Compare commits

..

10 Commits

Author SHA1 Message Date
Florian Ruynat
75d648cae5 Fix unintended SIGPIPE (#6817) 2020-10-21 03:24:20 -07:00
bozzo
087d9c204f Fix cinder & external_openstack cacert deployment (#6745) (#6832)
The CA cert was only deployed on master nodes
2020-10-21 01:48:20 -07:00
Florian Ruynat
775cadda62 Update hashes and set default to 1.18.10 (#6842) 2020-10-21 01:30:23 -07:00
Kenichi Omichi
19c000c127 Set ansible_python_interpreter to python3 on debian (fix error with mitogen) (#6633) (#6744)
Co-authored-by: Florian Ruynat <16313165+floryut@users.noreply.github.com>
2020-10-01 06:16:54 -07:00
lukasz bielinski
b39a196cfb properly generate extravolumes in kubeadmconfig for centos (#6707) 2020-09-23 01:42:08 -07:00
Florent Monbillard
9fc14b3e6c Make sure node_ip is set if node is in etcd group (#6720) 2020-09-23 00:46:08 -07:00
Florian Ruynat
f9a7dce7ca Add Kubernetes hashes 1.19.2/1.18.9/1.17.12 and set default (#6699) 2020-09-18 14:44:28 -07:00
Florian Ruynat
fbbbd90732 fix kubelet_flexvolumes_plugins_dir undefined (#6645) (#6670)
Co-authored-by: w33dw0r7d <w33dw0r7d@gmail.com>
2020-09-18 02:14:46 -07:00
Florian Ruynat
9869b46432 Move from widehat.opensuse to download.opensuse for crio centos (#6682) (#6704) 2020-09-18 02:04:45 -07:00
spaced
6cd33700f5 NetworkManager lists must be separated by , (#6649) 2020-09-11 00:30:15 -07:00
990 changed files with 34980 additions and 35604 deletions

View File

@@ -18,13 +18,3 @@ skip_list:
# While it can be useful to have these metadata available, they are also available in the existing documentation.
# (Disabled in May 2019)
- '701'
# [role-name] "meta/main.yml" Role name role-name does not match ``^+$`` pattern
# Meta roles in Kubespray don't need proper names
# (Disabled in June 2021)
- 'role-name'
# [var-naming] "defaults/main.yml" File defines variable 'apiVersion' that violates variable naming standards
# In Kubespray we use variables that use camelCase to match their k8s counterparts
# (Disabled in June 2021)
- 'var-naming'

View File

@@ -1,7 +1,7 @@
---
name: Support Request
about: Support request or question relating to Kubespray
labels: kind/support
labels: triage/support
---

11
.gitignore vendored
View File

@@ -15,7 +15,6 @@ contrib/terraform/aws/credentials.tfvars
**/*.sw[pon]
*~
vagrant/
plugins/mitogen
# Ansible inventory
inventory/*
@@ -99,13 +98,3 @@ target/
# virtualenv
venv/
ENV/
# molecule
roles/**/molecule/**/__pycache__/
roles/**/molecule/**/*.conf
# macOS
.DS_Store
# Temp location used by our scripts
scripts/tmp/

View File

@@ -8,15 +8,13 @@ stages:
- deploy-special
variables:
KUBESPRAY_VERSION: v2.17.1
KUBESPRAY_VERSION: v2.13.3
FAILFASTCI_NAMESPACE: 'kargo-ci'
GITLAB_REPOSITORY: 'kargo-ci/kubernetes-sigs-kubespray'
ANSIBLE_FORCE_COLOR: "true"
MAGIC: "ci check this"
TEST_ID: "$CI_PIPELINE_ID-$CI_BUILD_ID"
CI_TEST_VARS: "./tests/files/${CI_JOB_NAME}.yml"
CI_TEST_REGISTRY_MIRROR: "./tests/common/_docker_hub_registry_mirror.yml"
CI_TEST_SETTING: "./tests/common/_kubespray_test_settings.yml"
GS_ACCESS_KEY_ID: $GS_KEY
GS_SECRET_ACCESS_KEY: $GS_SECRET
CONTAINER_ENGINE: docker
@@ -31,15 +29,12 @@ variables:
MITOGEN_ENABLE: "false"
ANSIBLE_LOG_LEVEL: "-vv"
RECOVER_CONTROL_PLANE_TEST: "false"
RECOVER_CONTROL_PLANE_TEST_GROUPS: "etcd[2:],kube_control_plane[1:]"
TERRAFORM_VERSION: 1.0.8
ANSIBLE_MAJOR_VERSION: "2.10"
RECOVER_CONTROL_PLANE_TEST_GROUPS: "etcd[2:],kube-master[1:]"
before_script:
- ./tests/scripts/rebase.sh
- update-alternatives --install /usr/bin/python python /usr/bin/python3 1
- python -m pip uninstall -y ansible ansible-base ansible-core
- python -m pip install -r tests/requirements-${ANSIBLE_MAJOR_VERSION}.txt
- python -m pip install -r tests/requirements.txt
- mkdir -p /.ssh
.job: &job
@@ -53,7 +48,6 @@ before_script:
.testcases: &testcases
<<: *job
retry: 1
before_script:
- update-alternatives --install /usr/bin/python python /usr/bin/python3 1
- ./tests/scripts/rebase.sh

View File

@@ -14,7 +14,7 @@ vagrant-validate:
stage: unit-tests
tags: [light]
variables:
VAGRANT_VERSION: 2.2.19
VAGRANT_VERSION: 2.2.4
script:
- ./tests/scripts/vagrant-validate.sh
except: ['triggers', 'master']
@@ -53,7 +53,6 @@ tox-inventory-builder:
- ./tests/scripts/rebase.sh
- apt-get update && apt-get install -y python3-pip
- update-alternatives --install /usr/bin/python python /usr/bin/python3 10
- python -m pip uninstall -y ansible
- python -m pip install -r tests/requirements.txt
script:
- pip3 install tox
@@ -65,9 +64,9 @@ markdownlint:
tags: [light]
image: node
before_script:
- npm install -g markdownlint-cli@0.22.0
- npm install -g markdownlint-cli
script:
- markdownlint $(find . -name '*.md' | grep -vF './.git') --ignore docs/_sidebar.md --ignore contrib/dind/README.md
- markdownlint README.md docs --ignore docs/_sidebar.md
ci-matrix:
stage: unit-tests

View File

@@ -1,143 +1,88 @@
---
.packet:
.packet: &packet
extends: .testcases
variables:
ANSIBLE_TIMEOUT: "120"
CI_PLATFORM: packet
SSH_USER: kubespray
CI_PLATFORM: "packet"
SSH_USER: "kubespray"
tags:
- packet
except: [triggers]
# CI template for PRs
.packet_pr:
only: [/^pr-.*$/]
extends: .packet
except: ['triggers']
# CI template for periodic CI jobs
# Enabled when PERIODIC_CI_ENABLED var is set
.packet_periodic:
only:
variables:
- $PERIODIC_CI_ENABLED
allow_failure: true
packet_ubuntu18-calico-aio:
stage: deploy-part1
extends: .packet
when: on_success
# The ubuntu20-calico-aio jobs are meant as early stages to prevent running the full CI if something is horribly broken
# Future AIO job
packet_ubuntu20-calico-aio:
stage: deploy-part1
extends: .packet_pr
extends: .packet
when: on_success
variables:
RESET_CHECK: "true"
# Exericse ansible variants
packet_ubuntu20-calico-aio-ansible-2_9:
stage: deploy-part1
extends: .packet_pr
when: on_success
variables:
ANSIBLE_MAJOR_VERSION: "2.9"
RESET_CHECK: "true"
packet_ubuntu20-calico-aio-ansible-2_11:
stage: deploy-part1
extends: .packet_pr
when: on_success
variables:
ANSIBLE_MAJOR_VERSION: "2.11"
RESET_CHECK: "true"
# ### PR JOBS PART2
packet_ubuntu18-aio-docker:
packet_centos7-flannel-containerd-addons-ha:
extends: .packet
stage: deploy-part2
extends: .packet_pr
when: on_success
variables:
MITOGEN_ENABLE: "true"
packet_ubuntu20-aio-docker:
stage: deploy-part2
extends: .packet_pr
when: on_success
packet_ubuntu18-calico-aio:
stage: deploy-part2
extends: .packet_pr
when: on_success
packet_centos7-flannel-addons-ha:
extends: .packet_pr
stage: deploy-part2
when: on_success
packet_almalinux8-crio:
extends: .packet_pr
packet_centos7-crio:
extends: .packet
stage: deploy-part2
when: on_success
variables:
MITOGEN_ENABLE: "true"
packet_ubuntu18-crio:
extends: .packet_pr
extends: .packet
stage: deploy-part2
when: manual
variables:
MITOGEN_ENABLE: "true"
packet_fedora35-crio:
extends: .packet_pr
packet_ubuntu16-canal-kubeadm-ha:
stage: deploy-part2
when: manual
packet_ubuntu16-canal-ha:
stage: deploy-part2
extends: .packet_periodic
extends: .packet
when: on_success
packet_ubuntu16-canal-sep:
stage: deploy-special
extends: .packet_pr
extends: .packet
when: manual
packet_ubuntu16-flannel-ha:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: manual
packet_ubuntu16-kube-router-sep:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: manual
packet_ubuntu16-kube-router-svc-proxy:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: manual
packet_debian10-cilium-svc-proxy:
stage: deploy-part2
extends: .packet_periodic
when: on_success
extends: .packet
when: manual
packet_debian10-calico:
packet_debian10-containerd:
stage: deploy-part2
extends: .packet_pr
when: on_success
packet_debian10-docker:
stage: deploy-part2
extends: .packet_pr
when: on_success
packet_debian11-calico:
stage: deploy-part2
extends: .packet_pr
when: on_success
packet_debian11-docker:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: on_success
variables:
MITOGEN_ENABLE: "true"
packet_centos7-calico-ha-once-localhost:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: on_success
variables:
# This will instruct Docker not to start over TLS.
@@ -145,168 +90,140 @@ packet_centos7-calico-ha-once-localhost:
services:
- docker:19.03.9-dind
packet_almalinux8-kube-ovn:
packet_centos8-kube-ovn:
stage: deploy-part2
extends: .packet_periodic
extends: .packet
when: on_success
packet_almalinux8-calico:
packet_centos8-calico:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: on_success
packet_almalinux8-docker:
packet_fedora32-weave:
stage: deploy-part2
extends: .packet_pr
when: on_success
packet_fedora34-docker-weave:
stage: deploy-part2
extends: .packet_pr
when: on_success
packet_fedora35-kube-router:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: on_success
packet_opensuse-canal:
stage: deploy-part2
extends: .packet_periodic
extends: .packet
when: on_success
packet_opensuse-docker-cilium:
packet_ubuntu18-ovn4nfv:
stage: deploy-part2
extends: .packet_pr
when: manual
extends: .packet
when: on_success
# Contiv does not work in k8s v1.16
# packet_ubuntu16-contiv-sep:
# stage: deploy-part2
# extends: .packet
# when: on_success
# ### MANUAL JOBS
packet_ubuntu16-docker-weave-sep:
packet_ubuntu16-weave-sep:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: manual
packet_ubuntu18-cilium-sep:
stage: deploy-special
extends: .packet_pr
extends: .packet
when: manual
packet_ubuntu18-flannel-ha:
packet_ubuntu18-flannel-containerd-ha:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: manual
packet_ubuntu18-flannel-ha-once:
packet_ubuntu18-flannel-containerd-ha-once:
stage: deploy-part2
extends: .packet_pr
when: manual
# Calico HA eBPF
packet_almalinux8-calico-ha-ebpf:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: manual
packet_debian9-macvlan:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: manual
packet_centos7-calico-ha:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: manual
packet_centos7-kube-router:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: manual
packet_centos7-multus-calico:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: manual
packet_oracle7-canal-ha:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: manual
packet_fedora35-docker-calico:
packet_fedora31-flannel:
stage: deploy-part2
extends: .packet_periodic
extends: .packet
when: on_success
variables:
RESET_CHECK: "true"
packet_fedora34-calico-selinux:
stage: deploy-part2
extends: .packet_periodic
when: on_success
packet_fedora35-calico-swap-selinux:
stage: deploy-part2
extends: .packet_pr
when: manual
MITOGEN_ENABLE: "true"
packet_amazon-linux-2-aio:
stage: deploy-part2
extends: .packet_pr
extends: .packet
when: manual
packet_almalinux8-calico-nodelocaldns-secondary:
packet_fedora32-kube-ovn-containerd:
stage: deploy-part2
extends: .packet_pr
when: manual
packet_fedora34-kube-ovn:
stage: deploy-part2
extends: .packet_periodic
extends: .packet
when: on_success
# ### PR JOBS PART3
# Long jobs (45min+)
packet_centos7-docker-weave-upgrade-ha:
packet_centos7-weave-upgrade-ha:
stage: deploy-part3
extends: .packet_periodic
extends: .packet
when: on_success
variables:
UPGRADE_TEST: basic
MITOGEN_ENABLE: "false"
# Calico HA Wireguard
packet_ubuntu20-calico-ha-wireguard:
stage: deploy-part2
extends: .packet_pr
when: manual
packet_debian10-calico-upgrade:
packet_debian9-calico-upgrade:
stage: deploy-part3
extends: .packet_pr
extends: .packet
when: on_success
variables:
UPGRADE_TEST: graceful
MITOGEN_ENABLE: "false"
packet_debian10-calico-upgrade-once:
packet_debian9-calico-upgrade-once:
stage: deploy-part3
extends: .packet_periodic
extends: .packet
when: on_success
variables:
UPGRADE_TEST: graceful
MITOGEN_ENABLE: "false"
packet_ubuntu18-calico-ha-recover:
stage: deploy-part3
extends: .packet_periodic
extends: .packet
when: on_success
variables:
RECOVER_CONTROL_PLANE_TEST: "true"
RECOVER_CONTROL_PLANE_TEST_GROUPS: "etcd[2:],kube_control_plane[1:]"
RECOVER_CONTROL_PLANE_TEST_GROUPS: "etcd[2:],kube-master[1:]"
packet_ubuntu18-calico-ha-recover-noquorum:
stage: deploy-part3
extends: .packet_periodic
extends: .packet
when: on_success
variables:
RECOVER_CONTROL_PLANE_TEST: "true"
RECOVER_CONTROL_PLANE_TEST_GROUPS: "etcd[1:],kube_control_plane[1:]"
RECOVER_CONTROL_PLANE_TEST_GROUPS: "etcd[1:],kube-master[1:]"

