diff options
Diffstat (limited to 'playbooks/common/openshift-cluster/upgrades')
7 files changed, 68 insertions, 34 deletions
diff --git a/playbooks/common/openshift-cluster/upgrades/docker/docker_upgrade.yml b/playbooks/common/openshift-cluster/upgrades/docker/docker_upgrade.yml index 07db071ce..02b8a9d3c 100644 --- a/playbooks/common/openshift-cluster/upgrades/docker/docker_upgrade.yml +++ b/playbooks/common/openshift-cluster/upgrades/docker/docker_upgrade.yml @@ -52,9 +52,13 @@ - name: Drain Node for Kubelet upgrade command: > - {{ openshift.common.admin_binary }} drain {{ openshift.node.nodename }} --force --delete-local-data --ignore-daemonsets + {{ openshift.common.admin_binary }} drain {{ openshift.node.nodename }} --config={{ openshift.common.config_base }}/master/admin.kubeconfig --force --delete-local-data --ignore-daemonsets delegate_to: "{{ groups.oo_first_master.0 }}" when: l_docker_upgrade is defined and l_docker_upgrade | bool and inventory_hostname in groups.oo_nodes_to_upgrade + register: l_docker_upgrade_drain_result + until: not l_docker_upgrade_drain_result | failed + retries: 60 + delay: 60 - include: upgrade.yml when: l_docker_upgrade is defined and l_docker_upgrade | bool diff --git a/playbooks/common/openshift-cluster/upgrades/docker/restart.yml b/playbooks/common/openshift-cluster/upgrades/docker/restart.yml index 1b418920f..13313377e 100644 --- a/playbooks/common/openshift-cluster/upgrades/docker/restart.yml +++ b/playbooks/common/openshift-cluster/upgrades/docker/restart.yml @@ -1,6 +1,10 @@ --- - name: Restart docker service: name=docker state=restarted + register: l_docker_restart_docker_in_upgrade_result + until: not l_docker_restart_docker_in_upgrade_result | failed + retries: 3 + delay: 30 - name: Update docker facts openshift_facts: @@ -24,4 +28,5 @@ state: started delay: 10 port: "{{ openshift.master.api_port }}" + timeout: 600 when: inventory_hostname in groups.oo_masters_to_config diff --git a/playbooks/common/openshift-cluster/upgrades/docker/upgrade.yml b/playbooks/common/openshift-cluster/upgrades/docker/upgrade.yml index 17f8fc6e9..35d000e49 100644 --- a/playbooks/common/openshift-cluster/upgrades/docker/upgrade.yml +++ b/playbooks/common/openshift-cluster/upgrades/docker/upgrade.yml @@ -32,7 +32,13 @@ - debug: var=docker_image_count.stdout when: docker_upgrade_nuke_images is defined and docker_upgrade_nuke_images | bool -- service: name=docker state=stopped +- service: + name: docker + state: stopped + register: l_pb_docker_upgrade_stop_result + until: not l_pb_docker_upgrade_stop_result | failed + retries: 3 + delay: 30 - name: Upgrade Docker package: name=docker{{ '-' + docker_version }} state=present diff --git a/playbooks/common/openshift-cluster/upgrades/etcd/upgrade.yml b/playbooks/common/openshift-cluster/upgrades/etcd/upgrade.yml index 0431c1ce0..39e82498d 100644 --- a/playbooks/common/openshift-cluster/upgrades/etcd/upgrade.yml +++ b/playbooks/common/openshift-cluster/upgrades/etcd/upgrade.yml @@ -36,7 +36,7 @@ - not openshift.common.is_etcd_system_container | bool - name: Record containerized etcd version (runc) - command: runc exec etcd_container rpm -qa --qf '%{version}' etcd\* + command: runc exec etcd rpm -qa --qf '%{version}' etcd\* register: etcd_container_version_runc failed_when: false # AUDIT:changed_when: `false` because we are only inspecting diff --git a/playbooks/common/openshift-cluster/upgrades/initialize_nodes_to_upgrade.yml b/playbooks/common/openshift-cluster/upgrades/initialize_nodes_to_upgrade.yml index 046535680..72de63070 100644 --- a/playbooks/common/openshift-cluster/upgrades/initialize_nodes_to_upgrade.yml +++ b/playbooks/common/openshift-cluster/upgrades/initialize_nodes_to_upgrade.yml @@ -6,27 +6,32 @@ - lib_openshift tasks: - - name: Retrieve list of openshift nodes matching upgrade label - oc_obj: - state: list - kind: node - selector: "{{ openshift_upgrade_nodes_label }}" - register: nodes_to_upgrade - when: openshift_upgrade_nodes_label is defined + - when: openshift_upgrade_nodes_label is defined + block: + - name: Retrieve list of openshift nodes matching upgrade label + oc_obj: + state: list + kind: node + selector: "{{ openshift_upgrade_nodes_label }}" + register: nodes_to_upgrade - # We got a list of nodes with the label, now we need to match these with inventory hosts - # using their openshift.common.hostname fact. - - name: Map labelled nodes to inventory hosts - add_host: - name: "{{ item }}" - groups: temp_nodes_to_upgrade - ansible_ssh_user: "{{ g_ssh_user | default(omit) }}" - ansible_become: "{{ g_sudo | default(omit) }}" - with_items: " {{ groups['oo_nodes_to_config'] }}" - when: - - openshift_upgrade_nodes_label is defined - - hostvars[item].openshift.common.hostname in nodes_to_upgrade.results.results[0]['items'] | map(attribute='metadata.name') | list - changed_when: false + - name: Fail if no nodes match openshift_upgrade_nodes_label + fail: + msg: "openshift_upgrade_nodes_label was specified but no nodes matched" + when: nodes_to_upgrade.results.results[0]['items'] | length == 0 + + # We got a list of nodes