# Episode 37: kubernetes Topology Aware routing and rke2 with cilium
**Join us on [YouTube](https://youtu.be/P7WKlhyoyMA) on Friday 4th Febuary, 7pm GMT**
### Kubecon EU!!
[kubecon eu ebpf day!](https://twitter.com/mauilion/status/1494741512696975361?s=20&t=tu-3KTij6iXiVNMopqlFPg)
### Resources:
[ep 35 notes](https://github.com/isovalent/eCHO/tree/main/episodes/035)
[rke2 network options](https://docs.rke2.io/install/network_options/)
`Vagrantfile`
```yaml
# Vagrantfile API/syntax version. Don't touch unless you know what you're doing!
VAGRANTFILE_API_VERSION = "2"
# Require a recent version of vagrant otherwise some have reported errors
# setting host names on boxes
Vagrant.require_version ">= 1.7.2"
CONFIG = {
"box" => ENV['box'] || "generic/ubuntu2004",
"domain" => ENV['CLUSTER_DOMAIN'] || "c1.k8s.work",
"network_name" => ENV['CLUSTER_NETWORK'] || "c1",
"network_cidr" => ENV['CLUSTER_CIDR'] || "192.168.201.0/24",
"domain_mac_seed" => ENV['DOMAIN_MAC_SEED'] || "52:54:30:00",
"num_cp_nodes" => ENV['NUM_CP'] || "1",
"cp_cores" => ENV['CP_CORES'] || "8",
"cp_memory" => ENV['CP_MEMORY'] || "16384",
"cp_disk" => ENV['CP_DISK'] || "100",
"cp_mac" => ENV['CP_MAC'] || ":02",
"num_worker_nodes" => ENV['NUM_WORKER'] || "3",
"worker_cores" => ENV['WORKER_CORES'] || "8",
"worker_memory" => ENV['WORKER_MEMORY'] || "16384",
"worker_disk" => ENV['WORKER_DISK'] || "100",
"worker_mac" => ENV['WORKER_MAC'] || ":03",
}
hwe_kernel = %{
apt update -qq
apt install -y -qq linux-{image,headers,tools}-generic-hwe-20.04
}
fix_dns = %{
sudo sed -i -e '/nameservers:/d' -e '/addresses:/d' /etc/netplan/01-netcfg.yaml
sudo netplan generate && sudo netplan apply
sudo sed -i 's/^[[:alpha:]]/#&/' /etc/systemd/resolved.conf
sudo systemctl restart systemd-resolved.service
}
server = %{
mkdir -p /etc/rancher/rke2/
echo "
write-kubeconfig-mode: "0644"
tls-san:
- "cp01.c1.k8s.work"
token: vagrant-rke2
cni: none
kube-apiserver-arg:
- "feature-gates=TopologyAwareHints=true,JobTrackingWithFinalizers=true"
kube-controller-manager-arg:
- "feature-gates=TopologyAwareHints=true,JobTrackingWithFinalizers=true"
disable-kube-proxy: true
disable:
- rke2-kube-proxy
" >> /etc/rancher/rke2/config.yaml
curl -sfL https://get.rke2.io | INSTALL_RKE2_VERSION=v1.23.3+rke2r1 sh -
systemctl enable --now rke2-server.service
}
agent = %{
mkdir -p /etc/rancher/rke2/
echo "
server: https://cp01.c1.k8s.work:9345
token: vagrant-rke2
" >> /etc/rancher/rke2/config.yaml
curl -sfL https://get.rke2.io | INSTALL_RKE2_VERSION=v1.23.3+rke2r1 sh -
systemctl enable --now rke2-agent.service
}
c1_xml = %{
<network xmlns:dnsmasq='http://libvirt.org/schemas/network/dnsmasq/1.0' ipv6='yes'>
<name>c1</name>
<forward mode='nat'>
<nat>
<port start='1024' end='65535'/>
</nat>
</forward>
<bridge name='br-c1' stp='on' delay='0'/>
<domain name='c1.k8s.work'/>
<dns>
<host ip='192.168.201.1'>
<hostname>api-int.c1.k8s.work</hostname>
<hostname>api.c1.k8s.work</hostname>
</host>
</dns>
<ip address='192.168.201.1' netmask='255.255.255.0' localPtr='yes'>
<dhcp>