View File

@@ -4,7 +4,7 @@ shellcheck:
stage: unit-tests
tags: [light]
variables:
SHELLCHECK_VERSION: v0.7.1
SHELLCHECK_VERSION: v0.6.0
before_script:
- ./tests/scripts/rebase.sh
- curl --silent --location "https://github.com/koalaman/shellcheck/releases/download/"${SHELLCHECK_VERSION}"/shellcheck-"${SHELLCHECK_VERSION}".linux.x86_64.tar.xz" | tar -xJv
@@ -12,5 +12,5 @@ shellcheck:
- shellcheck --version
script:
# Run shellcheck for all *.sh except contrib/
- find . -name '*.sh' -not -path './contrib/*' -not -path './.git/*' | xargs shellcheck --severity error
- find . -name '*.sh' -not -path './contrib/*' | xargs shellcheck --severity error
except: ['triggers', 'master']

View File

@@ -12,13 +12,13 @@
# Prepare inventory
- cp contrib/terraform/$PROVIDER/sample-inventory/cluster.tfvars .
- ln -s contrib/terraform/$PROVIDER/hosts
- terraform -chdir="contrib/terraform/$PROVIDER" init
- terraform init contrib/terraform/$PROVIDER
# Copy SSH keypair
- mkdir -p ~/.ssh
- echo "$PACKET_PRIVATE_KEY" | base64 -d > ~/.ssh/id_rsa
- chmod 400 ~/.ssh/id_rsa
- echo "$PACKET_PUBLIC_KEY" | base64 -d > ~/.ssh/id_rsa.pub
- mkdir -p contrib/terraform/$PROVIDER/group_vars
- mkdir -p group_vars
# Random subnet to avoid routing conflicts
- export TF_VAR_subnet_cidr="10.$(( $RANDOM % 256 )).$(( $RANDOM % 256 )).0/24"
@@ -28,8 +28,8 @@
tags: [light]
only: ['master', /^pr-.*$/]
script:
- terraform -chdir="contrib/terraform/$PROVIDER" validate
- terraform -chdir="contrib/terraform/$PROVIDER" fmt -check -diff
- terraform validate -var-file=cluster.tfvars contrib/terraform/$PROVIDER
- terraform fmt -check -diff contrib/terraform/$PROVIDER
.terraform_apply:
extends: .terraform_install
@@ -56,48 +56,28 @@
tf-validate-openstack:
extends: .terraform_validate
variables:
TF_VERSION: $TERRAFORM_VERSION
TF_VERSION: 0.12.24
PROVIDER: openstack
CLUSTER: $CI_COMMIT_REF_NAME
tf-validate-packet:
extends: .terraform_validate
variables:
TF_VERSION: $TERRAFORM_VERSION
TF_VERSION: 0.12.24
PROVIDER: packet
CLUSTER: $CI_COMMIT_REF_NAME
tf-validate-aws:
extends: .terraform_validate
variables:
TF_VERSION: $TERRAFORM_VERSION
TF_VERSION: 0.12.24
PROVIDER: aws
CLUSTER: $CI_COMMIT_REF_NAME
tf-validate-exoscale:
extends: .terraform_validate
variables:
TF_VERSION: $TERRAFORM_VERSION
PROVIDER: exoscale
tf-validate-vsphere:
extends: .terraform_validate
variables:
TF_VERSION: $TERRAFORM_VERSION
PROVIDER: vsphere
CLUSTER: $CI_COMMIT_REF_NAME
tf-validate-upcloud:
extends: .terraform_validate
variables:
TF_VERSION: $TERRAFORM_VERSION
PROVIDER: upcloud
CLUSTER: $CI_COMMIT_REF_NAME
# tf-packet-ubuntu16-default:
# extends: .terraform_apply
# variables:
# TF_VERSION: $TERRAFORM_VERSION
# TF_VERSION: 0.12.24
# PROVIDER: packet
# CLUSTER: $CI_COMMIT_REF_NAME
# TF_VAR_number_of_k8s_masters: "1"
@@ -111,7 +91,7 @@ tf-validate-upcloud:
# tf-packet-ubuntu18-default:
# extends: .terraform_apply
# variables:
# TF_VERSION: $TERRAFORM_VERSION
# TF_VERSION: 0.12.24
# PROVIDER: packet
# CLUSTER: $CI_COMMIT_REF_NAME
# TF_VAR_number_of_k8s_masters: "1"
@@ -146,6 +126,10 @@ tf-validate-upcloud:
OS_INTERFACE: public
OS_IDENTITY_API_VERSION: "3"
TF_VAR_router_id: "ab95917c-41fb-4881-b507-3a6dfe9403df"
# Since ELASTX is in Stockholm, Mitogen helps with latency
MITOGEN_ENABLE: "false"
# Mitogen doesn't support interpreter discovery yet
ANSIBLE_PYTHON_INTERPRETER: "/usr/bin/python3"
tf-elastx_cleanup:
stage: unit-tests
@@ -162,10 +146,9 @@ tf-elastx_ubuntu18-calico:
extends: .terraform_apply
stage: deploy-part3
when: on_success
allow_failure: true
variables:
<<: *elastx_variables
TF_VERSION: $TERRAFORM_VERSION
TF_VERSION: 0.12.24
PROVIDER: openstack
CLUSTER: $CI_COMMIT_REF_NAME
ANSIBLE_TIMEOUT: "60"
@@ -191,45 +174,44 @@ tf-elastx_ubuntu18-calico:
TF_VAR_image: ubuntu-18.04-server-latest
TF_VAR_k8s_allowed_remote_ips: '["0.0.0.0/0"]'
# OVH voucher expired, commenting job until things are sorted out
# tf-ovh_cleanup:
# stage: unit-tests
# tags: [light]
# image: python
# environment: ovh
# variables:
# <<: *ovh_variables
# before_script:
# - pip install -r scripts/openstack-cleanup/requirements.txt
# script:
# - ./scripts/openstack-cleanup/main.py
tf-ovh_cleanup:
stage: unit-tests
tags: [light]
image: python
environment: ovh
variables:
<<: *ovh_variables
before_script:
- pip install -r scripts/openstack-cleanup/requirements.txt
script:
- ./scripts/openstack-cleanup/main.py
# tf-ovh_ubuntu18-calico:
# extends: .terraform_apply
# when: on_success
# environment: ovh
# variables:
# <<: *ovh_variables
# TF_VERSION: $TERRAFORM_VERSION
# PROVIDER: openstack
# CLUSTER: $CI_COMMIT_REF_NAME
# ANSIBLE_TIMEOUT: "60"
# SSH_USER: ubuntu
# TF_VAR_number_of_k8s_masters: "0"
# TF_VAR_number_of_k8s_masters_no_floating_ip: "1"
# TF_VAR_number_of_k8s_masters_no_floating_ip_no_etcd: "0"
# TF_VAR_number_of_etcd: "0"
# TF_VAR_number_of_k8s_nodes: "0"
# TF_VAR_number_of_k8s_nodes_no_floating_ip: "1"
# TF_VAR_number_of_gfs_nodes_no_floating_ip: "0"
# TF_VAR_number_of_bastions: "0"
# TF_VAR_number_of_k8s_masters_no_etcd: "0"
# TF_VAR_use_neutron: "0"
# TF_VAR_floatingip_pool: "Ext-Net"
# TF_VAR_external_net: "6011fbc9-4cbf-46a4-8452-6890a340b60b"
# TF_VAR_network_name: "Ext-Net"
# TF_VAR_flavor_k8s_master: "defa64c3-bd46-43b4-858a-d93bbae0a229" # s1-8
# TF_VAR_flavor_k8s_node: "defa64c3-bd46-43b4-858a-d93bbae0a229" # s1-8
# TF_VAR_image: "Ubuntu 18.04"
# TF_VAR_k8s_allowed_remote_ips: '["0.0.0.0/0"]'
tf-ovh_ubuntu18-calico:
extends: .terraform_apply
when: on_success
environment: ovh
variables:
<<: *ovh_variables
TF_VERSION: 0.12.24
PROVIDER: openstack
CLUSTER: $CI_COMMIT_REF_NAME
ANSIBLE_TIMEOUT: "60"
SSH_USER: ubuntu
TF_VAR_number_of_k8s_masters: "0"
TF_VAR_number_of_k8s_masters_no_floating_ip: "1"
TF_VAR_number_of_k8s_masters_no_floating_ip_no_etcd: "0"
TF_VAR_number_of_etcd: "0"
TF_VAR_number_of_k8s_nodes: "0"
TF_VAR_number_of_k8s_nodes_no_floating_ip: "1"
TF_VAR_number_of_gfs_nodes_no_floating_ip: "0"
TF_VAR_number_of_bastions: "0"
TF_VAR_number_of_k8s_masters_no_etcd: "0"
TF_VAR_use_neutron: "0"
TF_VAR_floatingip_pool: "Ext-Net"
TF_VAR_external_net: "6011fbc9-4cbf-46a4-8452-6890a340b60b"
TF_VAR_network_name: "Ext-Net"
TF_VAR_flavor_k8s_master: "defa64c3-bd46-43b4-858a-d93bbae0a229" # s1-8
TF_VAR_flavor_k8s_node: "defa64c3-bd46-43b4-858a-d93bbae0a229" # s1-8
TF_VAR_image: "Ubuntu 18.04"
TF_VAR_k8s_allowed_remote_ips: '["0.0.0.0/0"]'

View File

@@ -4,24 +4,17 @@ molecule_tests:
tags: [c3.small.x86]
only: [/^pr-.*$/]
except: ['triggers']
image: quay.io/kubespray/vagrant:v2.18.0
image: quay.io/kubespray/vagrant:$KUBESPRAY_VERSION
services: []
stage: deploy-part1
before_script:
- tests/scripts/rebase.sh
- apt-get update && apt-get install -y python3-pip
- update-alternatives --install /usr/bin/python python /usr/bin/python3 10
- python -m pip uninstall -y ansible
- python -m pip install -r tests/requirements.txt
- ./tests/scripts/vagrant_clean.sh
script:
- ./tests/scripts/molecule_run.sh
after_script:
- chronic ./tests/scripts/molecule_logs.sh
artifacts:
when: always
paths:
- molecule_logs/
.vagrant:
extends: .testcases
@@ -33,24 +26,17 @@ molecule_tests:
tags: [c3.small.x86]
only: [/^pr-.*$/]
except: ['triggers']
image: quay.io/kubespray/vagrant:v2.18.0
image: quay.io/kubespray/vagrant:$KUBESPRAY_VERSION
services: []
before_script:
- apt-get update && apt-get install -y python3-pip
- update-alternatives --install /usr/bin/python python /usr/bin/python3 10
- python -m pip uninstall -y ansible
- python -m pip install -r tests/requirements.txt
- ./tests/scripts/vagrant_clean.sh
script:
- ./tests/scripts/testcases_run.sh
after_script:
- chronic ./tests/scripts/testcases_cleanup.sh
allow_failure: true
vagrant_ubuntu18-calico-dual-stack:
stage: deploy-part2
extends: .vagrant
when: on_success
vagrant_ubuntu18-flannel:
stage: deploy-part2

View File

@@ -1,9 +1,6 @@
---
extends: default
ignore: |
.git/
rules:
braces:
min-spaces-inside: 0

View File

@@ -6,17 +6,11 @@
It is recommended to use filter to manage the GitHub email notification, see [examples for setting filters to Kubernetes Github notifications](https://github.com/kubernetes/community/blob/master/communication/best-practices.md#examples-for-setting-filters-to-kubernetes-github-notifications)
To install development dependencies you can set up a python virtual env with the necessary dependencies:
```ShellSession
virtualenv venv
source venv/bin/activate
pip install -r tests/requirements.txt
```
To install development dependencies you can use `pip install -r tests/requirements.txt`
#### Linting
Kubespray uses `yamllint` and `ansible-lint`. To run them locally use `yamllint .` and `ansible-lint`. It is a good idea to add call these tools as part of your pre-commit hook and avoid a lot of back end forth on fixing linting issues (<https://support.gitkraken.com/working-with-repositories/githooksexample/>).
Kubespray uses `yamllint` and `ansible-lint`. To run them locally use `yamllint .` and `./tests/scripts/ansible-lint.sh`
#### Molecule
@@ -35,5 +29,3 @@ Vagrant with VirtualBox or libvirt driver helps you to quickly spin test cluster
3. Fork the desired repo, develop and test your code changes.
4. Sign the CNCF CLA (<https://git.k8s.io/community/CLA.md#the-contributor-license-agreement>)
5. Submit a pull request.
6. Work with the reviewers on their suggestions.
7. Ensure to rebase to the HEAD of your target branch and squash un-necessary commits (<https://blog.carbonfive.com/always-squash-and-rebase-your-git-commits/>) before final merger of your contribution.

View File

@@ -1,33 +1,21 @@
# Use imutable image tags rather than mutable tags (like ubuntu:20.04)
FROM ubuntu:focal-20220316
FROM ubuntu:18.04
RUN apt update -y \
&& apt install -y \
RUN mkdir /kubespray
WORKDIR /kubespray
RUN apt update -y && \
apt install -y \
libssl-dev python3-dev sshpass apt-transport-https jq moreutils \
ca-certificates curl gnupg2 software-properties-common python3-pip unzip rsync git \
&& rm -rf /var/lib/apt/lists/*
RUN curl -fsSL https://download.docker.com/linux/ubuntu/gpg | apt-key add - \
&& add-apt-repository \
ca-certificates curl gnupg2 software-properties-common python3-pip rsync
RUN curl -fsSL https://download.docker.com/linux/ubuntu/gpg | apt-key add - && \
add-apt-repository \
"deb [arch=amd64] https://download.docker.com/linux/ubuntu \
$(lsb_release -cs) \
stable" \
&& apt update -y && apt-get install --no-install-recommends -y docker-ce \
&& rm -rf /var/lib/apt/lists/*
&& apt update -y && apt-get install docker-ce -y
COPY . .
RUN /usr/bin/python3 -m pip install pip -U && /usr/bin/python3 -m pip install -r tests/requirements.txt && python3 -m pip install -r requirements.txt && update-alternatives --install /usr/bin/python python /usr/bin/python3 1
RUN curl -LO https://storage.googleapis.com/kubernetes-release/release/v1.17.5/bin/linux/amd64/kubectl \
&& chmod a+x kubectl && cp kubectl /usr/local/bin/kubectl
# Some tools like yamllint need this
# Pip needs this as well at the moment to install ansible
# (and potentially other packages)
# See: https://github.com/pypa/pip/issues/10219
ENV LANG=C.UTF-8
WORKDIR /kubespray
COPY . .
RUN /usr/bin/python3 -m pip install --no-cache-dir pip -U \
&& /usr/bin/python3 -m pip install --no-cache-dir -r tests/requirements.txt \
&& python3 -m pip install --no-cache-dir -r requirements.txt \
&& update-alternatives --install /usr/bin/python python /usr/bin/python3 1
RUN KUBE_VERSION=$(sed -n 's/^kube_version: //p' roles/kubespray-defaults/defaults/main.yaml) \
&& curl -LO https://storage.googleapis.com/kubernetes-release/release/$KUBE_VERSION/bin/linux/amd64/kubectl \
&& chmod a+x kubectl \
&& mv kubectl /usr/local/bin/kubectl

