How to enable memory/CPU hot plugging on my Linux server?
Posted by Dovydas Š., Last modified by Vytautas K. on 14 April 2018 09:36 AM

With the release of the updated OS templates comes a major feature: memory/CPU hot plugging. Only customers with then newest OS template version will have this enabled by default. Clients using older OS templates should follow these steps to enable it:

STEP 1. Make sure the kernel version is >= 3.2.0 (Note: Debian 7 did not work with kernel 3.2.0 in our testing. Seems like a kernel issue, but we didn't want to taint the Debian template by using custom/3rd party kernels)

STEP 2.  Add/update udev rules. Generally the user rules reside in /etc/udev/rules.d/. Create a text file to that directory, called 40-cpu-mem-hotplug.rules (so full path looks like /etc/udev/rules.d/40-cpu-mem-hotplug.rules

Add the following content to that file:

# CPU hotadd request
SUBSYSTEM=="cpu", ACTION=="add", TEST=="online", ATTR{online}=="0", ATTR{online}="1"

# Memory hotadd request
SUBSYSTEM=="memory", ACTION=="add", ATTR{state}=="offline", ATTR{state}="online"

STEP 3. Depending on the distribution, these udev rules might be already present. For example CentOS 6, 7. You can check the following directories for udev rules:


On CentOS, the already present rules file is called 40-redhat.rules.

STEP 4. Then you need to reboot your server once, using the controls in the client area (not from inside the server). This is required to regenerate your server configuration in the backend, so that the backend can then hotplug cpu/memory during upgrades.

(0 vote(s))
Not helpful