v1.21.1

kubernetes v1.21.1

• dqd:
  • ssst0n3/docker_archive:kubernetes-v1.21.1 -> ssst0n3/docker_archive:kubernetes-v1.21.1_v0.1.0
  • ssst0n3/docker_archive:kubernetes-v1.21.1_v0.1.0
• ctr:
  • ssst0n3/docker_archive:ctr_kubernetes-v1.21.1 -> ssst0n3/docker_archive:ctr_kubernetes-v1.21.1_v0.1.0
  • ssst0n3/docker_archive:ctr_kubernetes-v1.21.1_v0.1.0

usage

$ cd kubernetes/v1.21.1
$ docker compose -f docker-compose.yml -f docker-compose.kvm.yml up -d

$ kubectl --kubeconfig=kubeconfig get pods -A
NAMESPACE     NAME                                        READY   STATUS    RESTARTS   AGE
kube-system   coredns-558bd4d5db-l2btw                    0/1     Pending   0          161m
kube-system   coredns-558bd4d5db-ptzxt                    0/1     Pending   0          161m
kube-system   etcd-kubernetes-1-21-1                      1/1     Running   1          161m
kube-system   kube-apiserver-kubernetes-1-21-1            1/1     Running   1          161m
kube-system   kube-controller-manager-kubernetes-1-21-1   1/1     Running   1          161m
kube-system   kube-proxy-tp9vv                            1/1     Running   1          161m
kube-system   kube-scheduler-kubernetes-1-21-1            1/1     Running   1          161m

$ ./ssh
root@kubernetes-1-21-1:~# helm version
version.BuildInfo{Version:"v3.6.0", GitCommit:"7f2df6467771a75f5646b7f12afb408590ed1755", GitTreeState:"clean", GoVersion:"go1.16.3"}
root@kubernetes-1-21-1:~# kubectl version
Client Version: version.Info{Major:"1", Minor:"21", GitVersion:"v1.21.1", GitCommit:"5e58841cce77d4bc13713ad2b91fa0d961e69192", GitTreeState:"clean", BuildDate:"2021-05-12T14:18:45Z", GoVersion:"go1.16.4", Compiler:"gc", Platform:"linux/amd64"}
Server Version: version.Info{Major:"1", Minor:"21", GitVersion:"v1.21.1", GitCommit:"5e58841cce77d4bc13713ad2b91fa0d961e69192", GitTreeState:"clean", BuildDate:"2021-05-12T14:12:29Z", GoVersion:"go1.16.4", Compiler:"gc", Platform:"linux/amd64"}
root@kubernetes-1-21-1:~# containerd --version
containerd github.com/containerd/containerd v1.5.1 12dca9790f4cb6b18a6a7a027ce420145cb98ee7
root@kubernetes-1-21-1:~# runc --version
runc version 1.0.0-rc94
spec: 1.0.2-dev
go: go1.14.15
libseccomp: 2.5.1
root@kubernetes-1-21-1:~# uname -a
Linux kubernetes-1-21-1 5.4.0-216-generic #236-Ubuntu SMP Fri Apr 11 19:53:21 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
root@kubernetes-1-21-1:~# cat /etc/os-release 
NAME="Ubuntu"
VERSION="20.04.6 LTS (Focal Fossa)"
ID=ubuntu
ID_LIKE=debian
PRETTY_NAME="Ubuntu 20.04.6 LTS"
VERSION_ID="20.04"
HOME_URL="https://www.ubuntu.com/"
SUPPORT_URL="https://help.ubuntu.com/"
BUG_REPORT_URL="https://bugs.launchpad.net/ubuntu/"
PRIVACY_POLICY_URL="https://www.ubuntu.com/legal/terms-and-policies/privacy-policy"
VERSION_CODENAME=focal
UBUNTU_CODENAME=focal

build

make all DIR=kubernetes/v1.21.1

for developers

# syntax=docker/dockerfile:1-labs
FROM ssst0n3/docker_archive:ctr_kubernetes-v1.21.1_v0.1.0
...
RUN --security=insecure ["/sbin/init", "--log-target=kmsg"]

• use dmesg -w to see build logs.
• use systemd service to execute commands
• ssh root/root 10.0.2.16 to debug

Tricks

1. Use a cache mount to avoid overlayfs-on-overlayfs errors

If you encounter the following error:

[ 2237.540485] overlayfs: filesystem on '/var/lib/containerd/io.containerd.snapshotter.v1.overlayfs/snapshots/72/fs' not supported as upperdir

it typically means an overlayfs is being mounted over another overlayfs.
You can solve this issue by using a cache mount within your build process.

Example:

# copy image snapshots
RUN --mount=type=cache,id=kubernetes-v1.21.1-snapshots,target=/trick \
    cp -a /var/lib/containerd/io.containerd.snapshotter.v1.overlayfs/* /trick/
# kubeadm init under ext4 fs
RUN --mount=type=cache,id=kubernetes-v1.21.1-snapshots,target=/var/lib/containerd/io.containerd.snapshotter.v1.overlayfs \
    # fix kube-proxy `Failed to load kernel module`
    --mount=type=bind,src=/modules,target=/lib/modules \
    --security=insecure \
    ["/sbin/init", "--log-target=kmsg"]
# skip overlayfs whiteout files (c 0,0)
RUN --mount=type=cache,id=kubernetes-v1.21.1-snapshots,target=/trick \
    # all these files are safe to delete, list each file path here for more clear
    cd /trick/snapshots/32/fs && rm \
        var/lib/apt/lists/security.debian.org_debian-security_dists_buster_updates_main_binary-amd64_Packages.lz4 \
        var/lib/apt/lists/deb.debian.org_debian_dists_buster-updates_main_binary-amd64_Packages.lz4 \
        var/lib/apt/lists/deb.debian.org_debian_dists_buster-backports_InRelease \
        var/lib/apt/lists/lock \
        var/lib/apt/lists/deb.debian.org_debian_dists_buster_main_binary-amd64_Packages.lz4 \
        var/lib/apt/lists/partial \
        var/lib/apt/lists/deb.debian.org_debian_dists_buster-updates_InRelease \
        var/lib/apt/lists/deb.debian.org_debian_dists_buster_InRelease \
        var/lib/apt/lists/deb.debian.org_debian_dists_buster-backports_main_binary-amd64_Packages.lz4 \
        var/lib/apt/lists/auxfiles \
        var/lib/apt/lists/security.debian.org_debian-security_dists_buster_updates_InRelease \
        var/log/alternatives.log \
        var/log/dpkg.log \
        var/log/apt \
        var/cache/debconf/templates.dat \
        var/cache/debconf/passwords.dat \
        var/cache/debconf/config.dat && \
    # use tar to preserve file capabilities
    tar -C /trick -cf - . | tar -C /var/lib/containerd/io.containerd.snapshotter.v1.overlayfs/ -xf -

This approach stores containerd snapshots in a temp cache directory, avoiding nested overlayfs layers and improving build consistency.

NOTE: mknod is not allowed on overlayfs, even when running as a privileged container.

2. Fix systemctl exit 0 inside containers

When running systemctl exit 0, you might see this warning:

ExitCode can only be set for user service managers or in containers.

This happens because systemd only allows this command in a detected "container environment".
To fix this, define the container environment variable before running systemd:

# Workaround for systemd bug affecting `systemctl exit 0`
ENV container=docker

Setting ENV container=docker ensures that systemd recognizes the environment as a container, allowing graceful exits via systemctl exit 0.