<range start='192.168.201.2' end='192.168.201.9'/>
<host mac='52:54:30:00:02:01' name='cp01.c1.k8s.work' ip='192.168.201.21'/>
<host mac='52:54:30:00:03:01' name='worker01.c1.k8s.work' ip='192.168.201.31'/>
<host mac='52:54:30:00:03:02' name='worker02.c1.k8s.work' ip='192.168.201.32'/>
<host mac='52:54:30:00:03:03' name='worker03.c1.k8s.work' ip='192.168.201.33'/>
<bootp file='pxelinux.0' server='192.168.201.1'/>
</dhcp>
</ip>
<dnsmasq:options>
<dnsmasq:option value='cname=*.apps.c1.k8s.work,apps.c1.k8s.work,api.c1.k8s.work'/>
<dnsmasq:option value='auth-zone=c1.k8s.work'/>
<dnsmasq:option value='auth-server=c1.k8s.work,*'/>
</dnsmasq:options>
</network>
}
Vagrant.configure(VAGRANTFILE_API_VERSION) do |config|
config.ssh.insert_key = false
config.vm.box = "#{CONFIG['box']}"
config.vm.provision "shell", name: "fix_dns", run: "once", inline: fix_dns
# config.vm.provision "shell", name: "hwe_kernel", run: "once", privileged: true, reboot: true, inline: hwe_kernel
config.trigger.before :up do |trigger|
trigger.name = "vmnet"
trigger.info = "defining management network"
trigger.ruby do |env|
require 'libvirt'
conn = Libvirt::open('qemu:///system')
begin
net = conn.define_network_xml(c1_xml)
net.create
rescue Libvirt::DefinitionError
p "network already defined"
end
conn.close
end
end
CP = (CONFIG['num_cp_nodes']).to_i
(1..CP).each do |i|
vm_name = "cp0#{i}"
vm_fqdn = "#{vm_name}.#{CONFIG['domain']}"
vm_cpu = CONFIG['cp_cores']
vm_cidr = CONFIG['network_cidr']
vm_memory = CONFIG['cp_memory']
vm_disk = CONFIG['cp_disk']
vm_mac = "#{CONFIG['domain_mac_seed']}#{CONFIG['cp_mac']}:0#{i}"
config.vm.define vm_name do |node|
node.vm.hostname = "#{vm_fqdn}"
node.vm.provision "shell", name: "rke2-server", run: "once", privileged: true, inline: server
node.vm.provider :libvirt do |domain|
domain.cpus = "#{vm_cpu}".to_i
domain.driver = 'kvm'
domain.machine_virtual_size = "#{vm_disk}".to_i
domain.management_network_mac = vm_mac
domain.management_network_name = CONFIG['network_name']
domain.management_network_address = CONFIG['network_cidr']
domain.memory = "#{vm_memory}".to_i
domain.uri = 'qemu+unix:///system'
end
end
end
WORKER = (CONFIG['num_worker_nodes']).to_i
(1..WORKER).each do |i|
vm_name = "worker0#{i}"
vm_fqdn = "#{vm_name}.#{CONFIG['domain']}"
vm_cpu = CONFIG['worker_cores']
vm_memory = CONFIG['worker_memory']
vm_disk = CONFIG['worker_disk']
vm_mac = "#{CONFIG['domain_mac_seed']}#{CONFIG['worker_mac']}:0#{i}"
config.vm.define vm_name, autostart: true do |node|
node.vm.hostname = "#{vm_fqdn}"
node.vm.provision "shell", name: "rke2-agent", run: "once", privileged: true, inline: agent
node.vm.provider :libvirt do |domain|
domain.cpus = "#{vm_cpu}".to_i
domain.driver = 'kvm'
domain.machine_virtual_size = "#{vm_disk}".to_i
domain.management_network_mac = vm_mac
domain.management_network_name = CONFIG['network_name']
domain.management_network_address = CONFIG['network_cidr']
domain.memory = "#{vm_memory}".to_i
domain.uri = 'qemu+unix:///system'
end
end
end
end
```
Sign in with Wallet
Connect another wallet