View File

@@ -1,7 +1,5 @@
mitogen:
@echo Mitogen support is deprecated.
@echo Please run the following command manually:
@echo ansible-playbook -c local mitogen.yml -vv
ansible-playbook -c local mitogen.yml -vv
clean:
rm -rf dist/
rm *.retry

2
OWNERS
View File

@@ -4,5 +4,3 @@ approvers:
- kubespray-approvers
reviewers:
- kubespray-reviewers
emeritus_approvers:
- kubespray-emeritus_approvers

View File

@@ -1,21 +1,19 @@
aliases:
kubespray-approvers:
- ant31
- mattymo
- atoms
- chadswen
- mirwan
- miouge1
- riverzhang
- verwilst
- woopstar
- luckysb
- floryut
- oomichi
kubespray-reviewers:
- jjungnickel
- archifleks
- holmsten
- bozzo
- floryut
- eppo
- oomichi
- jayonlau
- cristicalin
kubespray-emeritus_approvers:
- riverzhang
- atoms
- ant31

View File

@@ -5,7 +5,7 @@
If you have questions, check the documentation at [kubespray.io](https://kubespray.io) and join us on the [kubernetes slack](https://kubernetes.slack.com), channel **\#kubespray**.
You can get your invite [here](http://slack.k8s.io/)
- Can be deployed on **[AWS](docs/aws.md), GCE, [Azure](docs/azure.md), [OpenStack](docs/openstack.md), [vSphere](docs/vsphere.md), [Equinix Metal](docs/equinix-metal.md) (bare metal), Oracle Cloud Infrastructure (Experimental), or Baremetal**
- Can be deployed on **[AWS](docs/aws.md), GCE, [Azure](docs/azure.md), [OpenStack](docs/openstack.md), [vSphere](docs/vsphere.md), [Packet](docs/packet.md) (bare metal), Oracle Cloud Infrastructure (Experimental), or Baremetal**
- **Highly available** cluster
- **Composable** (Choice of the network plugin for instance)
- Supports most popular **Linux distributions**
@@ -32,7 +32,7 @@ CONFIG_FILE=inventory/mycluster/hosts.yaml python3 contrib/inventory_builder/inv
# Review and change parameters under ``inventory/mycluster/group_vars``
cat inventory/mycluster/group_vars/all/all.yml
cat inventory/mycluster/group_vars/k8s_cluster/k8s-cluster.yml
cat inventory/mycluster/group_vars/k8s-cluster/k8s-cluster.yml
# Deploy Kubespray with Ansible Playbook - run the playbook as root
# The option `--become` is required, as for example writing SSL keys in /etc/,
@@ -48,23 +48,11 @@ As a consequence, `ansible-playbook` command will fail with:
ERROR! no action detected in task. This often indicates a misspelled module name, or incorrect module path.
```
probably pointing on a task depending on a module present in requirements.txt.
probably pointing on a task depending on a module present in requirements.txt (i.e. "unseal vault").
One way of solving this would be to uninstall the Ansible package and then, to install it via pip but it is not always possible.
A workaround consists of setting `ANSIBLE_LIBRARY` and `ANSIBLE_MODULE_UTILS` environment variables respectively to the `ansible/modules` and `ansible/module_utils` subdirectories of pip packages installation location, which can be found in the Location field of the output of `pip show [package]` before executing `ansible-playbook`.
A simple way to ensure you get all the correct version of Ansible is to use the [pre-built docker image from Quay](https://quay.io/repository/kubespray/kubespray?tab=tags).
You will then need to use [bind mounts](https://docs.docker.com/storage/bind-mounts/) to get the inventory and ssh key into the container, like this:
```ShellSession
docker pull quay.io/kubespray/kubespray:v2.17.1
docker run --rm -it --mount type=bind,source="$(pwd)"/inventory/sample,dst=/inventory \
--mount type=bind,source="${HOME}"/.ssh/id_rsa,dst=/root/.ssh/id_rsa \
quay.io/kubespray/kubespray:v2.17.1 bash
# Inside the container you may now run the kubespray playbooks:
ansible-playbook -i /inventory/inventory.ini --private-key /root/.ssh/id_rsa cluster.yml
```
### Vagrant
For Vagrant we need to install python dependencies for provisioning tasks.
@@ -105,7 +93,7 @@ vagrant up
- [AWS](docs/aws.md)
- [Azure](docs/azure.md)
- [vSphere](docs/vsphere.md)
- [Equinix Metal](docs/equinix-metal.md)
- [Packet Host](docs/packet.md)
- [Large deployments](docs/large-deployments.md)
- [Adding/replacing a node](docs/nodes.md)
- [Upgrades basics](docs/upgrades.md)
@@ -115,56 +103,53 @@ vagrant up
## Supported Linux Distributions
- **Flatcar Container Linux by Kinvolk**
- **Debian** Bullseye, Buster, Jessie, Stretch
- **Debian** Buster, Jessie, Stretch, Wheezy
- **Ubuntu** 16.04, 18.04, 20.04
- **CentOS/RHEL** 7, [8](docs/centos8.md)
- **Fedora** 34, 35
- **Fedora CoreOS** (see [fcos Note](docs/fcos.md))
- **openSUSE** Leap 15.x/Tumbleweed
- **Oracle Linux** 7, [8](docs/centos8.md)
- **Alma Linux** [8](docs/centos8.md)
- **Rocky Linux** [8](docs/centos8.md)
- **Amazon Linux 2** (experimental: see [amazon linux notes](docs/amazonlinux.md))
- **CentOS/RHEL** 7, 8 (experimental: see [centos 8 notes](docs/centos8.md))
- **Fedora** 31, 32
- **Fedora CoreOS** (experimental: see [fcos Note](docs/fcos.md))
- **openSUSE** Leap 42.3/Tumbleweed
- **Oracle Linux** 7, 8 (experimental: [centos 8 notes](docs/centos8.md) apply)
Note: Upstart/SysV init based OS types are not supported.
## Supported Components
- Core
- [kubernetes](https://github.com/kubernetes/kubernetes) v1.22.8
- [etcd](https://github.com/coreos/etcd) v3.5.0
- [docker](https://www.docker.com/) v20.10 (see note)
- [containerd](https://containerd.io/) v1.5.8
- [cri-o](http://cri-o.io/) v1.22 (experimental: see [CRI-O Note](docs/cri-o.md). Only on fedora, ubuntu and centos based OS)
- [kubernetes](https://github.com/kubernetes/kubernetes) v1.18.10
- [etcd](https://github.com/coreos/etcd) v3.4.3
- [docker](https://www.docker.com/) v19.03 (see note)
- [containerd](https://containerd.io/) v1.2.13
- [cri-o](http://cri-o.io/) v1.17 (experimental: see [CRI-O Note](docs/cri-o.md). Only on fedora, ubuntu and centos based OS)
- Network Plugin
- [cni-plugins](https://github.com/containernetworking/plugins) v1.0.1
- [calico](https://github.com/projectcalico/calico) v3.20.3
- [cni-plugins](https://github.com/containernetworking/plugins) v0.8.6
- [calico](https://github.com/projectcalico/calico) v3.15.2
- [canal](https://github.com/projectcalico/canal) (given calico/flannel versions)
- [cilium](https://github.com/cilium/cilium) v1.9.11
- [flanneld](https://github.com/flannel-io/flannel) v0.15.1
- [kube-ovn](https://github.com/alauda/kube-ovn) v1.8.1
- [kube-router](https://github.com/cloudnativelabs/kube-router) v1.3.2
- [multus](https://github.com/intel/multus-cni) v3.8
- [weave](https://github.com/weaveworks/weave) v2.8.1
- [cilium](https://github.com/cilium/cilium) v1.8.3
- [contiv](https://github.com/contiv/install) v1.2.1
- [flanneld](https://github.com/coreos/flannel) v0.12.0
- [kube-ovn](https://github.com/alauda/kube-ovn) v1.3.0
- [kube-router](https://github.com/cloudnativelabs/kube-router) v1.0.1
- [multus](https://github.com/intel/multus-cni) v3.6.0
- [ovn4nfv](https://github.com/opnfv/ovn4nfv-k8s-plugin) v1.1.0
- [weave](https://github.com/weaveworks/weave) v2.7.0
- Application
- [ambassador](https://github.com/datawire/ambassador): v1.5
- [cephfs-provisioner](https://github.com/kubernetes-incubator/external-storage) v2.1.0-k8s1.11
- [rbd-provisioner](https://github.com/kubernetes-incubator/external-storage) v2.1.1-k8s1.11
- [cert-manager](https://github.com/jetstack/cert-manager) v1.5.4
- [coredns](https://github.com/coredns/coredns) v1.8.0
- [ingress-nginx](https://github.com/kubernetes/ingress-nginx) v1.0.4
- [cert-manager](https://github.com/jetstack/cert-manager) v0.16.1
- [coredns](https://github.com/coredns/coredns) v1.6.7
- [ingress-nginx](https://github.com/kubernetes/ingress-nginx) v0.35.0
## Container Runtime Notes
- The list of available docker version is 18.09, 19.03 and 20.10. The recommended docker version is 20.10. The kubelet might break on docker's non-standard version numbering (it no longer uses semantic versioning). To ensure auto-updates don't break your cluster look into e.g. yum versionlock plugin or apt pin).
- The cri-o version should be aligned with the respective kubernetes version (i.e. kube_version=1.20.x, crio_version=1.20)
Note: The list of validated [docker versions](https://kubernetes.io/docs/setup/production-environment/container-runtimes/#docker) is 1.13.1, 17.03, 17.06, 17.09, 18.06, 18.09 and 19.03. The recommended docker version is 19.03. The kubelet might break on docker's non-standard version numbering (it no longer uses semantic versioning). To ensure auto-updates don't break your cluster look into e.g. yum versionlock plugin or apt pin).
## Requirements
- **Minimum required version of Kubernetes is v1.20**
- **Ansible v2.9.x, Jinja 2.11+ and python-netaddr is installed on the machine that will run Ansible commands, Ansible 2.10.x is experimentally supported for now**
- **Minimum required version of Kubernetes is v1.17**
- **Ansible v2.9+, Jinja 2.11+ and python-netaddr is installed on the machine that will run Ansible commands**
- The target servers must have **access to the Internet** in order to pull docker images. Otherwise, additional configuration is required (See [Offline Environment](docs/offline-environment.md))
- The target servers are configured to allow **IPv4 forwarding**.
- If using IPv6 for pods and services, the target servers are configured to allow **IPv6 forwarding**.
- **Your ssh key must be copied** to all the servers part of your inventory.
- The **firewalls are not managed**, you'll need to implement your own rules the way you used to.
in order to avoid any issue during deployment you should disable your firewall.
- If kubespray is ran from non-root user account, correct privilege escalation method
@@ -194,6 +179,11 @@ You can choose between 10 network plugins. (default: `calico`, except Vagrant us
- [cilium](http://docs.cilium.io/en/latest/): layer 3/4 networking (as well as layer 7 to protect and secure application protocols), supports dynamic insertion of BPF bytecode into the Linux kernel to implement security services, networking and visibility logic.
- [contiv](docs/contiv.md): supports vlan, vxlan, bgp and Cisco SDN networking. This plugin is able to
apply firewall policies, segregate containers in multiple network and bridging pods onto physical networks.
- [ovn4nfv](docs/ovn4nfv.md): [ovn4nfv-k8s-plugins](https://github.com/opnfv/ovn4nfv-k8s-plugin) is the network controller, OVS agent and CNI server to offer basic SFC and OVN overlay networking.
- [weave](docs/weave.md): Weave is a lightweight container overlay network that doesn't require an external K/V database cluster.
(Please refer to `weave` [troubleshooting documentation](https://www.weave.works/docs/net/latest/troubleshooting/)).
@@ -214,9 +204,9 @@ See also [Network checker](docs/netcheck.md).
## Ingress Plugins
- [nginx](https://kubernetes.github.io/ingress-nginx): the NGINX Ingress Controller.
- [ambassador](docs/ambassador.md): the Ambassador Ingress Controller and API gateway.
- [metallb](docs/metallb.md): the MetalLB bare-metal service LoadBalancer provider.
- [nginx](https://kubernetes.github.io/ingress-nginx): the NGINX Ingress Controller.
## Community docs and resources
@@ -234,6 +224,6 @@ See also [Network checker](docs/netcheck.md).
[![Build graphs](https://gitlab.com/kargo-ci/kubernetes-sigs-kubespray/badges/master/pipeline.svg)](https://gitlab.com/kargo-ci/kubernetes-sigs-kubespray/pipelines)
CI/end-to-end tests sponsored by: [CNCF](https://cncf.io), [Equinix Metal](https://metal.equinix.com/), [OVHcloud](https://www.ovhcloud.com/), [ELASTX](https://elastx.se/).
CI/end-to-end tests sponsored by: [CNCF](https://cncf.io), [Packet](https://www.packet.com/), [OVHcloud](https://www.ovhcloud.com/), [ELASTX](https://elastx.se/).
See the [test matrix](docs/test_cases.md) for details.

66
Vagrantfile vendored
View File

@@ -26,16 +26,12 @@ SUPPORTED_OS = {
"centos-bento" => {box: "bento/centos-7.6", user: "vagrant"},
"centos8" => {box: "centos/8", user: "vagrant"},
"centos8-bento" => {box: "bento/centos-8", user: "vagrant"},
"almalinux8" => {box: "almalinux/8", user: "vagrant"},
"almalinux8-bento" => {box: "bento/almalinux-8", user: "vagrant"},
"fedora34" => {box: "fedora/34-cloud-base", user: "vagrant"},
"fedora35" => {box: "fedora/35-cloud-base", user: "vagrant"},
"opensuse" => {box: "bento/opensuse-leap-15.2", user: "vagrant"},
"fedora31" => {box: "fedora/31-cloud-base", user: "vagrant"},
"fedora32" => {box: "fedora/32-cloud-base", user: "vagrant"},
"opensuse" => {box: "bento/opensuse-leap-15.1", user: "vagrant"},
"opensuse-tumbleweed" => {box: "opensuse/Tumbleweed.x86_64", user: "vagrant"},
"oraclelinux" => {box: "generic/oracle7", user: "vagrant"},
"oraclelinux8" => {box: "generic/oracle8", user: "vagrant"},
"rhel7" => {box: "generic/rhel7", user: "vagrant"},
"rhel8" => {box: "generic/rhel8", user: "vagrant"},
}
if File.exist?(CONFIG)
@@ -51,13 +47,12 @@ $vm_cpus ||= 2
$shared_folders ||= {}
$forwarded_ports ||= {}
$subnet ||= "172.18.8"
$subnet_ipv6 ||= "fd3c:b398:0698:0756"
$os ||= "ubuntu1804"
$network_plugin ||= "flannel"
