WIP to be squashed properly

enix · Dec 14, 2023 · cd3fec6 · cd3fec6
1 parent 8181515
commit cd3fec6
Show file tree

Hide file tree

Showing 14 changed files with 88 additions and 112 deletions.
diff --git a/docs/guides/join_nodes.md b/docs/guides/join_nodes.md
@@ -27,44 +27,30 @@ node-3
 If you don't have provision a load-balancer and require the local haproxy to be deployed:
 
 ```
-ansible-playbook -i inventory enix.kubeadm.00_apiserver_proxy -l kube_control_plane:nodes-3
+ansible-playbook -i inventory enix.kubeadm.00_apiserver_proxy -e limit=nodes-3
 ```
-You can skip the `-l` argument, if you're cluster doesn't have pending change you want to preserve on other nodes.
-Don't forget to put all control_plane or it will fail to provision the apiserver proxy
+You need to specify the `limit` variable via "extra-vars", because `-l` cannot really work in the context of ansible-kubeadm
+(you need to connect to all the masters to get the IP needed to configure the loadbalancer)
 
+### Joining nodes
 
-### Create bootstrap-token
-
-Then create a bootstrap token by adding using the `bootstrap_token` tag.
-Don't use a limit that skip control plane nodes.
+You can join a node and skip other changes on other nodes by specify the limit variable.
 
 ```
-ansible-playbook -i inventory.cfg enix.kubeadm.01_site -t bootstrap_token
+ansible-play -i inventory.cfg enix.kubeadm.01_site -e limit=nodes-3
 ```
 
-No need to retrieve it by yourself, it will be discovered when joining the node
-The token has a validity of 1H, so you don't need to repeat this step each time you try to join nodes
-
-### Joining nodes
-
-You can join a node and skip other changes to the cluster by using the `join` tag.
-With the tag, you can limit to hosts you want to join.
 
-```
-ansible-play -i inventory.cfg enix.kubeadm.01_site -t join -l nodes-3
-```
 
-## Alternative method
+### Create bootstrap-token
 
-You can merge the creation of the boostrap token with the joining of the action of join:
+Then create a bootstrap token by adding using the `bootstrap_token` tag.
+Don't use a limit that skip control plane nodes.
 
 ```
-ansible-playbook -i inventory.cfg enix.kubeadm.01_site -t bootstap_token,join -l kube_control_plane:node-3
+ansible-playbook -i inventory.cfg enix.kubeadm.01_site -t bootstrap_token
 ```
 
-Please note that you need to include a least one control plane node in the limit host pattern,
-You can also skip the limit host pattern to apply to all nodes as those step are indempotent on their own: it will not mess with the current nodes.
-
-# To join control-plane nodes
+No need to retrieve it by yourself, it will be discovered when joining the node
+The token has a validity of 1H, so you don't need to repeat this step each time you try to join nodes
 
-There is no tag for this operation, you need to apply the entire playbook for this
diff --git a/playbooks/00_apiserver_proxy.yml b/playbooks/00_apiserver_proxy.yml
@@ -10,15 +10,10 @@
   any_errors_fatal: '{{ any_errors_fatal|default(true) }}'
   vars:
     _control_plane: true
-  pre_tasks:
-    - name: 'Fail if not all master in the specified limit'
-      fail:
-        msg: 'Not all control_plane provided, ajust --limit to provid all control_plane'
-      when: groups[kube_cp_group|default("kube_control_plane")]|difference(ansible_play_hosts)|length > 0
   roles:
     - role: find_ip
 
-- hosts: '{{ kube_cp_group|default("kube_control_plane") }}:{{ kube_worker_group|default("kube_workers") }}'
+- hosts: '{{ kube_cp_group|default("kube_control_plane") }}:{{ kube_worker_group|default("kube_workers") }}{{ ":" ~ limit if limit is defined else "" }}'
   any_errors_fatal: '{{ any_errors_fatal|default(true) }}'
   pre_tasks:
     - include_role:
@@ -36,7 +31,7 @@
       vars:
         kubeadm_hook_list: ['post_apiserver_proxy']
 
-- hosts: 'haproxy_upgrade_group:&{{ kube_cp_group|default("kube_control_plane") }}'
+- hosts: 'haproxy_upgrade_group:&{{ kube_cp_group|default("kube_control_plane") }}{{ ":" ~ limit if limit is defined else "" }}'
   serial: '{{ upgrade_cp_serial|default(1) }}'
   any_errors_fatal: '{{ any_errors_fatal|default(true) }}'
   pre_tasks:
@@ -52,7 +47,7 @@
       vars:
         kubeadm_hook_list: ['post_proxy_upgrade_haproxy']
 
