VMware Memory Hotplug Linux: Ubuntu Memory Hotplug VMware, VMware Memory CPU Hotplug CentOS, and VMware Hotplug Memory Linux — Complete Per-Distro Guide

Dynamic scaling of VM resources is critical in production workloads where downtime is unacceptable. VMware Memory Hotplug allows administrators to add RAM to a running Linux VM without rebooting — eliminating service interruptions during demand spikes. Unlike CPU hotplug, which has limited OS edition support, memory hotplug is broadly supported across modern Linux distributions.

This guide focuses on Ubuntu, CentOS, and Debian — the three most common Linux server platforms in VMware environments. You’ll learn:

• How to enable memory hotplug in vSphere and confirm guest OS support.
• Kernel and configuration requirements for each distribution.
• Step‑by‑step verification that the guest recognizes newly added memory.
• Limitations and best practices to avoid performance or NUMA issues.

📌 Core takeaway → With proper configuration, Linux guests on VMware can scale memory live, ensuring workloads remain responsive without planned downtime.

Enabling VMware Memory Hotplug in vSphere: Step‑by‑Step

(Keyword: vmware hotplug memory linux)

Enable Memory and CPU Hot‑Add via vSphere Client

1. 1. 🔌 Power off the target VM.
2. 2. In the vSphere Client, right‑click the VM → Edit Settings.
3. 3. On the Virtual Hardware tab:

• Expand Memory → check Enable Hot Plug.
• Expand CPU → check Enable CPU Hot Add.

1. 4. Click OK and power on the VM.
2. 5. ✅ Verify hot‑add is enabled by reopening Edit Settings while the VM is running — the checkboxes should reflect the configured values.

Enable Memory Hot‑Add via VMX File (ESXi SSH)

For environments without vSphere Client access or when automating template configuration:

1. 1. 🔌 Power off the VM.
2. 2. SSH into the ESXi host.
3. 3. Navigate to the VM directory:

<code class="language-Bash">cd /vmfs/volumes/[datastore]/[VMname]</code>

1. 4. Edit the VMX file:

<code class="language-Bash">vi VMname.vmx</code>

1. 5. Add or confirm these lines:

<code class="language-Bash">mem.hotadd = "TRUE"
    vcpu.hotadd = "TRUE"</code>

1. 6. Save the file and power on the VM.

Hot‑Adding Memory After the VM Is Running

1. 1. With the VM running and hot‑add enabled, right‑click the VM in vSphere → Edit Settings.
2. 2. In the Memory field, increase the allocation to the desired value. This field is editable on a running VM only when hot‑add is enabled.
3. 3. Click OK. vSphere delivers the new memory blocks to the VM via ACPI hotplug events.
4. 4. 🐧 Inside the guest OS, verify whether the memory was auto‑onlined or requires manual intervention (see per‑distribution sections below).

<code class="language-Bash">grep -i offline /sys/devices/system/memory/*/state</code>

<code class="language-Bash">/sys/devices/system/memory/memory100/state: offline</code>

<code class="language-Bash">echo online &gt; /sys/devices/system/memory/memory100/state</code>

<code class="language-Bash">for mem in /sys/devices/system/memory/*/state; do
        if [ "$(cat $mem)" = "offline" ]; then
        echo online &gt; "$mem"
        echo "Onlined: $mem"
        fi
        done</code>

<code class="language-Bash">free -m</code>

<code class="language-Bash">sudo nano /etc/default/grub</code>

<code class="language-Code">GRUB_CMDLINE_LINUX_DEFAULT="quiet splash memhp_default_state=online"</code>

<code class="language-Bash">sudo update-grub</code>

<code class="language-Bash">echo 'SUBSYSTEM=="memory", ACTION=="add", ATTR{state}=="offline", ATTR{state}="online"' | \
        sudo tee /etc/udev/rules.d/40-hotplug-memory.rules</code>

<code class="language-Bash">sudo udevadm control --reload-rules</code>

VMware Memory CPU Hotplug CentOS: RHEL and CentOS Automatic Onlining

🤖 Why RHEL and CentOS Handle Hotplug Automatically

Red Hat Enterprise Linux, CentOS, Rocky Linux, AlmaLinux, and CentOS Stream include default udev rules and systemd configurations that automatically online hotplugged memory and CPUs.

• When ESXi delivers a new memory block via ACPI, the RHEL‑family kernel brings it online immediately — no administrator action required.
• Running free -m after a vSphere hot‑add shows the updated total memory instantly.
• CPU hot‑add works the same way: new cores appear in /proc/cpuinfo and lscpu without manual onlining.

📌 Key point → RHEL‑based distributions are designed for seamless hotplug operations, making them ideal for production workloads.

🛠️ Step‑by‑Step: Verify and Online Hotplugged Memory in CentOS/RHEL (Manual Path)

In most cases, auto‑onlining works. Use this manual procedure only if it does not (older CentOS 6/7 or custom kernels):

1. 1. After vSphere hot‑adds memory, check for offline blocks:

<code class="language-Bash">grep -i offline /sys/devices/system/memory/*/state</code>

Look for entries marked offline.

