Skip to content

Month: November 2024

The Problem: ESXi Host Loses Network Connectivity

Published by Luciano Batalha on November 16, 2024

Step-by-Step Troubleshooting Guide

  1. Verify Network Configuration on the Host
  • Log in via DCUI (Direct Console User Interface) or SSH.
  • Check the IP address, gateway, and DNS settings:

esxcli network ip interface ipv4 get

  • Confirm that the host can ping the gateway or vCenter server:

ping <gateway-IP>

  1. Check vSwitch and Port Group Settings
  • In the vSphere Client, navigate to Networking > Virtual Switches.
  • Ensure the following:
    • The physical NICs (vmnics) are attached and active.
    • Port groups have the correct VLAN ID settings.
  • Verify the load balancing and failover policies for inconsistencies.
  1. Inspect Physical Network Connections
  • Check for issues with cables, switches, or ports.
  • Test connectivity using tools like link lights on NICs or port activity indicators on the switch.
  • Replace faulty cables or move connections to different switch ports if needed.
  1. Test and Reconfigure NICs
  • Verify the status of all NICs:

esxcli network nic list

  • Re-enable or restart any problematic NICs:

esxcli network nic down -n <vmnic-name>
esxcli network nic up -n <vmnic-name>

  1. Address Driver or Firmware Issues
  • Check the HCL (Hardware Compatibility List) for your ESXi version.
  • Update or reinstall the NIC driver and firmware if outdated or incompatible:

esxcli software vib install -v /path/to/driver.vib

  1. Monitor for IP Conflicts
  • Use tools like ARP tables on the switch or router to detect conflicting IP addresses.
  • Assign a new static IP address to the host if conflicts are found.
  1. Restart Management Network
  • Restart the management network via the DCUI:
    • Select Troubleshooting Options > Restart Management Network.
  • Alternatively, restart it via SSH:

services.sh restart

  1. Examine Logs for Deeper Insights
  • Review network-related logs for clues:

tail -f /var/log/vmkernel.log

tail -f /var/log/hostd.log

Resolving a Common VMware ESXi Issue – PSOD on Boot

Published by Luciano Batalha on November 15, 2024

The PSOD is VMware’s equivalent of the “blue screen” in Windows. It halts the ESXi host and displays diagnostic information in a purple background. One scenario that often triggers a PSOD is a hardware compatibility or driver issue, especially after an upgrade or new hardware deployment.

Root Cause Analysis

Common reasons for a PSOD on boot include:

  1. Incompatible Drivers: Using a driver version that doesn’t match the hardware or ESXi version.
  2. Faulty Hardware: Issues with RAM, storage controllers, or network adapters.
  3. Configuration Errors: Misconfigured BIOS or firmware settings.
  4. Corrupted Filesystem: Problems with the ESXi boot partition.

Step-by-Step Resolution

  1. Gather Information from the PSOD Screen
  • Note the error message and codes displayed on the PSOD.
  • Look for references to specific drivers, memory modules, or hardware.
  1. Reboot in Recovery Mode
  • Restart the host and enter Recovery Mode from the boot menu.
  • Check system logs using the esxcli command:

esxcli system syslog view –log /var/log/vmkernel.log

 

  1. Verify Hardware Compatibility
  • Cross-check the hardware against VMware’s Hardware Compatibility List (HCL) to ensure support for your ESXi version.
  • If issues arise, update firmware or replace problematic components.
  1. Roll Back Drivers
  • If a driver is causing the issue, try rolling back to a previous version:

software vib remove -n <driver-name>

 

  • Reboot the host and confirm stability.
  1. Check for Corrupted Boot Files
  • Boot the ESXi installation media and select Repair System.
  • Reinstall or repair the ESXi system files without wiping the datastore.
  1. Update or Patch ESXi
  • Check VMware’s knowledge base for patches or updates related to your error code.
  • Update your ESXi host:

esxcli software profile update -d <URL-to-depot>