# Setting multi_networking to true will install Multus: https://github.com/intel/multus-cni
$multi_networking ||= false
$download_run_once ||= "True"
$download_force_cache ||= "False"
$download_force_cache ||= "True"
# The first three nodes are etcd servers
$etcd_instances ||= $num_instances
# The first two nodes are kube masters
@@ -88,16 +83,16 @@ $inventory = File.absolute_path($inventory, File.dirname(__FILE__))
if ! File.exist?(File.join(File.dirname($inventory), "hosts.ini"))
$vagrant_ansible = File.join(File.dirname(__FILE__), ".vagrant", "provisioners", "ansible")
FileUtils.mkdir_p($vagrant_ansible) if ! File.exist?($vagrant_ansible)
$vagrant_inventory = File.join($vagrant_ansible,"inventory")
FileUtils.rm_f($vagrant_inventory)
FileUtils.ln_s($inventory, $vagrant_inventory)
if ! File.exist?(File.join($vagrant_ansible,"inventory"))
FileUtils.ln_s($inventory, File.join($vagrant_ansible,"inventory"))
end
end
if Vagrant.has_plugin?("vagrant-proxyconf")
$no_proxy = ENV['NO_PROXY'] || ENV['no_proxy'] || "127.0.0.1,localhost"
(1..$num_instances).each do |i|
$no_proxy += ",#{$subnet}.#{i+100}"
end
$no_proxy = ENV['NO_PROXY'] || ENV['no_proxy'] || "127.0.0.1,localhost"
(1..$num_instances).each do |i|
$no_proxy += ",#{$subnet}.#{i+100}"
end
end
Vagrant.configure("2") do |config|
@@ -147,7 +142,6 @@ Vagrant.configure("2") do |config|
vb.gui = $vm_gui
vb.linked_clone = true
vb.customize ["modifyvm", :id, "--vram", "8"] # ubuntu defaults to 256 MB which is a waste of precious RAM
vb.customize ["modifyvm", :id, "--audio", "none"]
end
node.vm.provider :libvirt do |lv|
@@ -182,39 +176,19 @@ Vagrant.configure("2") do |config|
node.vm.network "forwarded_port", guest: guest, host: host, auto_correct: true
end
if ["rhel7","rhel8"].include? $os
# Vagrant synced_folder rsync options cannot be used for RHEL boxes as Rsync package cannot
# be installed until the host is registered with a valid Red Hat support subscription
node.vm.synced_folder ".", "/vagrant", disabled: false
$shared_folders.each do |src, dst|
node.vm.synced_folder src, dst
end
else
node.vm.synced_folder ".", "/vagrant", disabled: false, type: "rsync", rsync__args: ['--verbose', '--archive', '--delete', '-z'] , rsync__exclude: ['.git','venv']
$shared_folders.each do |src, dst|
node.vm.synced_folder src, dst, type: "rsync", rsync__args: ['--verbose', '--archive', '--delete', '-z']
end
node.vm.synced_folder ".", "/vagrant", disabled: false, type: "rsync", rsync__args: ['--verbose', '--archive', '--delete', '-z'] , rsync__exclude: ['.git','venv']
$shared_folders.each do |src, dst|
node.vm.synced_folder src, dst, type: "rsync", rsync__args: ['--verbose', '--archive', '--delete', '-z']
end
ip = "#{$subnet}.#{i+100}"
node.vm.network :private_network, ip: ip,
:libvirt__guest_ipv6 => 'yes',
:libvirt__ipv6_address => "#{$subnet_ipv6}::#{i+100}",
:libvirt__ipv6_prefix => "64",
:libvirt__forward_mode => "none",
:libvirt__dhcp_enabled => false
node.vm.network :private_network, ip: ip
# Disable swap for each vm
node.vm.provision "shell", inline: "swapoff -a"
# ubuntu1804 and ubuntu2004 have IPv6 explicitly disabled. This undoes that.
if ["ubuntu1804", "ubuntu2004"].include? $os
node.vm.provision "shell", inline: "rm -f /etc/modprobe.d/local.conf"
node.vm.provision "shell", inline: "sed -i '/net.ipv6.conf.all.disable_ipv6/d' /etc/sysctl.d/99-sysctl.conf /etc/sysctl.conf"
end
# Disable firewalld on oraclelinux/redhat vms
if ["oraclelinux","oraclelinux8","rhel7","rhel8"].include? $os
# Disable firewalld on oraclelinux vms
if ["oraclelinux","oraclelinux8"].include? $os
node.vm.provision "shell", inline: "systemctl stop firewalld; systemctl disable firewalld"
end
@@ -255,9 +229,9 @@ Vagrant.configure("2") do |config|
#ansible.tags = ['download']
ansible.groups = {
"etcd" => ["#{$instance_name_prefix}-[1:#{$etcd_instances}]"],
"kube_control_plane" => ["#{$instance_name_prefix}-[1:#{$kube_master_instances}]"],
"kube_node" => ["#{$instance_name_prefix}-[1:#{$kube_node_instances}]"],
"k8s_cluster:children" => ["kube_control_plane", "kube_node"],
"kube-master" => ["#{$instance_name_prefix}-[1:#{$kube_master_instances}]"],
"kube-node" => ["#{$instance_name_prefix}-[1:#{$kube_node_instances}]"],
"k8s-cluster:children" => ["kube-master", "kube-node"],
}
end
end

View File

@@ -3,6 +3,7 @@ pipelining=True
ssh_args = -o ControlMaster=auto -o ControlPersist=30m -o ConnectionAttempts=100 -o UserKnownHostsFile=/dev/null
#control_path = ~/.ssh/ansible-%%r@%%h:%%p
[defaults]
strategy_plugins = plugins/mitogen/ansible_mitogen/plugins/strategy
# https://github.com/ansible/ansible/issues/56930 (to ignore group names with - and .)
force_valid_group_names = ignore

View File

@@ -3,43 +3,13 @@
gather_facts: false
become: no
vars:
minimal_ansible_version: 2.9.0
minimal_ansible_version_2_10: 2.10.11
maximal_ansible_version: 2.12.0
minimal_ansible_version: 2.8.0
ansible_connection: local
tags: always
tasks:
- name: "Check {{ minimal_ansible_version }} <= Ansible version < {{ maximal_ansible_version }}"
- name: "Check ansible version >={{ minimal_ansible_version }}"
assert:
msg: "Ansible must be between {{ minimal_ansible_version }} and {{ maximal_ansible_version }}"
msg: "Ansible must be {{ minimal_ansible_version }} or higher"
that:
- ansible_version.string is version(minimal_ansible_version, ">=")
- ansible_version.string is version(maximal_ansible_version, "<")
tags:
- check
- name: "Check Ansible version > {{ minimal_ansible_version_2_10 }} when using ansible 2.10"
assert:
msg: "When using Ansible 2.10, the minimum supported version is {{ minimal_ansible_version_2_10 }}"
that:
- ansible_version.string is version(minimal_ansible_version_2_10, ">=")
- ansible_version.string is version(maximal_ansible_version, "<")
when:
- ansible_version.string is version('2.10.0', ">=")
tags:
- check
- name: "Check that python netaddr is installed"
assert:
msg: "Python netaddr is not present"
that: "'127.0.0.1' | ipaddr"
tags:
- check
# CentOS 7 provides too old jinja version
- name: "Check that jinja is not too old (install via pip)"
assert:
msg: "Your Jinja version is too old, install via pip"
that: "{% set test %}It works{% endset %}{{ test == 'It works' }}"
tags:
- check

View File

@@ -2,21 +2,31 @@
- name: Check ansible version
import_playbook: ansible_version.yml
- name: Ensure compatibility with old groups
import_playbook: legacy_groups.yml
- hosts: all
gather_facts: false
tags: always
tasks:
- name: "Set up proxy environment"
set_fact:
proxy_env:
http_proxy: "{{ http_proxy | default ('') }}"
HTTP_PROXY: "{{ http_proxy | default ('') }}"
https_proxy: "{{ https_proxy | default ('') }}"
HTTPS_PROXY: "{{ https_proxy | default ('') }}"
no_proxy: "{{ no_proxy | default ('') }}"
NO_PROXY: "{{ no_proxy | default ('') }}"
no_log: true
- hosts: bastion[0]
gather_facts: False
environment: "{{ proxy_disable_env }}"
roles:
- { role: kubespray-defaults }
- { role: bastion-ssh-config, tags: ["localhost", "bastion"] }
- hosts: k8s_cluster:etcd
- hosts: k8s-cluster:etcd
strategy: linear
any_errors_fatal: "{{ any_errors_fatal | default(true) }}"
gather_facts: false
environment: "{{ proxy_disable_env }}"
roles:
- { role: kubespray-defaults }
- { role: bootstrap-os, tags: bootstrap-os}
@@ -25,20 +35,19 @@
tags: always
import_playbook: facts.yml
- hosts: k8s_cluster:etcd
- hosts: k8s-cluster:etcd
gather_facts: False
any_errors_fatal: "{{ any_errors_fatal | default(true) }}"
environment: "{{ proxy_disable_env }}"
roles:
- { role: kubespray-defaults }
- { role: kubernetes/preinstall, tags: preinstall }
- { role: "container-engine", tags: "container-engine", when: deploy_container_engine }
- { role: "container-engine", tags: "container-engine", when: deploy_container_engine|default(true) }
- { role: download, tags: download, when: "not skip_downloads" }
environment: "{{ proxy_env }}"
- hosts: etcd
gather_facts: False
any_errors_fatal: "{{ any_errors_fatal | default(true) }}"
environment: "{{ proxy_disable_env }}"
roles:
- { role: kubespray-defaults }
- role: etcd
@@ -48,10 +57,9 @@
etcd_events_cluster_setup: "{{ etcd_events_cluster_enabled }}"
when: not etcd_kubeadm_enabled| default(false)
- hosts: k8s_cluster
- hosts: k8s-cluster
gather_facts: False
any_errors_fatal: "{{ any_errors_fatal | default(true) }}"
environment: "{{ proxy_disable_env }}"
roles:
- { role: kubespray-defaults }
- role: etcd
@@ -61,54 +69,50 @@
etcd_events_cluster_setup: false
when: not etcd_kubeadm_enabled| default(false)
- hosts: k8s_cluster
- hosts: k8s-cluster
gather_facts: False
any_errors_fatal: "{{ any_errors_fatal | default(true) }}"
environment: "{{ proxy_disable_env }}"
roles:
- { role: kubespray-defaults }
- { role: kubernetes/node, tags: node }
environment: "{{ proxy_env }}"
- hosts: kube_control_plane
- hosts: kube-master
gather_facts: False
any_errors_fatal: "{{ any_errors_fatal | default(true) }}"
environment: "{{ proxy_disable_env }}"
roles:
- { role: kubespray-defaults }
- { role: kubernetes/control-plane, tags: master }
- { role: kubernetes/master, tags: master }
- { role: kubernetes/client, tags: client }
- { role: kubernetes-apps/cluster_roles, tags: cluster-roles }
- hosts: k8s_cluster
- hosts: k8s-cluster
gather_facts: False
any_errors_fatal: "{{ any_errors_fatal | default(true) }}"
environment: "{{ proxy_disable_env }}"
roles:
- { role: kubespray-defaults }
- { role: kubernetes/kubeadm, tags: kubeadm}
- { role: kubernetes/node-label, tags: node-label }
- { role: network_plugin, tags: network }
- { role: kubernetes/node-label, tags: node-label }
- hosts: calico_rr
- hosts: calico-rr
gather_facts: False
any_errors_fatal: "{{ any_errors_fatal | default(true) }}"
environment: "{{ proxy_disable_env }}"
roles:
- { role: kubespray-defaults }
- { role: network_plugin/calico/rr, tags: ['network', 'calico_rr'] }
- hosts: kube_control_plane[0]
- hosts: kube-master[0]
gather_facts: False
any_errors_fatal: "{{ any_errors_fatal | default(true) }}"
environment: "{{ proxy_disable_env }}"
roles:
- { role: kubespray-defaults }
- { role: kubernetes-apps/rotate_tokens, tags: rotate_tokens, when: "secret_changed|default(false)" }
- { role: win_nodes/kubernetes_patch, tags: ["master", "win_nodes"] }
- hosts: kube_control_plane
- hosts: kube-master
gather_facts: False
any_errors_fatal: "{{ any_errors_fatal | default(true) }}"
environment: "{{ proxy_disable_env }}"
roles:
- { role: kubespray-defaults }
- { role: kubernetes-apps/external_cloud_controller, tags: external-cloud-controller }
@@ -116,12 +120,18 @@
- { role: kubernetes-apps/policy_controller, tags: policy-controller }
- { role: kubernetes-apps/ingress_controller, tags: ingress-controller }
- { role: kubernetes-apps/external_provisioner, tags: external-provisioner }
- { role: kubernetes-apps, tags: apps }
- hosts: k8s_cluster
- hosts: kube-master
gather_facts: False
any_errors_fatal: "{{ any_errors_fatal | default(true) }}"
roles:
- { role: kubespray-defaults }
- { role: kubernetes-apps, tags: apps }
environment: "{{ proxy_env }}"
- hosts: k8s-cluster
gather_facts: False
any_errors_fatal: "{{ any_errors_fatal | default(true) }}"
environment: "{{ proxy_disable_env }}"
roles:
- { role: kubespray-defaults }
- { role: kubernetes/preinstall, when: "dns_mode != 'none' and resolvconf_mode == 'host_resolvconf'", tags: resolvconf, dns_late: true }

View File

@@ -35,7 +35,7 @@ class SearchEC2Tags(object):
hosts['_meta'] = { 'hostvars': {} }
##Search ec2 three times to find nodes of each group type. Relies on kubespray-role key/value.
for group in ["kube_control_plane", "kube_node", "etcd"]:
for group in ["kube-master", "kube-node", "etcd"]:
hosts[group] = []
tag_key = "kubespray-role"
tag_value = ["*"+group+"*"]
@@ -69,8 +69,8 @@ class SearchEC2Tags(object):
hosts[group].append(dns_name)
hosts['_meta']['hostvars'][dns_name] = ansible_host
hosts['k8s_cluster'] = {'children':['kube_control_plane', 'kube_node']}
hosts['k8s-cluster'] = {'children':['kube-master', 'kube-node']}
print(json.dumps(hosts, sort_keys=True, indent=2))
SearchEC2Tags()

View File

@@ -1 +0,0 @@
boto3 # Apache-2.0

View File

@@ -24,14 +24,14 @@ experience.
You can enable the use of a Bastion Host by changing **use_bastion** in group_vars/all to **true**. The generated
templates will then include an additional bastion VM which can then be used to connect to the masters and nodes. The option
also removes all public IPs from all other VMs.
also removes all public IPs from all other VMs.
## Generating and applying
To generate and apply the templates, call:
```shell
./apply-rg.sh <resource_group_name>
$ ./apply-rg.sh <resource_group_name>
```
If you change something in the configuration (e.g. number of nodes) later, you can call this again and Azure will
@@ -42,23 +42,25 @@ take care about creating/modifying whatever is needed.
If you need to delete all resources from a resource group, simply call:
```shell
./clear-rg.sh <resource_group_name>
$ ./clear-rg.sh <resource_group_name>
