Add Cilium CNI option (#435)

* Add Cilium CNI option

* Tweak version checks and add BGP resource verify

* Update metallb detection for kube-vip feat compat

.github/workflows/test.yml

@@ -13,6 +13,7 @@ jobs:
           - ipv6
           - single_node
           - calico
+          - cilium
           - kube-vip
       fail-fast: false
     env:

inventory/sample/group_vars/all.yml

@@ -13,9 +13,27 @@ flannel_iface: "eth0"
 # uncomment calico_iface to use tigera operator/calico cni instead of flannel https://docs.tigera.io/calico/latest/about
 # calico_iface: "eth0"
 calico_ebpf: false           # use eBPF dataplane instead of iptables
-calico_cidr: "10.52.0.0/16"  # calico cluster pod cidr pool
 calico_tag: "v3.27.0"        # calico version tag
 
+# uncomment cilium_iface to use cilium cni instead of flannel or calico
+# ensure v4.19.57, v5.1.16, v5.2.0 or more recent kernel
+# cilium_iface: "eth0"
+cilium_mode: "native"        # native when nodes on same subnet or using bgp, else set routed
+cilium_tag: "v1.14.6"        # cilium version tag
+cilium_hubble: true          # enable hubble observability relay and ui
+
+# if using calico or cilium, you may specify the cluster pod cidr pool
+cluster_cidr: "10.52.0.0/16"
+
+# enable cilium bgp control plane for lb services and pod cidrs. disables metallb.
+cilium_bgp: false
+
+# bgp parameters for cilium cni. only active when cilium_iface is defined and cilium_bgp is true.
+cilium_bgp_my_asn: "64513"
+cilium_bgp_peer_asn: "64512"
+cilium_bgp_peer_address: "192.168.30.1"
+cilium_bgp_lb_cidr: "192.168.31.0/24"   # cidr for cilium loadbalancer ipam
+
 # apiserver_endpoint is virtual ip-address which will be configured on each master
 apiserver_endpoint: "192.168.30.222"
 
@@ -26,25 +44,25 @@ k3s_token: "some-SUPER-DEDEUPER-secret-password"
 # The IP on which the node is reachable in the cluster.
 # Here, a sensible default is provided, you can still override
 # it for each of your hosts, though.
-k3s_node_ip: "{{ ansible_facts[(calico_iface | default(flannel_iface))]['ipv4']['address'] }}"
+k3s_node_ip: "{{ ansible_facts[(cilium_iface | default(calico_iface | default(flannel_iface)))]['ipv4']['address'] }}"
 
 # Disable the taint manually by setting: k3s_master_taint = false
 k3s_master_taint: "{{ true if groups['node'] | default([]) | length >= 1 else false }}"
 
 # these arguments are recommended for servers as well as agents:
 extra_args: >-
-  {{ '--flannel-iface=' + flannel_iface if calico_iface is not defined else '' }}
+  {{ '--flannel-iface=' + flannel_iface if calico_iface is not defined and cilium_iface is not defined else '' }}
   --node-ip={{ k3s_node_ip }}
 
 # change these to your liking, the only required are: --disable servicelb, --tls-san {{ apiserver_endpoint }}
-# the contents of the if block is also required if using calico
+# the contents of the if block is also required if using calico or cilium
 extra_server_args: >-
   {{ extra_args }}
   {{ '--node-taint node-role.kubernetes.io/master=true:NoSchedule' if k3s_master_taint else '' }}
-  {% if calico_iface is defined %}
+  {% if calico_iface is defined or cilium_iface is defined %}
   --flannel-backend=none
   --disable-network-policy
-  --cluster-cidr={{ calico_cidr | default('10.52.0.0/16') }}
+  --cluster-cidr={{ cluster_cidr | default('10.52.0.0/16') }}
   {% endif %}
   --tls-san {{ apiserver_endpoint }}
   --disable servicelb

molecule/README.md

@@ -15,6 +15,8 @@ We have these scenarios:
   Very similar to the default scenario, but uses only a single node for all cluster functionality.
 - **calico**:
   The same as single node, but uses calico cni instead of flannel.
+- **cilium**:
+  The same as single node, but uses cilium cni instead of flannel.
 - **kube-vip**
   The same as single node, but uses kube-vip as service loadbalancer instead of MetalLB
 

molecule/cilium/molecule.yml

@@ -0,0 +1,49 @@
+---
+dependency:
+  name: galaxy
+driver:
+  name: vagrant
+platforms:
+  - name: control1
+    box: generic/ubuntu2204
+    memory: 4096
+    cpus: 4
+    config_options:
+      # We currently can not use public-key based authentication on Ubuntu 22.04,
+      # see: https://github.com/chef/bento/issues/1405
+      ssh.username: "vagrant"
+      ssh.password: "vagrant"
+    groups:
+      - k3s_cluster
+      - master
+    interfaces:
+      - network_name: private_network
+        ip: 192.168.30.63
+provisioner:
+  name: ansible
+  env:
+    ANSIBLE_VERBOSITY: 1
+  playbooks:
+    converge: ../resources/converge.yml
+    side_effect: ../resources/reset.yml
+    verify: ../resources/verify.yml
+  inventory:
+    links:
+      group_vars: ../../inventory/sample/group_vars
+scenario:
+  test_sequence:
+    - dependency
+    - cleanup
+    - destroy
+    - syntax
+    - create
+    - prepare
+    - converge
+    # idempotence is not possible with the playbook in its current form.
+    - verify
+    # We are repurposing side_effect here to test the reset playbook.
+    # This is why we do not run it before verify (which tests the cluster),
+    # but after the verify step.
+    - side_effect
+    - cleanup
+    - destroy

molecule/cilium/overrides.yml

@@ -0,0 +1,16 @@
+---
+- name: Apply overrides
+  hosts: all
+  tasks:
+    - name: Override host variables
+      ansible.builtin.set_fact:
+        # See:
+        # https://github.com/flannel-io/flannel/blob/67d603aaf45ef80f5dd39f43714fc5e6f8a637eb/Documentation/troubleshooting.md#Vagrant
+        cilium_iface: eth1
+
+        # The test VMs might be a bit slow, so we give them more time to join the cluster:
+        retry_count: 45
+
+        # Make sure that our IP ranges do not collide with those of the other scenarios
+        apiserver_endpoint: "192.168.30.225"
+        metal_lb_ip_range: "192.168.30.110-192.168.30.119"

roles/k3s_server/tasks/main.yml

@@ -29,7 +29,7 @@
 - name: Deploy metallb manifest
   include_tasks: metallb.yml
   tags: metallb
-  when: kube_vip_lb_ip_range is not defined
+  when: kube_vip_lb_ip_range is not defined and (not cilium_bgp or cilium_iface is not defined)
 
 - name: Deploy kube-vip manifest
   include_tasks: kube-vip.yml