icon-cookie
The website uses cookies to optimize your user experience. Using this website grants us the permission to collect certain information essential to the provision of our services to you, but you may change the cookie settings within your browser any time you wish. Learn more
I agree
blank_error__heading
blank_error__body
Text direction?

Container runtimes

FEATURE STATE: Kubernetes v1.6 [stable]
To run containers in Pods, Kubernetes uses a container runtime. Here are the installation instructions for various runtimes.
Caution:

A flaw was found in the way runc handled system file descriptors when running containers. A malicious container could use this flaw to overwrite contents of the runc binary and consequently run arbitrary commands on the container host system.

Please refer to CVE-2019-5736 for more information about the issue.

Applicability

Note: This document is written for users installing CRI onto Linux. For other operating systems, look for documentation specific to your platform.

You should execute all the commands in this guide as root. For example, prefix commands with sudo , or become root and run the commands as that user.

Cgroup drivers

When systemd is chosen as the init system for a Linux distribution, the init process generates and consumes a root control group (cgroup) and acts as a cgroup manager. Systemd has a tight integration with cgroups and will allocate cgroups per process. It's possible to configure your container runtime and the kubelet to use cgroupfs. Using cgroupfs alongside systemd means that there will be two different cgroup managers.

Control groups are used to constrain resources that are allocated to processes. A single cgroup manager will simplify the view of what resources are being allocated and will by default have a more consistent view of the available and in-use resources. When we have two managers we end up with two views of those resources. We have seen cases in the field where nodes that are configured to use cgroupfs for the kubelet and Docker, and systemd for the rest of the processes running on the node becomes unstable under resource pressure.

Changing the settings such that your container runtime and kubelet use systemd as the cgroup driver stabilized the system. Please note the native.cgroupdriver=systemd option in the Docker setup below.

Caution: Changing the cgroup driver of a Node that has joined a cluster is highly unrecommended. If the kubelet has created Pods using the semantics of one cgroup driver, changing the container runtime to another cgroup driver can cause errors when trying to re-create the PodSandbox for such existing Pods. Restarting the kubelet may not solve such errors. The recommendation is to drain the Node from its workloads, remove it from the cluster and re-join it.

Docker

On each of your machines, install Docker. Version 19.03.11 is recommended, but 1.13.1, 17.03, 17.06, 17.09, 18.06 and 18.09 are known to work as well. Keep track of the latest verified Docker version in the Kubernetes release notes.

Use the following commands to install Docker on your system:

# (Install Docker CE)
## Set up the repository:
### Install packages to allow apt to use a repository over HTTPS
apt-get update && apt-get install -y \
  apt-transport-https ca-certificates curl software-properties-common gnupg2
# Add Docker’s official GPG key:
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | apt-key add -
# Add the Docker apt repository:
add-apt-repository \
  "deb [arch=amd64] https://download.docker.com/linux/ubuntu \
$(lsb_release -cs) \
  stable"
# Install Docker CE
apt-get update && apt-get install -y \
  containerd.io=1.2.13-2 \
  docker-ce=5:19.03.11~3-0~ubuntu-$(lsb_release -cs) \
  docker-ce-cli=5:19.03.11~3-0~ubuntu-$(lsb_release -cs)
# Set up the Docker daemon
cat > /etc/docker/daemon.json <<EOF
{
  "exec-opts": ["native.cgroupdriver=systemd"],
  "log-driver": "json-file",
  "log-opts": {
    "max-size": "100m"
  },
  "storage-driver": "overlay2"
}
EOF
mkdir -p /etc/systemd/system/docker.service.d
# Restart Docker
systemctl daemon-reload
systemctl restart docker
# (Install Docker CE)
## Set up the repository
### Install required packages
yum install -y yum-utils device-mapper-persistent-data lvm2
## Add the Docker repository
yum-config-manager --add-repo \
  https://download.docker.com/linux/centos/docker-ce.repo
# Install Docker CE
yum update -y && yum install -y \
  containerd.io-1.2.13 \
  docker-ce-19.03.11 \
  docker-ce-cli-19.03.11
## Create /etc/docker
mkdir /etc/docker
# Set up the Docker daemon
cat > /etc/docker/daemon.json <<EOF
{
  "exec-opts": ["native.cgroupdriver=systemd"],
  "log-driver": "json-file",
  "log-opts": {
    "max-size": "100m"
  },
  "storage-driver": "overlay2",
  "storage-opts": [
    "overlay2.override_kernel_check=true"
  ]
}
EOF
mkdir -p /etc/systemd/system/docker.service.d
# Restart Docker
systemctl daemon-reload
systemctl restart docker

If you want the docker service to start on boot, run the following command:

sudo systemctl enable docker

Refer to the official Docker installation guides for more information.

CRI-O

This section contains the necessary steps to install CRI-O as CRI runtime.

Use the following commands to install CRI-O on your system:

Note: The CRI-O major and minor versions must match the Kubernetes major and minor versions. For more information, see the CRI-O compatiblity matrix.