-- hosts: 'haproxy_upgrade_group:&{{ kube_worker_group|default("kube_workers") }}'
+- hosts: 'haproxy_upgrade_group:&{{ kube_worker_group|default("kube_workers") }}{{ ":" ~ limit if limit is defined else "" }}'
   serial: '{{ upgrade_worker_serial|default(1) }}'
   any_errors_fatal: '{{ any_errors_fatal|default(true) }}'
   pre_tasks:

diff --git a/playbooks/01_site.yml b/playbooks/01_site.yml
@@ -25,7 +25,7 @@
       vars:
         kubeadm_hook_list: ['post_preflight_cp']
 
-- hosts: '{{ kube_cp_group|default("kube_control_plane") }}:{{ kube_worker_group|default("kube_workers") }}'
+- hosts: '{{ kube_cp_group|default("kube_control_plane") }}:{{ kube_worker_group|default("kube_workers") }}{{ ":" ~ limit if limit is defined else "" }}'
   any_errors_fatal: '{{ any_errors_fatal|default(true) }}'
   roles:
     - role: find_ip
@@ -42,7 +42,7 @@
   roles:
     - role: process_reasons
 
-- hosts: '{{ kube_cp_group|default("kube_control_plane") }}'
+- hosts: '{{ kube_cp_group|default("kube_control_plane") }}{{ ":" ~ limit if limit is defined else "" }}{{ ":" ~ limit if limit is defined else "" }}'
   any_errors_fatal: '{{ any_errors_fatal|default(true) }}'
   gather_facts: false
   roles:
@@ -92,7 +92,7 @@
         kubeadm_hook_list: ['post_config_update']
 
 # This has to be overly cautious on package upgade
-- hosts: cp_upgrade
+- hosts: 'cp_upgrade{{ ":" ~ limit if limit is defined else "" }}'
   any_errors_fatal: '{{ any_errors_fatal|default(true) }}'
   gather_facts: false
   pre_tasks:
@@ -118,7 +118,7 @@
 
 # Upgrade conrol-plane nodes
 - name: 'Upgrade to control plane nodes'
-  hosts: '{{ kube_cp_group|default("kube_control_plane") }}:&nodes_upgrade'
+  hosts: '{{ kube_cp_group|default("kube_control_plane") }}:&nodes_upgrade{{ ":" ~ limit if limit is defined else "" }}'
   any_errors_fatal: '{{ any_errors_fatal|default(true) }}'
   serial: '{{ upgrade_cp_serial|default(1) }}'
   gather_facts: false
@@ -147,7 +147,7 @@
 
 # Upgrade worker nodes
 - name: 'Upgrade to workers nodes'
-  hosts: '{{ kube_worker_group|default("kube_workers") }}:&nodes_upgrade'
+  hosts: '{{ kube_worker_group|default("kube_workers") }}:&nodes_upgrade{{ ":" ~ limit if limit is defined else "" }}'
   any_errors_fatal: '{{ any_errors_fatal|default(true) }}'
   serial: '{{ upgrade_worker_serial|default(1) }}'
   gather_facts: false
@@ -174,7 +174,7 @@
 
 # Join control-plane nodes
 - name: 'Join new control plane nodes'
-  hosts: '{{ kube_cp_group|default("kube_control_plane") }}'
+  hosts: '{{ kube_cp_group|default("kube_control_plane") }}{{ ":" ~ limit if limit is defined else "" }}'
   any_errors_fatal: '{{ any_errors_fatal|default(true) }}'
   gather_facts: false
   vars:
@@ -198,7 +198,7 @@
 
 # Join worker nodes
 - name: 'Join new workers nodes'
-  hosts: '{{ kube_worker_group|default("kube_workers") }}'
+  hosts: '{{ kube_worker_group|default("kube_workers") }}{{ ":" ~ limit if limit is defined else "" }}'
   any_errors_fatal: '{{ any_errors_fatal|default(true) }}'
   gather_facts: false
   tags: ['join']
@@ -218,7 +218,7 @@
         kubeadm_hook_list: ['post_workers_join', 'post_nodes_join']
 
