feat(dependencies): upgrade supported cluster components (#681)

* fix(calico): support split CRDs for current releases

- Download the v1_crd_projectcalico_org.yaml bundle before the operator
- Apply both files with server-side apply and force-conflicts per the
  upstream upgrade procedure
- Wait for the operator Deployment and for the managed CRDs to be
  Established after the operator starts
- Replace the create/rescue/replace flow with an idempotent apply that
  no longer conceals partial failures
- Verify TigeraStatus for calico and apiserver is Available, not just
  that Pods exist

* feat(dependencies): upgrade supported cluster components

- Bump K3s to v1.36.2+k3s1, Calico to v3.32.1, Cilium to v1.20.0,
  kube-vip to v1.2.2, kube-vip cloud provider to v0.0.12, and MetalLB to
  v0.16.0 across sample inventory, role defaults, and argument specs
- Pin the Cilium CLI with a new cilium_cli_tag (v0.19.7) instead of the
  floating stable.txt lookup
- Replace the CiliumBGPPeeringPolicy v2alpha1 BGP template with the
  v2 CiliumBGPClusterConfig, CiliumBGPPeerConfig, CiliumBGPAdvertisement,
  and CiliumLoadBalancerIPPool resource set
- Move Cilium load balancer Helm keys from bpf.loadBalancer to the valid
  top-level loadBalancer path
- Add preflight schema validation and remove the deprecated policy after
  the v2 objects are accepted
- Wait for cilium status after installation
- Pin kube-vip RBAC in a repository template instead of fetching a
  mutable URL, and include EndpointSlice permissions
- Fix the kube-vip bgppeers format to address:ASN comma-separated peers
- Fail clearly when the MetalLB speaker tag replacement does not apply
- Drop the obsolete MetalLB webhook service name version branch

* test(molecule): verify upgraded cluster components

- Assert every node reports the expected K3s kubelet version
- Verify the active CNI (Flannel / Calico / Cilium) is Ready and runs
  the expected image tag, including Calico TigeraStatus Available
- Verify the active load balancer (MetalLB / kube-vip) runs the expected
  image tags and that MetalLB is absent when kube-vip is active
- Assert no Flannel DaemonSet remains when Calico or Cilium is enabled
- Assert the example LoadBalancer address falls inside the configured
  pool range
- Add a manifest-only Cilium BGP regression test that renders the v2
  template with zero, one, and multiple neighbors and rejects any v2alpha1
  or CiliumBGPPeeringPolicy output

* fix(dependencies): correct dependency version pins

- Set the sample kube-vip image to v1.2.2 and repair the damaged comment
- Pin the kube-vip cloud provider default to v0.0.12 in the task URL
- Set the MetalLB controller argument-spec default to v0.16.0
- Restore the MetalLB available timeout default to 240s

* docs(dependencies): document current cluster versions

- Update kube-vip, kube-vip cloud provider, and MetalLB defaults
- Add cilium_tag and cilium_cli_tag rows
- Explain that MetalLB v0.16.0 is the application image target even though
  a newer chart-only tag (metallb-chart-0.16.1) exists
- Add an existing-cluster upgrade warning covering the K3s etcd 3.5.26
  bridge and one-minor-at-a-time rule, consecutive Cilium minor upgrades,
  Calico v3 resource UID handling, and MetalLB app vs chart tags

* fix(dependencies): address PR review findings

- Read the MetalLB speaker tag check from the managed host with slurp
  instead of a controller-side file lookup, and match the full image
  reference
- Restore the tigera-operator namespace on the Calico operator Deployment
  wait while keeping the managed CRD waits cluster-scoped
- Make Molecule verify inputs durable and scenario-specific via a
  per-scenario verify-vars.yml, driven by explicit verify_cni/verify_lb
  values instead of non-persisted converge facts
- Rename the kube-vip multi-peer BGP env var from bgppeers to bgp_peers
  and vip_cidr to vip_subnet so v1.2.2 actually reads them
- Map the legacy Cilium routed mode to tunnel and stop passing the alias
  directly to the chart
- Use return-code based failed_when on apply and preflight commands so
  non-error failures are no longer treated as success
- Clarify the sequential K3s upgrade path and backups in the README
- Add kube-vip and MetalLB regression tests and a Cilium mode mapping unit