1. 2. Online a specific block (replace memoryN with the block number):

<code class="language-Bash">echo online &gt; /sys/devices/system/memory/memoryN/state</code>

1. 3. Verify memory is registered:

<code class="language-Bash">free -m</code>

1. 4. For CPU hotplug, verify new CPUs:

<code class="language-Bash">lscpu
    cat /proc/cpuinfo | grep "processor"</code>

If CPUs appear offline:

<code class="language-Bash">echo 1 &gt; /sys/devices/system/cpu/cpuN/online</code>

⚙️ Enable acpiphp Module on Older CentOS/RHEL for CPU Hotplug

On CentOS 6 and early CentOS 7, the acpiphp module must be loaded for the kernel to detect hot‑added CPUs and PCI devices:

<code class="language-Bash">sudo modprobe acpiphp</code>

To load automatically at boot:

<code class="language-Bash">echo "acpiphp" | sudo tee /etc/modules-load.d/acpiphp.conf</code>

📌 Modern RHEL 8/9 and CentOS Stream handle this automatically — manual module loading is only required on legacy distributions.

<code class="language-Bash">grep -i offline /sys/devices/system/memory/*/state</code>

<code class="language-Bash">for mem in /sys/devices/system/memory/*/state; do
        [ "$(cat $mem)" = "offline" ] &amp;&amp; echo online &gt; "$mem"
        done</code>

<code class="language-Bash">free -m</code>

SLES: Loading Required Modules for Memory Hotplug

🧩 SLES‑Specific Module Requirements

SUSE Linux Enterprise Server (SLES) requires two kernel modules to be loaded before memory hotplug events are processed (per Broadcom KB 309841). Without them, ESXi delivers the ACPI event but the guest kernel ignores it.

Step‑by‑Step:

1. 1. 📦 Load the ACPI hotplug module:

<code class="language-Bash">sudo modprobe acpiphp</code>

1. 2. 📦 Load the ACPI memory hotplug module:

<code class="language-Bash">sudo modprobe acpi_memhotplug</code>

1. 3. ⚙️ In the vSphere Client, increase the VM’s memory allocation while the VM is running.
2. 4. 🔍 Check for offline memory blocks:

<code class="language-Bash">grep -i offline /sys/devices/system/memory/*/state</code>

1. 5. ➕ Online any offline blocks:

<code class="language-Bash">echo online &gt; /sys/devices/system/memory/memory[number]/state</code>

1. 6. ✅ Verify memory is registered:

<code class="language-Bash">free -m</code>

Auto‑load at boot: To ensure these modules are always available, add both to /etc/modules-load.d/acpi-hotplug.conf:

<code class="language-Code">acpiphp
    acpi_memhotplug</code>

📌 Key takeaway → Unlike Ubuntu, Debian, or RHEL‑based systems, SLES requires explicit module loading for hotplug to function. Once configured, memory hotplug works seamlessly in VMware environments.

Memory Hotplug Linux VMware: Troubleshooting Common Failures

🛑 Memory Added in vSphere but Not Visible in /sys/devices/system/memory/

• Cause → VMware Tools not installed or outdated. The guest OS is not receiving ACPI hotplug events from ESXi.
• Fix →
  • Ubuntu/Debian:

<code class="language-Bash">sudo apt install open-vm-tools</code>

• CentOS/RHEL:

After installing VMware Tools, retry the hot‑add operation in vSphere.

⚙️ Hot‑Add Option Is Greyed Out in vSphere Edit Settings

• Cause → VM is powered on and hot‑add was not enabled before power‑on, or the virtual hardware version does not support hot‑add.
• Fix →
  • Power off the VM.
  • Check virtual hardware version (minimum version 7 required).
  • Enable hot‑add while powered off, then power on again.

📉 Memory Appears Online in /sys but free -m Shows No Change

• Cause → Memory blocks were onlined but the kernel allocator has not refreshed. Rare on modern kernels but possible.
• Fix →

<code class="language-Bash">sudo sync &amp;&amp; echo 3 &gt; /proc/sys/vm/drop_caches</code>

Then re‑run:

<code class="language-Bash">free -m</code>

If the issue persists, the online operation may have silently failed. Re‑check block states:

<code class="language-Bash">grep -i offline /sys/devices/system/memory/*/state</code>

🚦 vNUMA Performance Degradation After Enabling Hot‑Add

• Cause → Hot‑add disables vNUMA. VMs spanning multiple physical NUMA nodes experience degraded memory bandwidth when topology is hidden from the guest OS.
• Fix →
  • Do not enable hot‑add on VMs with >8 vCPUs spanning multiple NUMA nodes.
  • Provision adequate resources at initial deployment.
  • Use planned maintenance windows for resource changes instead of hot‑add.

📌 Key takeaway → Most hotplug failures stem from missing VMware Tools, unsupported hardware versions, allocator refresh issues, or vNUMA trade‑offs. Systematic checks of /sys, free -m, and vSphere settings resolve the majority of problems.

Ready to get your data back?

To start VMware data recovery (recovering your data, documents, databases, images, videos, and other files), press the FREE DOWNLOAD button below to get the latest version of DiskInternals VMFS Recovery® and begin the step-by-step recovery process. You can preview all recovered files absolutely for FREE. To check the current prices, please press the Get Prices button. If you need any assistance, please feel free to contact Technical Support. The team is here to help you recover deleted VMware virtual machine!

FAQ

Related articles

FREE DOWNLOADVer 4.27, WinBUY NOWFrom $699