidance describes the default timeout window as approximately 60 seconds, meaning that multiple consecutive heartbeat messages must be missed before the host is marked as unavailable. The important point is…

Linux swap usage can sometimes look confusing, especially when a server still has several gigabytes of available memory. At first glance, this may seem like a sign of memory pressure or poor system performance. In reality, Linux memory management is designed to use RAM efficiently rather than keep swap permanently at zero. The kernel may move inactive pages to swap even when physical memory is still available. Understanding Linux swap usage requires looking beyond the amount of free memory. Swap activity, available RAM, page faults, workload behavior, and kernel settings provide a much clearer picture of what is actually happening.…

The Lustre file system is one of the most widely used parallel storage solutions in High-Performance Computing (HPC) environments. In this guide, I introduce the core concepts behind Lustre, explaining how its distributed architecture enables multiple clients to access massive datasets efficiently while delivering the high throughput required by modern HPC clusters. Before deploying or administering a Lustre environment, it is important to understand how its architecture differs from a traditional filesystem. Unlike local filesystems such as ext4 or XFS, the Lustre file system distributes metadata and file data across dedicated servers, allowing storage resources to scale while supporting parallel…

Recover ESXi Host Network shows all the necessary steps to recover an ESXi management network when connected to a vSphere Distributed Switch (VDS). First things first: Describing the Scenario So, in this example, we have a four-node vSAN cluster, and the first ESXi host, for some reason (wrong configuration such as VLAN tagging, etc), is with “not responding” status, as we can see: All nodes are using a vSphere Distributed Switch (VDS), and all VMkernels adapters communicate through Port Groups attached to this VDS: Each ESXi host has four physical adapters: Each Port group uses two physical adapters: one Active…

tmux on Linux is one of the tools I rely on most when working on remote servers and managing multiple terminal tasks simultaneously. Instead of opening several terminal windows, I use tmux to organize sessions, split the screen into panes, and keep long-running processes active even after disconnecting from an SSH session. In day-to-day Linux administration, tmux makes it much easier to monitor logs, edit configuration files, execute maintenance tasks, and work with multiple remote systems from a single terminal window. Its lightweight design and wide availability across Linux distributions make it an essential utility for system administrators and developers…

A Slurm node DRAINED state means the scheduler has intentionally prevented a compute node from receiving new jobs. This usually happens because Slurm, a health-check mechanism, or a cluster administrator identified a condition that makes the node unsuitable for additional workloads. Seeing drain or drained in sinfo does not immediately tell you what failed. The node may have a hardware problem, a full filesystem, an invalid Slurm configuration, a failed Prolog or Epilog, or simply have been drained manually for maintenance. Fortunately, Slurm normally stores a reason with the node state. By combining sinfo, scontrol, system logs, and a few Linux checks, an administrator can usually determine exactly why the…