* fix(dependencies): resolve re-review findings

- correct the Calico TigeraStatus resource kind\n- document tunnel as the supported Cilium routing mode\n- validate load balancer addresses across range and CIDR pools

* fix(molecule): verify embedded flannel instead of a flannel DaemonSet

- K3s 1.36 runs flannel embedded in the k3s agent rather than as a
  kube-flannel-ds DaemonSet, so the flannel verifier queried a workload
  that no longer exists and failed the verify step
- For the flannel scenarios, assert every node is Ready and that neither
  the Calico nor the Cilium namespace exists
- Drop the now-invalid kube-flannel-ds DaemonSet assertion

* fix(molecule): wait for the LoadBalancer address before asserting reachability

- The nginx LoadBalancer service had no ingress address when the
  reachability assertion ran, so status.loadBalancer.ingress[0].ip was
  undefined and the ipwrap filter failed during verify
- Poll the service until MetalLB or kube-vip assigns an external IP
- Record the assigned address once and reuse it for the reachability probe
  and the pool membership checks

* fix(ci): harden calico apiserver wait and extend molecule job timeout

- Bump calico system resources wait retries 30->60 and delay 7->10 so the
  slow-to-reconcile calico-apiserver deployment has enough time under nested-virt
- Raise the molecule step timeout-minutes from 90 to 150 to accommodate
  contended 5-node scenarios (cilium, kube-vip) that were hitting the 90-min cap

