infra/modules/create-template-vm/cloud_init.yaml

102 lines
3.4 KiB
YAML

#cloud-config
hostname: terraform-vm
users:
- name: wizard
sudo: ALL=(ALL) NOPASSWD:ALL
ssh_authorized_keys:
- ${authorized_ssh_key}
passwd: ${passwd}
lock_passwd: false # enable passwd login
shell: /bin/bash
package_update: true
package_upgrade: true
packages:
- htop
- vim
- curl
- jq
- tcpdump
- tree
- tmux
- wget
- net-tools
- zsh
- apt-transport-https
- ca-certificates
- gpg
- isc-dhcp-client
- cloud-guest-utils # to enable resizing of disk via growpart
- qemu-guest-agent
- nginx
# docker
- docker-ce
- docker-ce-cli
- containerd.io
- docker-buildx-plugin
- docker-compose-plugin
%{if is_k8s_template}
# kubernetes
- kubeadm
- kubelet
# iSCSI client for CSI-backed database storage
- open-iscsi
%{endif}
apt:
sources:
%{if is_k8s_template}
kubernetes:
source: "deb https://pkgs.k8s.io/core:/stable:/v1.32/deb/ /"
keyid: "DE15B14486CD377B9E876E1A234654DA9A296436"
filename: kubernetes.list
%{endif}
docker:
source: "deb https://download.docker.com/linux/ubuntu noble stable"
keyid: "9DC858229FC7DD38854AE2D88D81803C0EBFCD88"
filename: docker.list
runcmd:
# Enable persistent journald logging for crash forensics
- mkdir -p /var/log/journal
- sed -i 's/#Storage=auto/Storage=persistent/' /etc/systemd/journald.conf
- systemctl restart systemd-journald
%{if is_k8s_template}
# Disable unattended-upgrades to prevent unexpected kernel updates that can break containerd/kubelet
# (Root cause of 26h cluster outage: unattended-upgrades → kernel update → containerd failure)
- systemctl disable --now unattended-upgrades || true
- apt-get remove -y unattended-upgrades || true
- apt-mark hold kubelet kubeadm kubectl
- systemctl stop kubelet
- containerd config default | sudo tee /etc/containerd/config.toml
- ${containerd_config_update_command}
- systemctl restart containerd
- systemctl enable --now iscsid
# Harden iSCSI: increase recovery timeout (300s vs 120s default) and enable
# CRC32C data/header digests to detect bit flips over the network.
# Prevents SQLite corruption from transient iSCSI session drops.
- sed -i 's/^node.session.timeo.replacement_timeout = .*/node.session.timeo.replacement_timeout = 300/' /etc/iscsi/iscsid.conf
- sed -i 's/^node.conn\[0\].timeo.noop_out_interval = .*/node.conn[0].timeo.noop_out_interval = 10/' /etc/iscsi/iscsid.conf
- sed -i 's/^node.conn\[0\].timeo.noop_out_timeout = .*/node.conn[0].timeo.noop_out_timeout = 15/' /etc/iscsi/iscsid.conf
- |
if ! grep -q '^node.conn\[0\].iscsi.HeaderDigest' /etc/iscsi/iscsid.conf; then
echo 'node.conn[0].iscsi.HeaderDigest = CRC32C,None' >> /etc/iscsi/iscsid.conf
echo 'node.conn[0].iscsi.DataDigest = CRC32C,None' >> /etc/iscsi/iscsid.conf
fi
- systemctl restart iscsid
# Create /sentinel directory for kured reboot gating (sentinel gate DaemonSet)
- mkdir -p /sentinel
# Create 4Gi swap file for worker node memory pressure relief (NOT for master — etcd is latency-critical)
- fallocate -l 4G /swapfile
- chmod 600 /swapfile
- mkswap /swapfile
- swapon /swapfile
- echo '/swapfile none swap sw 0 0' >> /etc/fstab
- sysctl -w vm.swappiness=10
- echo 'vm.swappiness=10' >> /etc/sysctl.d/99-swap.conf
- ${k8s_join_command}
- systemctl enable kubelet
- systemctl start kubelet
%{ endif }
%{ for provision_cmd in provision_cmds ~}
- ${provision_cmd}
%{ endfor ~}