Prerequisites

modprobe overlay
modprobe br_netfilter

# Set up required sysctl params, these persist across reboots.
cat > /etc/sysctl.d/99-kubernetes-cri.conf <<EOF
net.bridge.bridge-nf-call-iptables  = 1
net.ipv4.ip_forward                 = 1
net.bridge.bridge-nf-call-ip6tables = 1
EOF

sysctl --system
# Debian Unstable/Sid
echo 'deb http://download.opensuse.org/repositories/devel:/kubic:/libcontainers:/stable/Debian_Unstable/ /' > /etc/apt/sources.list.d/devel:kubic:libcontainers:stable.list
wget -nv https://download.opensuse.org/repositories/devel:kubic:libcontainers:stable/Debian_Unstable/Release.key -O- | sudo apt-key add -
# Debian Testing
echo 'deb http://download.opensuse.org/repositories/devel:/kubic:/libcontainers:/stable/Debian_Testing/ /' > /etc/apt/sources.list.d/devel:kubic:libcontainers:stable.list
wget -nv https://download.opensuse.org/repositories/devel:kubic:libcontainers:stable/Debian_Testing/Release.key -O- | sudo apt-key add -
# Debian 10
echo 'deb http://download.opensuse.org/repositories/devel:/kubic:/libcontainers:/stable/Debian_10/ /' > /etc/apt/sources.list.d/devel:kubic:libcontainers:stable.list
wget -nv https://download.opensuse.org/repositories/devel:kubic:libcontainers:stable/Debian_10/Release.key -O- | sudo apt-key add -
# Raspbian 10
echo 'deb http://download.opensuse.org/repositories/devel:/kubic:/libcontainers:/stable/Raspbian_10/ /' > /etc/apt/sources.list.d/devel:kubic:libcontainers:stable.list
wget -nv https://download.opensuse.org/repositories/devel:kubic:libcontainers:stable/Raspbian_10/Release.key -O- | sudo apt-key add -

and then install CRI-O:

sudo apt-get install cri-o-1.17
# Configure package repository
. /etc/os-release
sudo sh -c "echo 'deb http://download.opensuse.org/repositories/devel:/kubic:/libcontainers:/stable/x${NAME}_${VERSION_ID}/ /' > /etc/apt/sources.list.d/devel:kubic:libcontainers:stable.list"
wget -nv https://download.opensuse.org/repositories/devel:kubic:libcontainers:stable/x${NAME}_${VERSION_ID}/Release.key -O- | sudo apt-key add -
sudo apt-get update
# Install CRI-O
sudo apt-get install cri-o-1.17
# Install prerequisites
curl -L -o /etc/yum.repos.d/devel:kubic:libcontainers:stable.repo https://download.opensuse.org/repositories/devel:kubic:libcontainers:stable/CentOS_7/devel:kubic:libcontainers:stable.repo
curl -L -o /etc/yum.repos.d/devel:kubic:libcontainers:stable:cri-o:1.18.repo https://download.opensuse.org/repositories/devel:kubic:libcontainers:stable:cri-o:1.18/CentOS_7/devel:kubic:libcontainers:stable:cri-o:1.18.repo
# Install CRI-O
yum install -y cri-o
sudo zypper install cri-o

Start CRI-O

systemctl daemon-reload
systemctl start crio

Refer to the CRI-O installation guide for more information.

Containerd

This section contains the necessary steps to use containerd as CRI runtime.

Use the following commands to install Containerd on your system:

Prerequisites

cat > /etc/modules-load.d/containerd.conf <<EOF
overlay
br_netfilter
EOF

modprobe overlay
modprobe br_netfilter

# Setup required sysctl params, these persist across reboots.
cat > /etc/sysctl.d/99-kubernetes-cri.conf <<EOF
net.bridge.bridge-nf-call-iptables  = 1
net.ipv4.ip_forward                 = 1
net.bridge.bridge-nf-call-ip6tables = 1
EOF

sysctl --system

Install containerd

# (Install containerd)
## Set up the repository
### Install packages to allow apt to use a repository over HTTPS
apt-get update && apt-get install -y apt-transport-https ca-certificates curl software-properties-common
## Add Docker’s official GPG key
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | apt-key add -
## Add Docker apt repository.
add-apt-repository \
    "deb [arch=amd64] https://download.docker.com/linux/ubuntu \
$(lsb_release -cs) \
    stable"
## Install containerd
apt-get update && apt-get install -y containerd.io
# Configure containerd
mkdir -p /etc/containerd
containerd config default > /etc/containerd/config.toml
# Restart containerd
systemctl restart containerd
# (Install containerd)
## Set up the repository
### Install required packages
yum install -y yum-utils device-mapper-persistent-data lvm2
## Add docker repository
yum-config-manager \
    --add-repo \
    https://download.docker.com/linux/centos/docker-ce.repo
## Install containerd
yum update -y && yum install -y containerd.io
## Configure containerd
mkdir -p /etc/containerd
containerd config default > /etc/containerd/config.toml
# Restart containerd
systemctl restart containerd

systemd

To use the systemd cgroup driver, set plugins.cri.systemd_cgroup = true in /etc/containerd/config.toml. When using kubeadm, manually configure the cgroup driver for kubelet

Other CRI runtimes: frakti

Refer to the Frakti QuickStart guide for more information.

Measure
Measure
Related Notes
Get a free MyMarkup account to save this article and view it later on any device.
Create account

End User License Agreement

Summary | 7 Annotations
Cgroup drivers
2020/06/18 10:37
Systemd has a tight integration with cgroups and will allocate cgroups per process
2020/06/18 10:43
cgroupfs
2020/06/18 10:44
two different cgroup managers
2020/06/18 10:44
two views of those resources
2020/06/18 10:44
systemd as the cgroup driver stabilized the system
2020/06/18 10:45
drain the Node from its workloads, remove it from the cluster and re-join it.
2020/06/18 10:45