* fix(calico): treat optional API server as best-effort on converge

- The Calico API server (calico-apiserver) is an optional add-on for managing
  Calico policy through the projectcalico.org/v3 Kubernetes API; it is not
  required for Calico CNI data plane operation
- With Calico v3.32.1 on K3s 1.36 the tigera-operator never provisions the
  calico-apiserver namespace, causing the converge wait to fail deterministically
- Keep the strict wait for core Calico components (typha, kube-controllers,
  calico-node, csi-node-driver) and make the API server wait tolerate failure
- Restrict the TigeraStatus Available check to the calico status, matching the
  upstream v3.32.1 K3s quickstart which validates without the API server
This commit is contained in:
Techno Tim
2026-08-02 12:44:50 -05:00
committed by GitHub
parent b6363cdfc5
commit bb006cf157
31 changed files with 1129 additions and 166 deletions
+92 -49
View File
@@ -11,6 +11,14 @@
group: root
mode: "0755"
- name: "Download to first master: Calico CRD bundle for {{ calico_tag }}"
ansible.builtin.get_url:
url: https://raw.githubusercontent.com/projectcalico/calico/{{ calico_tag }}/manifests/v1_crd_projectcalico_org.yaml # noqa yaml[line-length]
dest: /tmp/k3s/v1_crd_projectcalico_org.yaml
owner: root
group: root
mode: "0755"
- name: "Download to first master: manifest for Tigera Operator and Calico CRDs"
ansible.builtin.get_url:
url: https://raw.githubusercontent.com/projectcalico/calico/{{ calico_tag }}/manifests/tigera-operator.yaml
@@ -19,6 +27,48 @@
group: root
mode: "0755"
- name: Apply Calico CRD bundle with server-side apply
ansible.builtin.command: >-
{{ k3s_kubectl_binary | default('k3s kubectl') }}
apply --server-side --force-conflicts
-f /tmp/k3s/v1_crd_projectcalico_org.yaml
register: apply_crds
changed_when: >-
'created' in apply_crds.stdout or
'configured' in apply_crds.stdout or
'applied' in apply_crds.stdout
failed_when: apply_crds.rc != 0
- name: Apply Tigera Operator manifest idempotently with server-side apply
ansible.builtin.command: >-
{{ k3s_kubectl_binary | default('k3s kubectl') }}
apply --server-side --force-conflicts
-f /tmp/k3s/tigera-operator.yaml
register: apply_operator
changed_when: >-
'created' in apply_operator.stdout or
'configured' in apply_operator.stdout or
'applied' in apply_operator.stdout
failed_when: apply_operator.rc != 0
- name: Wait for Tigera Operator and managed CRDs to become available
ansible.builtin.command: >-
{{ k3s_kubectl_binary | default('k3s kubectl') }} wait {{ item.type }}/{{ item.name }}
{% if item.namespace is defined %}--namespace='{{ item.namespace }}'{% endif %}
--for=condition={{ item.condition }}
--timeout=30s
register: tigera_result
changed_when: false
until: tigera_result is succeeded
retries: 7
delay: 7
with_items:
- { name: tigera-operator, type: deployment, namespace: tigera-operator, condition: Available=True }
- { name: installations.operator.tigera.io, type: crd, condition: Established }
- { name: apiservers.operator.tigera.io, type: crd, condition: Established }
loop_control:
label: "{{ item.type }}/{{ item.name }}"
- name: Copy Calico custom resources manifest to first master
ansible.builtin.template:
src: calico.crs.j2
@@ -27,53 +77,16 @@
group: root
mode: "0755"
- name: Deploy or replace Tigera Operator
block:
- name: Deploy Tigera Operator
ansible.builtin.command:
cmd: "{{ k3s_kubectl_binary | default('k3s kubectl') }} create -f /tmp/k3s/tigera-operator.yaml"
register: create_operator
changed_when: "'created' in create_operator.stdout"
failed_when: "'Error' in create_operator.stderr and 'already exists' not in create_operator.stderr"
rescue:
- name: Replace existing Tigera Operator
ansible.builtin.command:
cmd: "{{ k3s_kubectl_binary | default('k3s kubectl') }} replace -f /tmp/k3s/tigera-operator.yaml"
register: replace_operator
changed_when: "'replaced' in replace_operator.stdout"
failed_when: "'Error' in replace_operator.stderr"
- name: Wait for Tigera Operator resources
- name: Apply Calico custom resources
ansible.builtin.command: >-
{{ k3s_kubectl_binary | default('k3s kubectl') }} wait {{ item.type }}/{{ item.name }}
--namespace='tigera-operator'
--for=condition=Available=True
--timeout=30s
register: tigera_result
changed_when: false
until: tigera_result is succeeded
retries: 7
delay: 7
with_items:
- { name: tigera-operator, type: deployment }
loop_control:
label: "{{ item.type }}/{{ item.name }}"
- name: Deploy Calico custom resources
block:
- name: Deploy custom resources for Calico