with the label, now we need to match these with inventory hosts + # using their openshift.common.hostname fact. + - name: Map labelled nodes to inventory hosts + add_host: + name: "{{ item }}" + groups: temp_nodes_to_upgrade + ansible_ssh_user: "{{ g_ssh_user | default(omit) }}" + ansible_become: "{{ g_sudo | default(omit) }}" + with_items: " {{ groups['oo_nodes_to_config'] }}" + when: + - hostvars[item].openshift.common.hostname in nodes_to_upgrade.results.results[0]['items'] | map(attribute='metadata.name') | list + changed_when: false # Build up the oo_nodes_to_upgrade group, use the list filtered by label if # present, otherwise hit all nodes: diff --git a/playbooks/common/openshift-cluster/upgrades/upgrade_control_plane.yml b/playbooks/common/openshift-cluster/upgrades/upgrade_control_plane.yml index 6738ce11f..2b2f10aee 100644 --- a/playbooks/common/openshift-cluster/upgrades/upgrade_control_plane.yml +++ b/playbooks/common/openshift-cluster/upgrades/upgrade_control_plane.yml @@ -5,13 +5,13 @@ # oc adm migrate storage should be run prior to etcd v3 upgrade # See: https://github.com/openshift/origin/pull/14625#issuecomment-308467060 -- name: Pre master upgrade - Upgrade job storage +- name: Pre master upgrade - Upgrade all storage hosts: oo_first_master tasks: - - name: Upgrade job storage + - name: Upgrade all storage command: > {{ openshift.common.client_binary }} adm --config={{ openshift.common.config_base }}/master/admin.kubeconfig - migrate storage --include=jobs --confirm + migrate storage --include=* --confirm # If facts cache were for some reason deleted, this fact may not be set, and if not set # it will always default to true. This causes problems for the etcd data dir fact detection @@ -143,13 +143,13 @@ - set_fact: master_update_complete: True -- name: Post master upgrade - Upgrade job storage +- name: Post master upgrade - Upgrade clusterpolicies storage hosts: oo_first_master tasks: - - name: Upgrade job storage + - name: Upgrade clusterpolicies storage command: > {{ openshift.common.client_binary }} adm --config={{ openshift.common.config_base }}/master/admin.kubeconfig - migrate storage --include=jobs --confirm + migrate storage --include=clusterpolicies --confirm ############################################################################## # Gate on master update complete @@ -223,13 +223,19 @@ - name: Reconcile Security Context Constraints command: > - {{ openshift.common.client_binary }} adm policy reconcile-sccs --confirm --additive-only=true -o name + {{ openshift.common.client_binary }} adm policy --config={{ openshift.common.config_base }}/master/admin.kubeconfig reconcile-sccs --confirm --additive-only=true -o name register: reconcile_scc_result changed_when: - reconcile_scc_result.stdout != '' - reconcile_scc_result.rc == 0 run_once: true + - name: Upgrade job storage + command: > + {{ openshift.common.client_binary }} adm --config={{ openshift.common.config_base }}/master/admin.kubeconfig + migrate storage --include=* --confirm + run_once: true + - set_fact: reconcile_complete: True @@ -288,15 +294,19 @@ - name: Drain Node for Kubelet upgrade command: > - {{ hostvars[groups.oo_first_master.0].openshift.common.admin_binary }} drain {{ openshift.node.nodename | lower }} --force --delete-local-data --ignore-daemonsets + {{ hostvars[groups.oo_first_master.0].openshift.common.admin_binary }} drain {{ openshift.node.nodename | lower }} --config={{ openshift.common.config_base }}/master/admin.kubeconfig --force --delete-local-data --ignore-daemonsets delegate_to: "{{ groups.oo_first_master.0 }}" + register: l_upgrade_control_plane_drain_result + until: not l_upgrade_control_plane_drain_result | failed + retries: 60 + delay: 60 roles: - lib_openshift - openshift_facts - docker - - openshift_node_upgrade - openshift_node_dnsmasq + - openshift_node_upgrade post_tasks: - name: Set node schedulability diff --git a/playbooks/common/openshift-cluster/upgrades/upgrade_nodes.yml b/playbooks/common/openshift-cluster/upgrades/upgrade_nodes.yml index 35a50cf4e..c93a5d89c 100644 --- a/playbooks/common/openshift-cluster/upgrades/upgrade_nodes.yml +++ b/playbooks/common/openshift-cluster/upgrades/upgrade_nodes.yml @@ -26,15 +26,19 @@ - name: Drain Node for Kubelet upgrade command: > - {{ hostvars[groups.oo_first_master.0].openshift.common.admin_binary }} drain {{ openshift.node.nodename | lower }} --force --delete-local-data --ignore-daemonsets + {{ hostvars[groups.oo_first_master.0].openshift.common.admin_binary }} drain {{ openshift.node.nodename | lower }} --config={{ openshift.common.config_base }}/master/admin.kubeconfig --force --delete-local-data --ignore-daemonsets delegate_to: "{{ groups.oo_first_master.0 }}" + register: l_upgrade_nodes_drain_result + until: not l_upgrade_nodes_drain_result | failed + retries: 60 + delay: 60 roles: - lib_openshift - openshift_facts - docker - - openshift_node_upgrade - openshift_node_dnsmasq + - openshift_node_upgrade - role: openshift_excluder r_openshift_excluder_action: enable r_openshift_excluder_service_type: "{{ openshift.common.service_type }}" |