```
**WARNING** this really deletes everything from your resource group, including everything that was later created by you!
## Generating an inventory for kubespray
After you have applied the templates, you can generate an inventory with this call:
```shell
./generate-inventory.sh <resource_group_name>
$ ./generate-inventory.sh <resource_group_name>
```
It will create the file ./inventory which can then be used with kubespray, e.g.:
```shell
cd kubespray-root-dir
sudo pip3 install -r requirements.txt
ansible-playbook -i contrib/azurerm/inventory -u devops --become -e "@inventory/sample/group_vars/all/all.yml" cluster.yml
$ cd kubespray-root-dir
$ sudo pip3 install -r requirements.txt
$ ansible-playbook -i contrib/azurerm/inventory -u devops --become -e "@inventory/sample/group_vars/all/all.yml" cluster.yml
```

View File

@@ -12,4 +12,3 @@
template:
src: inventory.j2
dest: "{{ playbook_dir }}/inventory"
mode: 0644

View File

@@ -7,9 +7,9 @@
{% endif %}
{% endfor %}
[kube_control_plane]
[kube-master]
{% for vm in vm_list %}
{% if 'kube_control_plane' in vm.tags.roles %}
{% if 'kube-master' in vm.tags.roles %}
{{ vm.name }}
{% endif %}
{% endfor %}
@@ -21,13 +21,13 @@
{% endif %}
{% endfor %}
[kube_node]
[kube-node]
{% for vm in vm_list %}
{% if 'kube_node' in vm.tags.roles %}
{% if 'kube-node' in vm.tags.roles %}
{{ vm.name }}
{% endif %}
{% endfor %}
[k8s_cluster:children]
kube_node
kube_control_plane
[k8s-cluster:children]
kube-node
kube-master

View File

@@ -22,10 +22,8 @@
template:
src: inventory.j2
dest: "{{ playbook_dir }}/inventory"
mode: 0644
- name: Generate Load Balancer variables
template:
src: loadbalancer_vars.j2
dest: "{{ playbook_dir }}/loadbalancer_vars.yml"
mode: 0644

View File

@@ -7,9 +7,9 @@
{% endif %}
{% endfor %}
[kube_control_plane]
[kube-master]
{% for vm in vm_roles_list %}
{% if 'kube_control_plane' in vm.tags.roles %}
{% if 'kube-master' in vm.tags.roles %}
{{ vm.name }}
{% endif %}
{% endfor %}
@@ -21,14 +21,14 @@
{% endif %}
{% endfor %}
[kube_node]
[kube-node]
{% for vm in vm_roles_list %}
{% if 'kube_node' in vm.tags.roles %}
{% if 'kube-node' in vm.tags.roles %}
{{ vm.name }}
{% endif %}
{% endfor %}
[k8s_cluster:children]
kube_node
kube_control_plane
[k8s-cluster:children]
kube-node
kube-master

View File

@@ -8,13 +8,11 @@
path: "{{ base_dir }}"
state: directory
recurse: true
mode: 0755
- name: Store json files in base_dir
template:
src: "{{ item }}"
dest: "{{ base_dir }}/{{ item }}"
mode: 0644
with_items:
- network.json
- storage.json

View File

@@ -144,7 +144,7 @@
"[concat('Microsoft.Network/networkInterfaces/', 'master-{{i}}-nic')]"
],
"tags": {
"roles": "kube_control_plane,etcd"
"roles": "kube-master,etcd"
},
"apiVersion": "{{apiVersion}}",
"properties": {

View File

@@ -61,7 +61,7 @@
"[concat('Microsoft.Network/networkInterfaces/', 'minion-{{i}}-nic')]"
],
"tags": {
"roles": "kube_node"
"roles": "kube-node"
},
"apiVersion": "{{apiVersion}}",
"properties": {
@@ -112,4 +112,4 @@
} {% if not loop.last %},{% endif %}
{% endfor %}
]
}
}

View File

@@ -6,7 +6,6 @@ to serve as Kubernetes "nodes", which in turn will run
called DIND (Docker-IN-Docker).
The playbook has two roles:
- dind-host: creates the "nodes" as containers in localhost, with
appropriate settings for DIND (privileged, volume mapping for dind
storage, etc).
@@ -28,7 +27,7 @@ See below for a complete successful run:
1. Create the node containers
```shell
~~~~
# From the kubespray root dir
cd contrib/dind
pip install -r requirements.txt
@@ -37,15 +36,15 @@ ansible-playbook -i hosts dind-cluster.yaml
# Back to kubespray root
cd ../..
```
~~~~
NOTE: if the playbook run fails with something like below error
message, you may need to specifically set `ansible_python_interpreter`,
see `./hosts` file for an example expanded localhost entry.
```shell
~~~
failed: [localhost] (item=kube-node1) => {"changed": false, "item": "kube-node1", "msg": "Failed to import docker or docker-py - No module named requests.exceptions. Try `pip install docker` or `pip install docker-py` (Python 2.6)"}
```
~~~
2. Customize kubespray-dind.yaml
@@ -53,33 +52,33 @@ Note that there's coupling between above created node containers
and `kubespray-dind.yaml` settings, in particular regarding selected `node_distro`
(as set in `group_vars/all/all.yaml`), and docker settings.
```shell
~~~
$EDITOR contrib/dind/kubespray-dind.yaml
```
~~~
3. Prepare the inventory and run the playbook
```shell
~~~
INVENTORY_DIR=inventory/local-dind
mkdir -p ${INVENTORY_DIR}
rm -f ${INVENTORY_DIR}/hosts.ini
CONFIG_FILE=${INVENTORY_DIR}/hosts.ini /tmp/kubespray.dind.inventory_builder.sh
ansible-playbook --become -e ansible_ssh_user=debian -i ${INVENTORY_DIR}/hosts.ini cluster.yml --extra-vars @contrib/dind/kubespray-dind.yaml
```
~~~
NOTE: You could also test other distros without editing files by
passing `--extra-vars` as per below commandline,
replacing `DISTRO` by either `debian`, `ubuntu`, `centos`, `fedora`:
```shell
~~~
cd contrib/dind
ansible-playbook -i hosts dind-cluster.yaml --extra-vars node_distro=DISTRO
cd ../..
CONFIG_FILE=inventory/local-dind/hosts.ini /tmp/kubespray.dind.inventory_builder.sh
ansible-playbook --become -e ansible_ssh_user=DISTRO -i inventory/local-dind/hosts.ini cluster.yml --extra-vars @contrib/dind/kubespray-dind.yaml --extra-vars bootstrap_os=DISTRO
```
~~~
## Resulting deployment
@@ -90,7 +89,7 @@ from the host where you ran kubespray playbooks.
Running from an Ubuntu Xenial host:
```shell
~~~
$ uname -a
Linux ip-xx-xx-xx-xx 4.4.0-1069-aws #79-Ubuntu SMP Mon Sep 24
15:01:41 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
@@ -150,14 +149,14 @@ kube-system weave-net-xr46t 2/2 Running 0
$ docker exec kube-node1 curl -s http://localhost:31081/api/v1/connectivity_check
{"Message":"All 10 pods successfully reported back to the server","Absent":null,"Outdated":null}
```
~~~
## Using ./run-test-distros.sh
You can use `./run-test-distros.sh` to run a set of tests via DIND,
and excerpt from this script, to get an idea:
```shell
~~~
# The SPEC file(s) must have two arrays as e.g.
# DISTROS=(debian centos)
# EXTRAS=(
@@ -170,7 +169,7 @@ and excerpt from this script, to get an idea:
#
# Each $EXTRAS element will be whitespace split, and passed as --extra-vars
# to main kubespray ansible-playbook run.
```
~~~
See e.g. `test-some_distros-most_CNIs.env` and
`test-some_distros-kube_router_combo.env` in particular for a richer

View File

@@ -35,7 +35,6 @@
path-exclude=/usr/share/doc/*
path-include=/usr/share/doc/*/copyright
dest: /etc/dpkg/dpkg.cfg.d/01_nodoc
mode: 0644
when:
- ansible_os_family == 'Debian'
@@ -64,7 +63,6 @@
copy:
content: "{{ distro_user }} ALL=(ALL) NOPASSWD:ALL"
dest: "/etc/sudoers.d/{{ distro_user }}"
mode: 0640
- name: Add my pubkey to "{{ distro_user }}" user authorized keys
authorized_key:

View File

@@ -46,7 +46,7 @@ test_distro() {
pass_or_fail "$prefix: netcheck" || return 1
}
NODES=($(egrep ^kube_node hosts))
NODES=($(egrep ^kube-node hosts))
NETCHECKER_HOST=localhost
: ${OUTPUT_DIR:=./out}

View File

@@ -44,11 +44,11 @@ import re
import subprocess
import sys
ROLES = ['all', 'kube_control_plane', 'kube_node', 'etcd', 'k8s_cluster',
'calico_rr']
ROLES = ['all', 'kube-master', 'kube-node', 'etcd', 'k8s-cluster',
'calico-rr']
PROTECTED_NAMES = ROLES
AVAILABLE_COMMANDS = ['help', 'print_cfg', 'print_ips', 'print_hostnames',
'load', 'add']
'load']
_boolean_states = {'1': True, 'yes': True, 'true': True, 'on': True,
'0': False, 'no': False, 'false': False, 'off': False}
yaml = YAML()
@@ -63,12 +63,10 @@ def get_var_as_bool(name, default):
CONFIG_FILE = os.environ.get("CONFIG_FILE", "./inventory/sample/hosts.yaml")
# Remove the reference of KUBE_MASTERS after some deprecation cycles.
KUBE_CONTROL_HOSTS = int(os.environ.get("KUBE_CONTROL_HOSTS",
os.environ.get("KUBE_MASTERS", 2)))
KUBE_MASTERS = int(os.environ.get("KUBE_MASTERS_MASTERS", 2))
# Reconfigures cluster distribution at scale
SCALE_THRESHOLD = int(os.environ.get("SCALE_THRESHOLD", 50))
MASSIVE_SCALE_THRESHOLD = int(os.environ.get("MASSIVE_SCALE_THRESHOLD", 200))
MASSIVE_SCALE_THRESHOLD = int(os.environ.get("SCALE_THRESHOLD", 200))
DEBUG = get_var_as_bool("DEBUG", True)
HOST_PREFIX = os.environ.get("HOST_PREFIX", "node")
@@ -82,46 +80,32 @@ class KubesprayInventory(object):
def __init__(self, changed_hosts=None, config_file=None):
self.config_file = config_file
self.yaml_config = {}
loadPreviousConfig = False
# See whether there are any commands to process
if changed_hosts and changed_hosts[0] in AVAILABLE_COMMANDS:
if changed_hosts[0] == "add":
loadPreviousConfig = True
changed_hosts = changed_hosts[1:]
else:
self.parse_command(changed_hosts[0], changed_hosts[1:])
sys.exit(0)
# If the user wants to remove a node, we need to load the config anyway
if changed_hosts and changed_hosts[0][0] == "-":
loadPreviousConfig = True
if self.config_file and loadPreviousConfig: # Load previous YAML file
if self.config_file:
try:
self.hosts_file = open(config_file, 'r')
self.yaml_config = yaml.load(self.hosts_file)
except OSError as e:
# I am assuming we are catching "cannot open file" exceptions
print(e)
sys.exit(1)
self.yaml_config = yaml.load_all(self.hosts_file)
except OSError:
pass
if changed_hosts and changed_hosts[0] in AVAILABLE_COMMANDS:
self.parse_command(changed_hosts[0], changed_hosts[1:])
sys.exit(0)
self.ensure_required_groups(ROLES)
if changed_hosts:
changed_hosts = self.range2ips(changed_hosts)
self.hosts = self.build_hostnames(changed_hosts,
loadPreviousConfig)
self.hosts = self.build_hostnames(changed_hosts)
self.purge_invalid_hosts(self.hosts.keys(), PROTECTED_NAMES)
self.set_all(self.hosts)
self.set_k8s_cluster()
etcd_hosts_count = 3 if len(self.hosts.keys()) >= 3 else 1
self.set_etcd(list(self.hosts.keys())[:etcd_hosts_count])
if len(self.hosts) >= SCALE_THRESHOLD:
self.set_kube_control_plane(list(self.hosts.keys())[
etcd_hosts_count:(etcd_hosts_count + KUBE_CONTROL_HOSTS)])
self.set_kube_master(list(self.hosts.keys())[
etcd_hosts_count:(etcd_hosts_count + KUBE_MASTERS)])
else:
self.set_kube_control_plane(
list(self.hosts.keys())[:KUBE_CONTROL_HOSTS])
self.set_kube_master(list(self.hosts.keys())[:KUBE_MASTERS])
self.set_kube_node(self.hosts.keys())
if len(self.hosts) >= SCALE_THRESHOLD:
self.set_calico_rr(list(self.hosts.keys())[:etcd_hosts_count])
@@ -171,29 +155,17 @@ class KubesprayInventory(object):
except IndexError:
raise ValueError("Host name must end in an integer")
# Keeps already specified hosts,
# and adds or removes the hosts provided as an argument
def build_hostnames(self, changed_hosts, loadPreviousConfig=False):
def build_hostnames(self, changed_hosts):
existing_hosts = OrderedDict()
highest_host_id = 0
# Load already existing hosts from the YAML
if loadPreviousConfig:
try:
for host in self.yaml_config['all']['hosts']:
# Read configuration of an existing host
hostConfig = self.yaml_config['all']['hosts'][host]
existing_hosts[host] = hostConfig
# If the existing host seems
# to have been created automatically, detect its ID
if host.startswith(HOST_PREFIX):
host_id = self.get_host_id(host)
if host_id > highest_host_id:
highest_host_id = host_id
except Exception as e:
# I am assuming we are catching automatically
# created hosts without IDs
print(e)
sys.exit(1)
try:
for host in self.yaml_config['all']['hosts']:
existing_hosts[host] = self.yaml_config['all']['hosts'][host]
host_id = self.get_host_id(host)
if host_id > highest_host_id:
highest_host_id = host_id
except Exception:
pass
# FIXME(mattymo): Fix condition where delete then add reuses highest id
next_host_id = highest_host_id + 1
@@ -201,7 +173,6 @@ class KubesprayInventory(object):
all_hosts = existing_hosts.copy()
for host in changed_hosts:
# Delete the host from config the hostname/IP has a "-" prefix
if host[0] == "-":
realhost = host[1:]
if self.exists_hostname(all_hosts, realhost):
@@ -210,8 +181,6 @@ class KubesprayInventory(object):
elif self.exists_ip(all_hosts, realhost):
self.debug("Marked {0} for deletion.".format(realhost))
self.delete_host_by_ip(all_hosts, realhost)
# Host/Argument starts with a digit,
# then we assume its an IP address
elif host[0].isdigit():
if ',' in host:
ip, access_ip = host.split(',')
@@ -231,15 +200,11 @@ class KubesprayInventory(object):
next_host = subprocess.check_output(cmd, shell=True)
next_host = next_host.strip().decode('ascii')
else:
# Generates a hostname because we have only an IP address
next_host = "{0}{1}".format(HOST_PREFIX, next_host_id)
next_host_id += 1
# Uses automatically generated node name
# in case we dont provide it.
all_hosts[next_host] = {'ansible_host': access_ip,
'ip': ip,
'access_ip': access_ip}
# Host/Argument starts with a letter, then we assume its a hostname
elif host[0].isalpha():
if ',' in host:
try:
@@ -258,7 +223,6 @@ class KubesprayInventory(object):
'access_ip': access_ip}
return all_hosts
# Expand IP ranges into individual addresses
def range2ips(self, hosts):