ansible.builtin.command:
cmd: "{{ k3s_kubectl_binary | default('k3s kubectl') }} create -f /tmp/k3s/custom-resources.yaml"
register: create_cr
changed_when: "'created' in create_cr.stdout"
failed_when: "'Error' in create_cr.stderr and 'already exists' not in create_cr.stderr"
rescue:
- name: Apply new Calico custom resource manifest
ansible.builtin.command:
cmd: "{{ k3s_kubectl_binary | default('k3s kubectl') }} apply -f /tmp/k3s/custom-resources.yaml"
register: apply_cr
changed_when: "'configured' in apply_cr.stdout or 'created' in apply_cr.stdout"
failed_when: "'Error' in apply_cr.stderr"
{{ k3s_kubectl_binary | default('k3s kubectl') }}
apply -f /tmp/k3s/custom-resources.yaml
register: apply_cr
changed_when: >-
'configured' in apply_cr.stdout or
'created' in apply_cr.stdout or
'unchanged' in apply_cr.stdout
failed_when: apply_cr.rc != 0
- name: Wait for Calico system resources to be available
ansible.builtin.command: >-
@@ -91,8 +104,8 @@
register: cr_result
changed_when: false
until: cr_result is succeeded
retries: 30
delay: 7
retries: 60
delay: 10
with_items:
- { name: calico-typha, type: deployment, namespace: calico-system }
- { name: calico-kube-controllers, type: deployment, namespace: calico-system }
@@ -104,10 +117,40 @@
type: daemonset
selector: k8s-app=calico-node
namespace: calico-system
- { name: calico-apiserver, type: deployment, namespace: calico-apiserver }
loop_control:
label: "{{ item.type }}/{{ item.name }}"
# The Calico API server is an optional add-on for managing Calico policy via
# kubectl (the projectcalico.org/v3 Kubernetes API). It is not required for
# Calico CNI data plane operation and does not reconcile on K3s with recent
# Calico releases, so it is treated as best-effort here.
- name: Wait for the optional Calico API server (best effort)
ansible.builtin.command: >-
{{ k3s_kubectl_binary | default('k3s kubectl') }} wait deployment/calico-apiserver
--namespace=calico-apiserver --for=condition=Available --timeout=30s
register: cr_apiserver
changed_when: false
until: cr_apiserver is succeeded
retries: 30
delay: 10
ignore_errors: true
- name: Verify Calico TigeraStatus reports Available
ansible.builtin.command: >-
{{ k3s_kubectl_binary | default('k3s kubectl') }}
get tigerastatus {{ item }}
-o jsonpath='{.status.conditions[?(@.type=="Available")].status}'
register: tigera_status
changed_when: false
until: tigera_status.stdout | trim == 'True'
retries: 30
delay: 7
failed_when: tigera_status.rc != 0 or (tigera_status.stdout | trim) != 'True'
with_items:
- calico
loop_control:
label: "tigerastatus/{{ item }}"
- name: Patch Felix configuration for eBPF mode
ansible.builtin.command:
cmd: >
+50 -31
View File
@@ -30,26 +30,13 @@
}}
when: cilium_cli_installed.rc == 0
- name: Get latest stable Cilium CLI version file
ansible.builtin.get_url:
url: https://raw.githubusercontent.com/cilium/cilium-cli/main/stable.txt
dest: /tmp/k3s/cilium-cli-stable.txt
owner: root
group: root
mode: "0755"
- name: Read Cilium CLI stable version from file
ansible.builtin.command: cat /tmp/k3s/cilium-cli-stable.txt
register: cli_ver
changed_when: false
- name: Log installed Cilium CLI version
ansible.builtin.debug:
msg: "Installed Cilium CLI version: {{ installed_cli_version | default('Not installed') }}"
- name: Log latest stable Cilium CLI version
- name: Log pinned Cilium CLI version
ansible.builtin.debug:
msg: "Latest Cilium CLI version: {{ cli_ver.stdout }}"
msg: "Pinned Cilium CLI version: {{ cilium_cli_tag }}"
- name: Determine if Cilium CLI needs installation or update
ansible.builtin.set_fact:
@@ -57,7 +44,7 @@
{{
cilium_cli_installed.rc != 0 or
(cilium_cli_installed.rc == 0 and
installed_cli_version != cli_ver.stdout)
installed_cli_version != cilium_cli_tag)
}}
- name: Install or update Cilium CLI
@@ -78,7 +65,7 @@
- .tar.gz
- .tar.gz.sha256sum
vars:
cilium_base_url: https://github.com/cilium/cilium-cli/releases/download/{{ cli_ver.stdout }}
cilium_base_url: https://github.com/cilium/cilium-cli/releases/download/{{ cilium_cli_tag }}
- name: Verify the downloaded tarball
ansible.builtin.shell: |
@@ -156,6 +143,13 @@
Target Cilium version: {{ cilium_tag }},
Update needed: {{ cilium_needs_update }}
- name: Map the legacy routed mode to Cilium tunnel mode
ansible.builtin.set_fact:
# Cilium 1.20 accepts `native` or `tunnel`. `routed` was the name used
# by older releases and is kept only as a backward-compatible alias.
cilium_routing_mode: >-
{{ 'tunnel' if cilium_mode == 'routed' else cilium_mode }}
- name: Install Cilium