 - name: 'Finally executing post_run hook on all hosts'
-  hosts: '{{ kube_cp_group|default("kube_control_plane") }}:{{ kube_worker_group|default("kube_workers") }}'
+  hosts: '{{ kube_cp_group|default("kube_control_plane") }}:{{ kube_worker_group|default("kube_workers") }}{{ ":" ~ limit if limit is defined else "" }}'
   any_errors_fatal: '{{ any_errors_fatal|default(true) }}'
   gather_facts: false
   tasks:

diff --git a/roles/bootstrap_token/tasks/main.yaml b/roles/bootstrap_token/tasks/main.yaml
@@ -4,6 +4,7 @@
     nodes_to_join: >-
       {{ q('inventory_hostnames', kube_cp_group ~ ':' ~ kube_worker_group)
          |map('extract', hostvars)
+         |selectattr('_kubelet_config_stat', 'defined')
          |rejectattr('_kubelet_config_stat.stat.exists')
          |map(attribute='inventory_hostname')|list }}
   run_once: true

diff --git a/roles/common_vars/defaults/main.yml b/roles/common_vars/defaults/main.yml
@@ -6,3 +6,6 @@ kube_cp_group: kube_control_plane
 kube_worker_group: kube_workers
 
 cp_node: '{{ (groups.cp_running|default(groups[kube_cp_group]))|first }}'
+
+_target_kube_version: '{{ hostvars[cp_node]._target_kube_version }}'
+_target_kubeadm_version: '{{ hostvars[cp_node]._target_kubeadm_version }}'
diff --git a/roles/packages/meta/main.yml b/roles/packages/meta/main.yml
@@ -1,19 +1,4 @@
 ---
 dependencies:
   - role: packages_common
-galaxy_info:
-  author: Julien Girardin
-  description: Install kubernetes related packages
-  company: Enix
-  license: Apache
-  min_ansible_version: 2.7
-  platforms:
-    - name: Ubuntu
-      versions:
-        - 18.04
-        - 20.04
-  galaxy_tags:
-    - kubernetes
-    - kubeadm
-    - kubelet
-    - kubectl
+  - role: common_vars
diff --git a/roles/upload_certs/tasks/main.yml b/roles/upload_certs/tasks/main.yml
@@ -15,7 +15,7 @@
     kubeadm init phase upload-certs
     --upload-certs
     --certificate-key {{ cert_encryption_key }}
-  no_log: '{{ sensitive_debug|bool }}'
+  no_log: '{{ not sensitive_debug|bool }}'
   run_once: true
   delegate_to: '{{ cp_node }}'
   when: _cp_to_join|length > 0
diff --git a/tests/conftest.py b/tests/conftest.py
@@ -95,8 +95,8 @@ def vagrant(tmpdir):
     return LocalVagrant(inventory_dir_copy=tmpdir)
 
 
-@then("Set cluster {variable} = {value}")
-@given("The cluster {variable) = {value}")
+@then(parsers.parse("Set cluster {variable} = {value}"))
+@given(parsers.parse("The cluster {variable} = {value}"))
 def cluster_set_param(provider, variable, value):
     provider.vars[variable] = value
     # Refresh infrastructure
@@ -160,7 +160,7 @@ def ansible_extra_args(request):
 
 @when(
     parsers.re(
-        r"I (?P<dry_run>dry-)?run the playbooks?:?\s+(?P<playbooks>.+?)(?P<with_err>\s+with error:?\s+)?(?(with_err)(?P<error>.+)|\Z)",
+        r"I (?P<dry_run>dry-)?run the playbooks?:?\s+(?P<arguments>.+?)(?P<with_err>\s+with error:?\s+)?(?(with_err)(?P<error>.+)|\Z)",
         re.DOTALL,
     )
 )
@@ -171,26 +171,20 @@ def ansible_playbook(
     galaxy_deps,
     ansible_extra_args,
     results,
-    playbooks,
+    arguments,
     dry_run,
     error,
 ):
     if dry_run == "dry-":
         dry_run = True
     else:
         dry_run = False
-    playbook_list = re.findall(r"[\w./]+", playbooks)
-    if not all(os.path.exists(p) for p in playbook_list):
-        playbook_list_subdir = [os.path.join("playbooks", p) for p in playbook_list]
-        if all(os.path.exists(p) for p in playbook_list_subdir):
-            playbook_list = playbook_list_subdir
-        else:
-            raise ValueError("All playbooks could not be found")
+    argument_list = re.findall(r"[^\s]+", arguments)
     result = run_ansible_playbook(
         virtualenv,
-        playbook_list,
-        ansible_extra_args=ansible_extra_args,
         inventory=inventory,
+        arguments=argument_list,
+        ansible_extra_args=ansible_extra_args,
         dry_run=dry_run,
     )
     if error:
@@ -206,9 +200,9 @@ def ansible_playbook(
 def ansible_kubeadm(inventory, virtualenv, galaxy_deps, ansible_extra_args, results):
     result = run_ansible_playbook(
         virtualenv,
+        inventory,
         ["tests/playbooks/verify.yml"],
         ansible_extra_args=ansible_extra_args,
-        inventory=inventory,
     )
     assert_ansible_error(result)
 

diff --git a/tests/features/haproxy.feature b/tests/features/haproxy.feature
@@ -18,23 +18,23 @@ Feature: Haproxy
             apiserver_proxy_use_docker: true
             kube_version: 1.23
         When I run the playbook tests/playbooks/prepare.yml
-        When I run the playbooks 00_apiserver_proxy.yml
-                                 01_site.yml
+        When I run the playbooks playbooks/00_apiserver_proxy.yml
+                                 playbooks/01_site.yml
         When I run the playbook tests/playbooks/cni.yml
         Then I should have a working cluster
 