reworked_hosts = []
@@ -302,7 +266,7 @@ class KubesprayInventory(object):
def purge_invalid_hosts(self, hostnames, protected_names=[]):
for role in self.yaml_config['all']['children']:
if role != 'k8s_cluster' and self.yaml_config['all']['children'][role]['hosts']: # noqa
if role != 'k8s-cluster' and self.yaml_config['all']['children'][role]['hosts']: # noqa
all_hosts = self.yaml_config['all']['children'][role]['hosts'].copy() # noqa
for host in all_hosts.keys():
if host not in hostnames and host not in protected_names:
@@ -323,54 +287,52 @@ class KubesprayInventory(object):
if self.yaml_config['all']['hosts'] is None:
self.yaml_config['all']['hosts'] = {host: None}
self.yaml_config['all']['hosts'][host] = opts
elif group != 'k8s_cluster:children':
elif group != 'k8s-cluster:children':
if self.yaml_config['all']['children'][group]['hosts'] is None:
self.yaml_config['all']['children'][group]['hosts'] = {
host: None}
else:
self.yaml_config['all']['children'][group]['hosts'][host] = None # noqa
def set_kube_control_plane(self, hosts):
def set_kube_master(self, hosts):
for host in hosts:
self.add_host_to_group('kube_control_plane', host)
self.add_host_to_group('kube-master', host)
def set_all(self, hosts):
for host, opts in hosts.items():
self.add_host_to_group('all', host, opts)
def set_k8s_cluster(self):
k8s_cluster = {'children': {'kube_control_plane': None,
'kube_node': None}}
self.yaml_config['all']['children']['k8s_cluster'] = k8s_cluster
k8s_cluster = {'children': {'kube-master': None, 'kube-node': None}}
self.yaml_config['all']['children']['k8s-cluster'] = k8s_cluster
def set_calico_rr(self, hosts):
for host in hosts:
if host in self.yaml_config['all']['children']['kube_control_plane']: # noqa
self.debug("Not adding {0} to calico_rr group because it "
"conflicts with kube_control_plane "
"group".format(host))
if host in self.yaml_config['all']['children']['kube-master']:
self.debug("Not adding {0} to calico-rr group because it "
"conflicts with kube-master group".format(host))
continue
if host in self.yaml_config['all']['children']['kube_node']:
self.debug("Not adding {0} to calico_rr group because it "
"conflicts with kube_node group".format(host))
if host in self.yaml_config['all']['children']['kube-node']:
self.debug("Not adding {0} to calico-rr group because it "
"conflicts with kube-node group".format(host))
continue
self.add_host_to_group('calico_rr', host)
self.add_host_to_group('calico-rr', host)
def set_kube_node(self, hosts):
for host in hosts:
if len(self.yaml_config['all']['hosts']) >= SCALE_THRESHOLD:
if host in self.yaml_config['all']['children']['etcd']['hosts']: # noqa
self.debug("Not adding {0} to kube_node group because of "
self.debug("Not adding {0} to kube-node group because of "
"scale deployment and host is in etcd "
"group.".format(host))
continue
if len(self.yaml_config['all']['hosts']) >= MASSIVE_SCALE_THRESHOLD: # noqa
if host in self.yaml_config['all']['children']['kube_control_plane']['hosts']: # noqa
self.debug("Not adding {0} to kube_node group because of "
"scale deployment and host is in "
"kube_control_plane group.".format(host))
if host in self.yaml_config['all']['children']['kube-master']['hosts']: # noqa
self.debug("Not adding {0} to kube-node group because of "
"scale deployment and host is in kube-master "
"group.".format(host))
continue
self.add_host_to_group('kube_node', host)
self.add_host_to_group('kube-node', host)
def set_etcd(self, hosts):
for host in hosts:
@@ -427,11 +389,9 @@ help - Display this message
print_cfg - Write inventory file to stdout
print_ips - Write a space-delimited list of IPs from "all" group
print_hostnames - Write a space-delimited list of Hostnames from "all" group
add - Adds specified hosts into an already existing inventory
Advanced usage:
Create new or overwrite old inventory file: inventory.py 10.10.1.5
Add another host after initial creation: inventory.py add 10.10.1.6
Add another host after initial creation: inventory.py 10.10.1.5
Add range of hosts: inventory.py 10.10.1.3-10.10.1.5
Add hosts with different ip and access ip: inventory.py 10.0.0.1,192.168.10.1 10.0.0.2,192.168.10.2 10.0.0.3,192.168.10.3
Add hosts with a specific hostname, ip, and optional access ip: first,10.0.0.1,192.168.10.1 second,10.0.0.2 last,10.0.0.3
@@ -442,9 +402,8 @@ Configurable env vars:
DEBUG Enable debug printing. Default: True
CONFIG_FILE File to write config to Default: ./inventory/sample/hosts.yaml
HOST_PREFIX Host prefix for generated hosts. Default: node
KUBE_CONTROL_HOSTS Set the number of kube-control-planes. Default: 2
SCALE_THRESHOLD Separate ETCD role if # of nodes >= 50
MASSIVE_SCALE_THRESHOLD Separate K8s control-plane and ETCD if # of nodes >= 200
MASSIVE_SCALE_THRESHOLD Separate K8s master and ETCD if # of nodes >= 200
''' # noqa
print(help_text)
@@ -465,7 +424,6 @@ def main(argv=None):
if not argv:
argv = sys.argv[1:]
KubesprayInventory(argv, CONFIG_FILE)
return 0
if __name__ == "__main__":

View File

@@ -13,8 +13,8 @@
# under the License.
import inventory
import mock
import unittest
from unittest import mock
from collections import OrderedDict
import sys
@@ -51,7 +51,7 @@ class TestInventory(unittest.TestCase):
groups = ['group1', 'group2']
self.inv.ensure_required_groups(groups)
for group in groups:
self.assertIn(group, self.inv.yaml_config['all']['children'])
self.assertTrue(group in self.inv.yaml_config['all']['children'])
def test_get_host_id(self):
hostnames = ['node99', 'no99de01', '01node01', 'node1.domain',
@@ -67,14 +67,23 @@ class TestInventory(unittest.TestCase):
self.assertRaisesRegex(ValueError, "Host name must end in an",
self.inv.get_host_id, hostname)
def test_build_hostnames_add_one(self):
changed_hosts = ['10.90.0.2']
expected = OrderedDict([('node1',
{'ansible_host': '10.90.0.2',
'ip': '10.90.0.2',
'access_ip': '10.90.0.2'})])
result = self.inv.build_hostnames(changed_hosts)
self.assertEqual(expected, result)
def test_build_hostnames_add_duplicate(self):
changed_hosts = ['10.90.0.2']
expected = OrderedDict([('node3',
expected = OrderedDict([('node1',
{'ansible_host': '10.90.0.2',
'ip': '10.90.0.2',
'access_ip': '10.90.0.2'})])
self.inv.yaml_config['all']['hosts'] = expected
result = self.inv.build_hostnames(changed_hosts, True)
result = self.inv.build_hostnames(changed_hosts)
self.assertEqual(expected, result)
def test_build_hostnames_add_two(self):
@@ -90,30 +99,6 @@ class TestInventory(unittest.TestCase):
result = self.inv.build_hostnames(changed_hosts)
self.assertEqual(expected, result)
def test_build_hostnames_add_three(self):
changed_hosts = ['10.90.0.2', '10.90.0.3', '10.90.0.4']
expected = OrderedDict([
('node1', {'ansible_host': '10.90.0.2',
'ip': '10.90.0.2',
'access_ip': '10.90.0.2'}),
('node2', {'ansible_host': '10.90.0.3',
'ip': '10.90.0.3',
'access_ip': '10.90.0.3'}),
('node3', {'ansible_host': '10.90.0.4',
'ip': '10.90.0.4',
'access_ip': '10.90.0.4'})])
result = self.inv.build_hostnames(changed_hosts)
self.assertEqual(expected, result)
def test_build_hostnames_add_one(self):
changed_hosts = ['10.90.0.2']
expected = OrderedDict([('node1',
{'ansible_host': '10.90.0.2',
'ip': '10.90.0.2',
'access_ip': '10.90.0.2'})])
result = self.inv.build_hostnames(changed_hosts)
self.assertEqual(expected, result)
def test_build_hostnames_delete_first(self):
changed_hosts = ['-10.90.0.2']
existing_hosts = OrderedDict([
@@ -128,24 +113,7 @@ class TestInventory(unittest.TestCase):
('node2', {'ansible_host': '10.90.0.3',
'ip': '10.90.0.3',
'access_ip': '10.90.0.3'})])
result = self.inv.build_hostnames(changed_hosts, True)
self.assertEqual(expected, result)
def test_build_hostnames_delete_by_hostname(self):
changed_hosts = ['-node1']
existing_hosts = OrderedDict([
('node1', {'ansible_host': '10.90.0.2',
'ip': '10.90.0.2',
'access_ip': '10.90.0.2'}),
('node2', {'ansible_host': '10.90.0.3',
'ip': '10.90.0.3',
'access_ip': '10.90.0.3'})])
self.inv.yaml_config['all']['hosts'] = existing_hosts
expected = OrderedDict([
('node2', {'ansible_host': '10.90.0.3',
'ip': '10.90.0.3',
'access_ip': '10.90.0.3'})])
result = self.inv.build_hostnames(changed_hosts, True)
result = self.inv.build_hostnames(changed_hosts)
self.assertEqual(expected, result)
def test_exists_hostname_positive(self):
@@ -241,8 +209,8 @@ class TestInventory(unittest.TestCase):
('doesnotbelong2', {'whateveropts=ilike'})])
self.inv.yaml_config['all']['hosts'] = existing_hosts
self.inv.purge_invalid_hosts(proper_hostnames)
self.assertNotIn(
bad_host, self.inv.yaml_config['all']['hosts'].keys())
self.assertTrue(
bad_host not in self.inv.yaml_config['all']['hosts'].keys())
def test_add_host_to_group(self):
group = 'etcd'
@@ -254,13 +222,13 @@ class TestInventory(unittest.TestCase):
self.inv.yaml_config['all']['children'][group]['hosts'].get(host),
None)
def test_set_kube_control_plane(self):
group = 'kube_control_plane'
def test_set_kube_master(self):
group = 'kube-master'
host = 'node1'
self.inv.set_kube_control_plane([host])
self.assertIn(
host, self.inv.yaml_config['all']['children'][group]['hosts'])
self.inv.set_kube_master([host])
self.assertTrue(
host in self.inv.yaml_config['all']['children'][group]['hosts'])
def test_set_all(self):
hosts = OrderedDict([
@@ -273,30 +241,30 @@ class TestInventory(unittest.TestCase):
self.inv.yaml_config['all']['hosts'].get(host), opt)
def test_set_k8s_cluster(self):
group = 'k8s_cluster'
expected_hosts = ['kube_node', 'kube_control_plane']
group = 'k8s-cluster'
expected_hosts = ['kube-node', 'kube-master']
self.inv.set_k8s_cluster()
for host in expected_hosts:
self.assertIn(
host,
self.assertTrue(
host in
self.inv.yaml_config['all']['children'][group]['children'])
def test_set_kube_node(self):
group = 'kube_node'
group = 'kube-node'
host = 'node1'
self.inv.set_kube_node([host])
self.assertIn(
host, self.inv.yaml_config['all']['children'][group]['hosts'])
self.assertTrue(
host in self.inv.yaml_config['all']['children'][group]['hosts'])
def test_set_etcd(self):
group = 'etcd'
host = 'node1'
self.inv.set_etcd([host])
self.assertIn(
host, self.inv.yaml_config['all']['children'][group]['hosts'])
self.assertTrue(
host in self.inv.yaml_config['all']['children'][group]['hosts'])
def test_scale_scenario_one(self):
num_nodes = 50
@@ -307,12 +275,12 @@ class TestInventory(unittest.TestCase):
self.inv.set_all(hosts)
self.inv.set_etcd(list(hosts.keys())[0:3])
self.inv.set_kube_control_plane(list(hosts.keys())[0:2])
self.inv.set_kube_master(list(hosts.keys())[0:2])
self.inv.set_kube_node(hosts.keys())
for h in range(3):
self.assertFalse(
list(hosts.keys())[h] in
self.inv.yaml_config['all']['children']['kube_node']['hosts'])
self.inv.yaml_config['all']['children']['kube-node']['hosts'])
def test_scale_scenario_two(self):
num_nodes = 500
@@ -323,12 +291,12 @@ class TestInventory(unittest.TestCase):
self.inv.set_all(hosts)
self.inv.set_etcd(list(hosts.keys())[0:3])
self.inv.set_kube_control_plane(list(hosts.keys())[3:5])
self.inv.set_kube_master(list(hosts.keys())[3:5])
self.inv.set_kube_node(hosts.keys())
for h in range(5):
self.assertFalse(
list(hosts.keys())[h] in
self.inv.yaml_config['all']['children']['kube_node']['hosts'])
self.inv.yaml_config['all']['children']['kube-node']['hosts'])
def test_range2ips_range(self):
changed_hosts = ['10.90.0.2', '10.90.0.4-10.90.0.6', '10.90.0.8']
@@ -345,7 +313,7 @@ class TestInventory(unittest.TestCase):
self.assertRaisesRegex(Exception, "Range of ip_addresses isn't valid",
self.inv.range2ips, host_range)
def test_build_hostnames_create_with_one_different_ips(self):
def test_build_hostnames_different_ips_add_one(self):
changed_hosts = ['10.90.0.2,192.168.0.2']
expected = OrderedDict([('node1',
{'ansible_host': '192.168.0.2',
@@ -354,7 +322,17 @@ class TestInventory(unittest.TestCase):
result = self.inv.build_hostnames(changed_hosts)
self.assertEqual(expected, result)
def test_build_hostnames_create_with_two_different_ips(self):
def test_build_hostnames_different_ips_add_duplicate(self):
changed_hosts = ['10.90.0.2,192.168.0.2']
expected = OrderedDict([('node1',
{'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'})])
self.inv.yaml_config['all']['hosts'] = expected
result = self.inv.build_hostnames(changed_hosts)
self.assertEqual(expected, result)
def test_build_hostnames_different_ips_add_two(self):
changed_hosts = ['10.90.0.2,192.168.0.2', '10.90.0.3,192.168.0.3']
expected = OrderedDict([
('node1', {'ansible_host': '192.168.0.2',
@@ -363,210 +341,6 @@ class TestInventory(unittest.TestCase):
('node2', {'ansible_host': '192.168.0.3',
'ip': '10.90.0.3',
'access_ip': '192.168.0.3'})])
self.inv.yaml_config['all']['hosts'] = OrderedDict()
result = self.inv.build_hostnames(changed_hosts)
self.assertEqual(expected, result)
def test_build_hostnames_create_with_three_different_ips(self):
changed_hosts = ['10.90.0.2,192.168.0.2',
'10.90.0.3,192.168.0.3',
'10.90.0.4,192.168.0.4']
expected = OrderedDict([
('node1', {'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'}),