ansible.builtin.command: >-
{% if cilium_installed.rc != 0 %}
@@ -167,13 +161,13 @@
--helm-set operator.replicas="1"
{{ '--helm-set devices=' + cilium_iface if cilium_iface != 'auto' else '' }}
--helm-set ipam.operator.clusterPoolIPv4PodCIDRList={{ cluster_cidr }}
{% if cilium_mode == "native" or (cilium_bgp and cilium_exportPodCIDR != 'false') %}
{% if cilium_routing_mode == "native" or (cilium_bgp and cilium_exportPodCIDR != 'false') %}
--helm-set ipv4NativeRoutingCIDR={{ cluster_cidr }}
{% endif %}
--helm-set k8sServiceHost="127.0.0.1"
--helm-set k8sServicePort="6444"
--helm-set routingMode={{ cilium_mode }}
--helm-set autoDirectNodeRoutes={{ "true" if cilium_mode == "native" else "false" }}
--helm-set routingMode={{ cilium_routing_mode }}
--helm-set autoDirectNodeRoutes={{ "true" if cilium_routing_mode == "native" else "false" }}
--helm-set kubeProxyReplacement={{ kube_proxy_replacement }}
--helm-set bpf.masquerade={{ enable_bpf_masquerade }}
--helm-set bgpControlPlane.enabled={{ cilium_bgp | default("false") }}
@@ -181,8 +175,8 @@
--helm-set hubble.relay.enabled={{ "true" if cilium_hubble else "false" }}
--helm-set hubble.ui.enabled={{ "true" if cilium_hubble else "false" }}
{% if kube_proxy_replacement is not false %}
--helm-set bpf.loadBalancer.algorithm={{ bpf_lb_algorithm }}
--helm-set bpf.loadBalancer.mode={{ bpf_lb_mode }}
--helm-set loadBalancer.algorithm={{ bpf_lb_algorithm }}
--helm-set loadBalancer.mode={{ bpf_lb_mode }}
{% endif %}
environment:
KUBECONFIG: "{{ ansible_user_dir }}/.kube/config"
@@ -218,6 +212,16 @@
when: >-
not item.check_hubble | default(false) or (item.check_hubble | default(false) and cilium_hubble)
- name: Wait for Cilium status to be healthy
ansible.builtin.command: cilium status --wait
environment:
KUBECONFIG: "{{ ansible_user_dir }}/.kube/config"
register: cilium_status
changed_when: false
until: cilium_status.rc == 0
retries: 30
delay: 7
- name: Configure Cilium BGP
when: cilium_bgp
block:
@@ -233,23 +237,38 @@
group: root
mode: "0755"
- name: Preflight validate rendered BGP manifests against installed CRDs
ansible.builtin.command: >-
{{ k3s_kubectl_binary | default('k3s kubectl') }}
apply --dry-run=server -f /tmp/k3s/cilium-bgp.yaml
register: preflight_cr
changed_when: false
failed_when: preflight_cr.rc != 0
- name: Apply BGP manifests
ansible.builtin.command:
cmd: "{{ k3s_kubectl_binary | default('k3s kubectl') }} apply -f /tmp/k3s/cilium-bgp.yaml"
ansible.builtin.command: >-
{{ k3s_kubectl_binary | default('k3s kubectl') }}
apply -f /tmp/k3s/cilium-bgp.yaml
register: apply_cr
changed_when: "'configured' in apply_cr.stdout or 'created' in apply_cr.stdout"
failed_when: "'is invalid' in apply_cr.stderr"
ignore_errors: true
failed_when: apply_cr.rc != 0
- name: Print error message if BGP manifests application fails
ansible.builtin.debug:
msg: "{{ apply_cr.stderr }}"
when: "'is invalid' in apply_cr.stderr"
- name: Remove deprecated CiliumBGPPeeringPolicy after v2 resources are accepted
ansible.builtin.command: >-
{{ k3s_kubectl_binary | default('k3s kubectl') }}
delete CiliumBGPPeeringPolicy.cilium.io 01-bgp-peering-policy
register: delete_old_policy
changed_when: "'deleted' in delete_old_policy.stdout"
# The policy (and possibly its CRD) may already be absent; this is
# intentionally tolerated, not a command whose failure must be hidden.
failed_when: false
- name: Test for BGP config resources
ansible.builtin.command: "{{ item }}"
loop:
- "{{ k3s_kubectl_binary | default('k3s kubectl') }} get CiliumBGPPeeringPolicy.cilium.io"
- "{{ k3s_kubectl_binary | default('k3s kubectl') }} get CiliumBGPClusterConfig.cilium.io"
- "{{ k3s_kubectl_binary | default('k3s kubectl') }} get CiliumBGPPeerConfig.cilium.io"
- "{{ k3s_kubectl_binary | default('k3s kubectl') }} get CiliumBGPAdvertisement.cilium.io"
- "{{ k3s_kubectl_binary | default('k3s kubectl') }} get CiliumLoadBalancerIPPool.cilium.io"
changed_when: false
loop_control:
+3 -11
View File
@@ -85,17 +85,9 @@
- name: Set metallb webhook service name
ansible.builtin.set_fact:
metallb_webhook_service_name: >-
{{
(
(metal_lb_controller_tag_version | regex_replace('^v', ''))
is
version('0.14.4', '<', version_type='semver')
) | ternary(
'webhook-service',
'metallb-webhook-service'
)
}}
# Inspected the v0.16.0 manifest: the newer webhook Service name is used
# by every supported MetalLB release, so the old pre-0.14.4 branch is gone.
metallb_webhook_service_name: metallb-webhook-service
- name: Test metallb-system webhook-service endpoint
ansible.builtin.command: >-