 
         When With those group_vars on group all:
             apiserver_proxy_use_docker:
         When I reset tasks counters
-        When I run the playbooks 00_apiserver_proxy.yml
-                                 01_site.yml
+        When I run the playbooks playbooks/00_apiserver_proxy.yml
+                                 playbooks/01_site.yml
              with error:
                As docker has been deprecated
 
         When With those group_vars on group all:
             apiserver_proxy_use_docker: false
         When I reset tasks counters
-        When I run the playbooks 00_apiserver_proxy.yml
-                                 01_site.yml
+        When I run the playbooks playbooks/00_apiserver_proxy.yml
+                                 playbooks/01_site.yml
         Then I should have a working cluster
diff --git a/tests/features/install.feature b/tests/features/install.feature
@@ -17,16 +17,16 @@ Feature: Install
               cgroupDriver: "systemd"
             kube_version: <version>
         When I run the playbook tests/playbooks/prepare.yml
-        When I dry-run the playbooks 00_apiserver_proxy.yml
-                                     01_site.yml
-        When I run the playbooks 00_apiserver_proxy.yml
-                                 01_site.yml
+        When I dry-run the playbooks playbooks/00_apiserver_proxy.yml
+                                     playbooks/01_site.yml
+        When I run the playbooks playbooks/00_apiserver_proxy.yml
+                                 playbooks/01_site.yml
         When I run the playbook tests/playbooks/cni.yml
         Then I should have a working cluster
 
         When I reset tasks counters
-        And  I run the playbooks 00_apiserver_proxy.yml
-                                 01_site.yml
+        And  I run the playbooks playbooks/00_apiserver_proxy.yml
+                                 playbooks/01_site.yml
         Then I should see no orange/yellow changed tasks
 
         Examples:

diff --git a/tests/features/join_nodes.feature b/tests/features/join_nodes.feature
@@ -1,10 +1,10 @@
-Feature: Upgrade
-    A test to upgrade a kubeadm cluster
+Feature: Join Nodes
+    A test to join nodes to a kubeadm cluster
 
-    Scenario: Upgrade via ansible-kubeadm
+    Scenario: Join nodes via ansible-kubeadm
         Given I want ansible 3
         Given The cluster control_plane_count = 1
-        Given The cluster worker_count = 1
+        Given The cluster worker_count = 0
         Given Some running VMs
 
         When With those group_vars on group all:
@@ -18,14 +18,19 @@ Feature: Upgrade
               cgroupDriver: "systemd"
             kube_version: 1.23
         When I run the playbook tests/playbooks/prepare.yml
-        When I run the playbooks 00_apiserver_proxy.yml
-                                 01_site.yml
+        When I run the playbooks playbooks/00_apiserver_proxy.yml
+                                 playbooks/01_site.yml
         When I run the playbook tests/playbooks/cni.yml
+        Then I should have a working cluster
 
-        Then Set cluster worker_count = 2
+        Then Set cluster worker_count = 1
+        When I run the playbook tests/playbooks/prepare.yml
+        When I run the playbooks playbooks/01_site.yml -e "limit=*-node-1"
+        Then I should have a working cluster
 
-        When With those group_vars on group all: kube_version: 1.24
-        When I run the playbooks 00_apiserver_proxy.yml
-                                 01_site.yml
+        Then Set cluster control_plane_count = 2
+        When I run the playbook tests/playbooks/prepare.yml
+        When I run the playbooks playbooks/00_apiserver_proxy.yml
+                                 playbooks/01_site.yml
 
         Then I should have a working cluster
diff --git a/tests/features/upgrade.feature b/tests/features/upgrade.feature
@@ -18,13 +18,13 @@ Feature: Upgrade
             kube_version: <from_version>
             action_reasons_review_skip: true
         When I run the playbook tests/playbooks/prepare.yml
-        When I run the playbooks 00_apiserver_proxy.yml
-                                 01_site.yml
+        When I run the playbooks playbooks/00_apiserver_proxy.yml
+                                 playbooks/01_site.yml
         When I run the playbook tests/playbooks/cni.yml
 
         When With those group_vars on group all: kube_version: <to_version>
-        When I run the playbooks 00_apiserver_proxy.yml
-                                 01_site.yml
+        When I run the playbooks playbooks/00_apiserver_proxy.yml
+                                 playbooks/01_site.yml
 
         Then I should have a working cluster