('node2', {'ansible_host': '192.168.0.3',
'ip': '10.90.0.3',
'access_ip': '192.168.0.3'}),
('node3', {'ansible_host': '192.168.0.4',
'ip': '10.90.0.4',
'access_ip': '192.168.0.4'})])
result = self.inv.build_hostnames(changed_hosts)
self.assertEqual(expected, result)
def test_build_hostnames_overwrite_one_with_different_ips(self):
changed_hosts = ['10.90.0.2,192.168.0.2']
expected = OrderedDict([('node1',
{'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'})])
existing = OrderedDict([('node5',
{'ansible_host': '192.168.0.5',
'ip': '10.90.0.5',
'access_ip': '192.168.0.5'})])
self.inv.yaml_config['all']['hosts'] = existing
result = self.inv.build_hostnames(changed_hosts)
self.assertEqual(expected, result)
def test_build_hostnames_overwrite_three_with_different_ips(self):
changed_hosts = ['10.90.0.2,192.168.0.2']
expected = OrderedDict([('node1',
{'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'})])
existing = OrderedDict([
('node3', {'ansible_host': '192.168.0.3',
'ip': '10.90.0.3',
'access_ip': '192.168.0.3'}),
('node4', {'ansible_host': '192.168.0.4',
'ip': '10.90.0.4',
'access_ip': '192.168.0.4'}),
('node5', {'ansible_host': '192.168.0.5',
'ip': '10.90.0.5',
'access_ip': '192.168.0.5'})])
self.inv.yaml_config['all']['hosts'] = existing
result = self.inv.build_hostnames(changed_hosts)
self.assertEqual(expected, result)
def test_build_hostnames_different_ips_add_duplicate(self):
changed_hosts = ['10.90.0.2,192.168.0.2']
expected = OrderedDict([('node3',
{'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'})])
existing = expected
self.inv.yaml_config['all']['hosts'] = existing
result = self.inv.build_hostnames(changed_hosts, True)
self.assertEqual(expected, result)
def test_build_hostnames_add_two_different_ips_into_one_existing(self):
changed_hosts = ['10.90.0.3,192.168.0.3', '10.90.0.4,192.168.0.4']
expected = OrderedDict([
('node2', {'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'}),
('node3', {'ansible_host': '192.168.0.3',
'ip': '10.90.0.3',
'access_ip': '192.168.0.3'}),
('node4', {'ansible_host': '192.168.0.4',
'ip': '10.90.0.4',
'access_ip': '192.168.0.4'})])
existing = OrderedDict([
('node2', {'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'})])
self.inv.yaml_config['all']['hosts'] = existing
result = self.inv.build_hostnames(changed_hosts, True)
self.assertEqual(expected, result)
def test_build_hostnames_add_two_different_ips_into_two_existing(self):
changed_hosts = ['10.90.0.4,192.168.0.4', '10.90.0.5,192.168.0.5']
expected = OrderedDict([
('node2', {'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'}),
('node3', {'ansible_host': '192.168.0.3',
'ip': '10.90.0.3',
'access_ip': '192.168.0.3'}),
('node4', {'ansible_host': '192.168.0.4',
'ip': '10.90.0.4',
'access_ip': '192.168.0.4'}),
('node5', {'ansible_host': '192.168.0.5',
'ip': '10.90.0.5',
'access_ip': '192.168.0.5'})])
existing = OrderedDict([
('node2', {'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'}),
('node3', {'ansible_host': '192.168.0.3',
'ip': '10.90.0.3',
'access_ip': '192.168.0.3'})])
self.inv.yaml_config['all']['hosts'] = existing
result = self.inv.build_hostnames(changed_hosts, True)
self.assertEqual(expected, result)
def test_build_hostnames_add_two_different_ips_into_three_existing(self):
changed_hosts = ['10.90.0.5,192.168.0.5', '10.90.0.6,192.168.0.6']
expected = OrderedDict([
('node2', {'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'}),
('node3', {'ansible_host': '192.168.0.3',
'ip': '10.90.0.3',
'access_ip': '192.168.0.3'}),
('node4', {'ansible_host': '192.168.0.4',
'ip': '10.90.0.4',
'access_ip': '192.168.0.4'}),
('node5', {'ansible_host': '192.168.0.5',
'ip': '10.90.0.5',
'access_ip': '192.168.0.5'}),
('node6', {'ansible_host': '192.168.0.6',
'ip': '10.90.0.6',
'access_ip': '192.168.0.6'})])
existing = OrderedDict([
('node2', {'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'}),
('node3', {'ansible_host': '192.168.0.3',
'ip': '10.90.0.3',
'access_ip': '192.168.0.3'}),
('node4', {'ansible_host': '192.168.0.4',
'ip': '10.90.0.4',
'access_ip': '192.168.0.4'})])
self.inv.yaml_config['all']['hosts'] = existing
result = self.inv.build_hostnames(changed_hosts, True)
self.assertEqual(expected, result)
# Add two IP addresses into a config that has
# three already defined IP addresses. One of the IP addresses
# is a duplicate.
def test_build_hostnames_add_two_duplicate_one_overlap(self):
changed_hosts = ['10.90.0.4,192.168.0.4', '10.90.0.5,192.168.0.5']
expected = OrderedDict([
('node2', {'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'}),
('node3', {'ansible_host': '192.168.0.3',
'ip': '10.90.0.3',
'access_ip': '192.168.0.3'}),
('node4', {'ansible_host': '192.168.0.4',
'ip': '10.90.0.4',
'access_ip': '192.168.0.4'}),
('node5', {'ansible_host': '192.168.0.5',
'ip': '10.90.0.5',
'access_ip': '192.168.0.5'})])
existing = OrderedDict([
('node2', {'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'}),
('node3', {'ansible_host': '192.168.0.3',
'ip': '10.90.0.3',
'access_ip': '192.168.0.3'}),
('node4', {'ansible_host': '192.168.0.4',
'ip': '10.90.0.4',
'access_ip': '192.168.0.4'})])
self.inv.yaml_config['all']['hosts'] = existing
result = self.inv.build_hostnames(changed_hosts, True)
self.assertEqual(expected, result)
# Add two duplicate IP addresses into a config that has
# three already defined IP addresses
def test_build_hostnames_add_two_duplicate_two_overlap(self):
changed_hosts = ['10.90.0.3,192.168.0.3', '10.90.0.4,192.168.0.4']
expected = OrderedDict([
('node2', {'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'}),
('node3', {'ansible_host': '192.168.0.3',
'ip': '10.90.0.3',
'access_ip': '192.168.0.3'}),
('node4', {'ansible_host': '192.168.0.4',
'ip': '10.90.0.4',
'access_ip': '192.168.0.4'})])
existing = OrderedDict([
('node2', {'ansible_host': '192.168.0.2',
'ip': '10.90.0.2',
'access_ip': '192.168.0.2'}),
('node3', {'ansible_host': '192.168.0.3',
'ip': '10.90.0.3',
'access_ip': '192.168.0.3'}),
('node4', {'ansible_host': '192.168.0.4',
'ip': '10.90.0.4',
'access_ip': '192.168.0.4'})])
self.inv.yaml_config['all']['hosts'] = existing
result = self.inv.build_hostnames(changed_hosts, True)
self.assertEqual(expected, result)

View File

@@ -5,7 +5,7 @@ deployment on VMs.
This playbook does not create Virtual Machines, nor does it run Kubespray itself.
## User creation
### User creation
If you want to create a user for running Kubespray deployment, you should specify
both `k8s_deployment_user` and `k8s_deployment_user_pkey_path`.

View File

@@ -1,7 +1,7 @@
---
- name: Install required packages
package:
yum:
name: "{{ item }}"
state: present
with_items:

View File

@@ -11,7 +11,6 @@
state: directory
owner: "{{ k8s_deployment_user }}"
group: "{{ k8s_deployment_user }}"
mode: 0700
- name: Configure sudo for deployment user
copy:

View File

@@ -8,19 +8,19 @@ In the same directory of this ReadMe file you should find a file named `inventor
Change that file to reflect your local setup (adding more machines or removing them and setting the adequate ip numbers), and save it to `inventory/sample/k8s_gfs_inventory`. Make sure that the settings on `inventory/sample/group_vars/all.yml` make sense with your deployment. Then execute change to the kubespray root folder, and execute (supposing that the machines are all using ubuntu):
```shell
```
ansible-playbook -b --become-user=root -i inventory/sample/k8s_gfs_inventory --user=ubuntu ./cluster.yml
```
This will provision your Kubernetes cluster. Then, to provision and configure the GlusterFS cluster, from the same directory execute:
```shell
```
ansible-playbook -b --become-user=root -i inventory/sample/k8s_gfs_inventory --user=ubuntu ./contrib/network-storage/glusterfs/glusterfs.yml
```
If your machines are not using Ubuntu, you need to change the `--user=ubuntu` to the correct user. Alternatively, if your Kubernetes machines are using one OS and your GlusterFS a different one, you can instead specify the `ansible_ssh_user=<correct-user>` variable in the inventory file that you just created, for each machine/VM:
```shell
```
k8s-master-1 ansible_ssh_host=192.168.0.147 ip=192.168.0.147 ansible_ssh_user=core
k8s-master-node-1 ansible_ssh_host=192.168.0.148 ip=192.168.0.148 ansible_ssh_user=core
k8s-master-node-2 ansible_ssh_host=192.168.0.146 ip=192.168.0.146 ansible_ssh_user=core
@@ -30,7 +30,7 @@ k8s-master-node-2 ansible_ssh_host=192.168.0.146 ip=192.168.0.146 ansible_ssh_us
First step is to fill in a `my-kubespray-gluster-cluster.tfvars` file with the specification desired for your cluster. An example with all required variables would look like:
```ini
```
cluster_name = "cluster1"
number_of_k8s_masters = "1"
number_of_k8s_masters_no_floating_ip = "2"
@@ -39,7 +39,7 @@ number_of_k8s_nodes = "0"
public_key_path = "~/.ssh/my-desired-key.pub"
image = "Ubuntu 16.04"
ssh_user = "ubuntu"
flavor_k8s_node = "node-flavor-id-in-your-openstack"
flavor_k8s_node = "node-flavor-id-in-your-openstack"
flavor_k8s_master = "master-flavor-id-in-your-openstack"
network_name = "k8s-network"
floatingip_pool = "net_external"
@@ -54,7 +54,7 @@ ssh_user_gfs = "ubuntu"
As explained in the general terraform/openstack guide, you need to source your OpenStack credentials file, add your ssh-key to the ssh-agent and setup environment variables for terraform:
```shell
```
$ source ~/.stackrc
$ eval $(ssh-agent -s)
$ ssh-add ~/.ssh/my-desired-key
@@ -67,7 +67,7 @@ $ echo Setting up Terraform creds && \
Then, standing on the kubespray directory (root base of the Git checkout), issue the following terraform command to create the VMs for the cluster:
```shell
```
terraform apply -state=contrib/terraform/openstack/terraform.tfstate -var-file=my-kubespray-gluster-cluster.tfvars contrib/terraform/openstack
```
@@ -75,18 +75,18 @@ This will create both your Kubernetes and Gluster VMs. Make sure that the ansibl
Then, provision your Kubernetes (kubespray) cluster with the following ansible call:
```shell
```
ansible-playbook -b --become-user=root -i contrib/terraform/openstack/hosts ./cluster.yml
```
Finally, provision the glusterfs nodes and add the Persistent Volume setup for GlusterFS in Kubernetes through the following ansible call:
```shell
```
ansible-playbook -b --become-user=root -i contrib/terraform/openstack/hosts ./contrib/network-storage/glusterfs/glusterfs.yml
```
If you need to destroy the cluster, you can run:
```shell
```
terraform destroy -state=contrib/terraform/openstack/terraform.tfstate -var-file=my-kubespray-gluster-cluster.tfvars contrib/terraform/openstack
```

View File

@@ -15,10 +15,10 @@
roles:
- { role: glusterfs/server }
- hosts: k8s_cluster
- hosts: k8s-cluster
roles:
- { role: glusterfs/client }
- hosts: kube_control_plane[0]
- hosts: kube-master[0]
roles:
- { role: kubernetes-pv }

View File

@@ -11,10 +11,10 @@
# ## Set disk_volume_device_1 to desired device for gluster brick, if different to /dev/vdb (default).
# ## As in the previous case, you can set ip to give direct communication on internal IPs
# gfs_node1 ansible_ssh_host=95.54.0.18 # disk_volume_device_1=/dev/vdc ip=10.3.0.7
# gfs_node2 ansible_ssh_host=95.54.0.19 # disk_volume_device_1=/dev/vdc ip=10.3.0.8
# gfs_node3 ansible_ssh_host=95.54.0.20 # disk_volume_device_1=/dev/vdc ip=10.3.0.9
# gfs_node2 ansible_ssh_host=95.54.0.19 # disk_volume_device_1=/dev/vdc ip=10.3.0.8
# gfs_node3 ansible_ssh_host=95.54.0.20 # disk_volume_device_1=/dev/vdc ip=10.3.0.9
# [kube_control_plane]
# [kube-master]
# node1
# node2
@@ -23,16 +23,16 @@
# node2
# node3
# [kube_node]
# [kube-node]
# node2
# node3
# node4
# node5
# node6
# [k8s_cluster:children]
# kube_node
# kube_control_plane
# [k8s-cluster:children]
# kube-node
# kube-master
# [gfs-cluster]
# gfs_node1

View File

@@ -8,7 +8,7 @@ Installs and configures GlusterFS on Linux.
For GlusterFS to connect between servers, TCP ports `24007`, `24008`, and `24009`/`49152`+ (that port, plus an additional incremented port for each additional server in the cluster; the latter if GlusterFS is version 3.4+), and TCP/UDP port `111` must be open. You can open these using whatever firewall you wish (this can easily be configured using the `geerlingguy.firewall` role).
This role performs basic installation and setup of Gluster, but it does not configure or mount bricks (volumes), since that step is easier to do in a series of plays in your own playbook. Ansible 1.9+ includes the [`gluster_volume`](https://docs.ansible.com/ansible/latest/collections/gluster/gluster/gluster_volume_module.html) module to ease the management of Gluster volumes.
This role performs basic installation and setup of Gluster, but it does not configure or mount bricks (volumes), since that step is easier to do in a series of plays in your own playbook. Ansible 1.9+ includes the [`gluster_volume`](https://docs.ansible.com/gluster_volume_module.html) module to ease the management of Gluster volumes.
## Role Variables

View File

@@ -1,10 +1,10 @@
---
- name: Install Prerequisites
package: name={{ item }} state=present
yum: name={{ item }} state=present
with_items:
- "centos-release-gluster{{ glusterfs_default_release }}"
- name: Install Packages
package: name={{ item }} state=present
yum: name={{ item }} state=present
with_items:
- glusterfs-client

View File

@@ -9,7 +9,7 @@
when: ansible_os_family == "Debian"
- name: install xfs RedHat
package: name=xfsprogs state=present
yum: name=xfsprogs state=present
when: ansible_os_family == "RedHat"
# Format external volumes in xfs
@@ -82,7 +82,6 @@
template:
dest: "{{ gluster_mount_dir }}/.test-file.txt"
src: test-file.txt
mode: 0644
when: groups['gfs-cluster'] is defined and inventory_hostname == groups['gfs-cluster'][0]
- name: Unmount glusterfs

View File

@@ -1,11 +1,11 @@
---
- name: Install Prerequisites
package: name={{ item }} state=present
yum: name={{ item }} state=present
with_items:
- "centos-release-gluster{{ glusterfs_default_release }}"
- name: Install Packages
package: name={{ item }} state=present
yum: name={{ item }} state=present
with_items:
- glusterfs-server
- glusterfs-client

View File

@@ -1,4 +1,5 @@
---
- hosts: all
roles:
- { role: prepare }
- role_under_test

View File

@@ -8,7 +8,7 @@
- { file: glusterfs-kubernetes-pv.yml.j2, type: pv, dest: glusterfs-kubernetes-pv.yml}
- { file: glusterfs-kubernetes-endpoint-svc.json.j2, type: svc, dest: glusterfs-kubernetes-endpoint-svc.json}
register: gluster_pv
when: inventory_hostname == groups['kube_control_plane'][0] and groups['gfs-cluster'] is defined and hostvars[groups['gfs-cluster'][0]].gluster_disk_size_gb is defined
when: inventory_hostname == groups['kube-master'][0] and groups['gfs-cluster'] is defined and hostvars[groups['gfs-cluster'][0]].gluster_disk_size_gb is defined
- name: Kubernetes Apps | Set GlusterFS endpoint and PV
kube:
@@ -19,4 +19,4 @@
filename: "{{ kube_config_dir }}/{{ item.item.dest }}"
state: "{{ item.changed | ternary('latest','present') }}"
with_items: "{{ gluster_pv.results }}"
when: inventory_hostname == groups['kube_control_plane'][0] and groups['gfs-cluster'] is defined
when: inventory_hostname == groups['kube-master'][0] and groups['gfs-cluster'] is defined

View File

@@ -1,26 +1,17 @@
# Deploy Heketi/Glusterfs into Kubespray/Kubernetes
This playbook aims to automate [this](https://github.com/heketi/heketi/blob/master/docs/admin/install-kubernetes.md) tutorial. It deploys heketi/glusterfs into kubernetes and sets up a storageclass.
## Important notice
> Due to resource limits on the current project maintainers and general lack of contributions we are considering placing Heketi into a [near-maintenance mode](https://github.com/heketi/heketi#important-notice)
## Client Setup
Heketi provides a CLI that provides users with a means to administer the deployment and configuration of GlusterFS in Kubernetes. [Download and install the heketi-cli](https://github.com/heketi/heketi/releases) on your client machine.
## Install
Copy the inventory.yml.sample over to inventory/sample/k8s_heketi_inventory.yml and change it according to your setup.
```shell
```
ansible-playbook --ask-become -i inventory/sample/k8s_heketi_inventory.yml contrib/network-storage/heketi/heketi.yml
```
## Tear down
```shell
```
ansible-playbook --ask-become -i inventory/sample/k8s_heketi_inventory.yml contrib/network-storage/heketi/heketi-tear-down.yml
```

View File

@@ -1,5 +1,5 @@
---
- hosts: kube_control_plane[0]
- hosts: kube-master[0]
roles:
- { role: tear-down }

View File

@@ -3,7 +3,7 @@
roles:
- { role: prepare }
- hosts: kube_control_plane[0]
- hosts: kube-master[0]
tags:
- "provision"
roles:

View File

@@ -2,25 +2,18 @@ all:
vars:
heketi_admin_key: "11elfeinhundertundelf"
heketi_user_key: "!!einseinseins"
glusterfs_daemonset:
readiness_probe:
timeout_seconds: 3
initial_delay_seconds: 3
liveness_probe:
timeout_seconds: 3
initial_delay_seconds: 10
children:
k8s_cluster:
k8s-cluster:
vars:
kubelet_fail_swap_on: false
children:
kube_control_plane:
kube-master:
hosts:
node1:
etcd:
hosts:
node2:
kube_node:
kube-node:
hosts: &kube_nodes
node1:
node2:

View File

@@ -11,7 +11,7 @@
- name: "Install glusterfs mount utils (RedHat)"
become: true
package:
yum:
name: "glusterfs-fuse"
state: "present"
when: "ansible_os_family == 'RedHat'"

View File

@@ -1,10 +1,7 @@
---
- name: "Kubernetes Apps | Lay Down Heketi Bootstrap"
become: true
template:
src: "heketi-bootstrap.json.j2"
dest: "{{ kube_config_dir }}/heketi-bootstrap.json"
mode: 0640
template: { src: "heketi-bootstrap.json.j2", dest: "{{ kube_config_dir }}/heketi-bootstrap.json" }
register: "rendering"
- name: "Kubernetes Apps | Install and configure Heketi Bootstrap"
kube:

View File

@@ -10,7 +10,6 @@
template:
src: "topology.json.j2"
dest: "{{ kube_config_dir }}/topology.json"
mode: 0644
- name: "Copy topology configuration into container."
changed_when: false
command: "{{ bin_dir }}/kubectl cp {{ kube_config_dir }}/topology.json {{ initial_heketi_pod_name }}:/tmp/topology.json"

View File

@@ -1,9 +1,6 @@
---
- name: "Kubernetes Apps | Lay Down GlusterFS Daemonset"
template:
src: "glusterfs-daemonset.json.j2"
dest: "{{ kube_config_dir }}/glusterfs-daemonset.json"
mode: 0644
template: { src: "glusterfs-daemonset.json.j2", dest: "{{ kube_config_dir }}/glusterfs-daemonset.json" }
become: true
register: "rendering"
- name: "Kubernetes Apps | Install and configure GlusterFS daemonset"
@@ -30,10 +27,7 @@
delay: 5
- name: "Kubernetes Apps | Lay Down Heketi Service Account"
template:
src: "heketi-service-account.json.j2"
dest: "{{ kube_config_dir }}/heketi-service-account.json"
mode: 0644
template: { src: "heketi-service-account.json.j2", dest: "{{ kube_config_dir }}/heketi-service-account.json" }
become: true
register: "rendering"
- name: "Kubernetes Apps | Install and configure Heketi Service Account"

View File

@@ -4,7 +4,6 @@
template:
src: "heketi-deployment.json.j2"
dest: "{{ kube_config_dir }}/heketi-deployment.json"
mode: 0644
register: "rendering"
- name: "Kubernetes Apps | Install and configure Heketi"

View File

@@ -5,7 +5,7 @@
changed_when: false
- name: "Kubernetes Apps | Deploy cluster role binding."
when: "clusterrolebinding_state.stdout | length == 0"
when: "clusterrolebinding_state.stdout == \"\""
command: "{{ bin_dir }}/kubectl create clusterrolebinding heketi-gluster-admin --clusterrole=edit --serviceaccount=default:heketi-service-account"
- name: Get clusterrolebindings again
@@ -15,7 +15,7 @@
- name: Make sure that clusterrolebindings are present now
assert:
that: "clusterrolebinding_state.stdout | length > 0"
that: "clusterrolebinding_state.stdout != \"\""
msg: "Cluster role binding is not present."
- name: Get the heketi-config-secret secret
@@ -28,10 +28,9 @@
template:
src: "heketi.json.j2"
dest: "{{ kube_config_dir }}/heketi.json"
mode: 0644
- name: "Deploy Heketi config secret"
when: "secret_state.stdout | length == 0"
when: "secret_state.stdout == \"\""
command: "{{ bin_dir }}/kubectl create secret generic heketi-config-secret --from-file={{ kube_config_dir }}/heketi.json"
- name: Get the heketi-config-secret secret again
@@ -41,5 +40,5 @@
- name: Make sure the heketi-config-secret secret exists now
assert:
that: "secret_state.stdout | length > 0"
that: "secret_state.stdout != \"\""
msg: "Heketi config secret is not present."

View File

@@ -2,10 +2,7 @@
- name: "Kubernetes Apps | Lay Down Heketi Storage"
become: true
vars: { nodes: "{{ groups['heketi-node'] }}" }
template:
src: "heketi-storage.json.j2"
dest: "{{ kube_config_dir }}/heketi-storage.json"
mode: 0644
template: { src: "heketi-storage.json.j2", dest: "{{ kube_config_dir }}/heketi-storage.json" }
register: "rendering"
- name: "Kubernetes Apps | Install and configure Heketi Storage"
kube:

View File

@@ -16,7 +16,6 @@
template:
src: "storageclass.yml.j2"
dest: "{{ kube_config_dir }}/storageclass.yml"
mode: 0644
register: "rendering"
- name: "Kubernetes Apps | Install and configure Storace Class"
kube:

View File

@@ -10,7 +10,6 @@
template:
src: "topology.json.j2"
dest: "{{ kube_config_dir }}/topology.json"
mode: 0644
- name: "Copy topology configuration into container." # noqa 503
when: "rendering.changed"
command: "{{ bin_dir }}/kubectl cp {{ kube_config_dir }}/topology.json {{ heketi_pod_name }}:/tmp/topology.json"

View File

@@ -73,8 +73,8 @@
"privileged": true
},
"readinessProbe": {
"timeoutSeconds": {{ glusterfs_daemonset.readiness_probe.timeout_seconds }},
"initialDelaySeconds": {{ glusterfs_daemonset.readiness_probe.initial_delay_seconds }},
"timeoutSeconds": 3,
"initialDelaySeconds": 3,
"exec": {
"command": [
"/bin/bash",
@@ -84,8 +84,8 @@
}
},
"livenessProbe": {
"timeoutSeconds": {{ glusterfs_daemonset.liveness_probe.timeout_seconds }},
"initialDelaySeconds": {{ glusterfs_daemonset.liveness_probe.initial_delay_seconds }},
"timeoutSeconds": 3,
"initialDelaySeconds": 10,
"exec": {
"command": [
"/bin/bash",

View File

@@ -1,7 +1,7 @@
---
- name: "Install lvm utils (RedHat)"
become: true
package:
yum:
name: "lvm2"
state: "present"
when: "ansible_os_family == 'RedHat'"
@@ -19,7 +19,7 @@
become: true
shell: "pvs {{ disk_volume_device_1 }} --option vg_name | tail -n+2"
register: "volume_groups"
ignore_errors: true # noqa ignore-errors
ignore_errors: true
changed_when: false
- name: "Remove volume groups." # noqa 301
@@ -35,11 +35,11 @@
PATH: "{{ ansible_env.PATH }}:/sbin" # Make sure we can workaround RH / CentOS conservative path management
become: true
command: "pvremove {{ disk_volume_device_1 }} --yes"
ignore_errors: true # noqa ignore-errors
ignore_errors: true
- name: "Remove lvm utils (RedHat)"
become: true
package:
yum:
name: "lvm2"
state: "absent"
when: "ansible_os_family == 'RedHat' and heketi_remove_lvm"

View File

@@ -1,51 +1,51 @@
---
- name: Remove storage class. # noqa 301
- name: "Remove storage class." # noqa 301
command: "{{ bin_dir }}/kubectl delete storageclass gluster"
ignore_errors: true # noqa ignore-errors
- name: Tear down heketi. # noqa 301
ignore_errors: true
- name: "Tear down heketi." # noqa 301
command: "{{ bin_dir }}/kubectl delete all,service,jobs,deployment,secret --selector=\"glusterfs=heketi-pod\""
ignore_errors: true # noqa ignore-errors
- name: Tear down heketi. # noqa 301
ignore_errors: true
- name: "Tear down heketi." # noqa 301
command: "{{ bin_dir }}/kubectl delete all,service,jobs,deployment,secret --selector=\"glusterfs=heketi-deployment\""
ignore_errors: true # noqa ignore-errors
- name: Tear down bootstrap.
ignore_errors: true
- name: "Tear down bootstrap."
include_tasks: "../../provision/tasks/bootstrap/tear-down.yml"
- name: Ensure there is nothing left over. # noqa 301
- name: "Ensure there is nothing left over." # noqa 301
command: "{{ bin_dir }}/kubectl get all,service,jobs,deployment,secret --selector=\"glusterfs=heketi-pod\" -o=json"
register: "heketi_result"
until: "heketi_result.stdout|from_json|json_query('items[*]')|length == 0"
retries: 60
delay: 5
- name: Ensure there is nothing left over. # noqa 301
- name: "Ensure there is nothing left over." # noqa 301
command: "{{ bin_dir }}/kubectl get all,service,jobs,deployment,secret --selector=\"glusterfs=heketi-deployment\" -o=json"
register: "heketi_result"
until: "heketi_result.stdout|from_json|json_query('items[*]')|length == 0"
retries: 60
delay: 5
- name: Tear down glusterfs. # noqa 301
- name: "Tear down glusterfs." # noqa 301
command: "{{ bin_dir }}/kubectl delete daemonset.extensions/glusterfs"
ignore_errors: true # noqa ignore-errors
- name: Remove heketi storage service. # noqa 301
ignore_errors: true
- name: "Remove heketi storage service." # noqa 301
command: "{{ bin_dir }}/kubectl delete service heketi-storage-endpoints"
ignore_errors: true # noqa ignore-errors
- name: Remove heketi gluster role binding # noqa 301
ignore_errors: true
- name: "Remove heketi gluster role binding" # noqa 301
command: "{{ bin_dir }}/kubectl delete clusterrolebinding heketi-gluster-admin"
ignore_errors: true # noqa ignore-errors
- name: Remove heketi config secret # noqa 301
ignore_errors: true
- name: "Remove heketi config secret" # noqa 301
command: "{{ bin_dir }}/kubectl delete secret heketi-config-secret"
ignore_errors: true # noqa ignore-errors
- name: Remove heketi db backup # noqa 301
ignore_errors: true
- name: "Remove heketi db backup" # noqa 301
command: "{{ bin_dir }}/kubectl delete secret heketi-db-backup"
ignore_errors: true # noqa ignore-errors
- name: Remove heketi service account # noqa 301
ignore_errors: true
- name: "Remove heketi service account" # noqa 301
command: "{{ bin_dir }}/kubectl delete serviceaccount heketi-service-account"
ignore_errors: true # noqa ignore-errors
- name: Get secrets
ignore_errors: true
- name: "Get secrets"
command: "{{ bin_dir }}/kubectl get secrets --output=\"json\""
register: "secrets"
changed_when: false
- name: Remove heketi storage secret
- name: "Remove heketi storage secret"
vars: { storage_query: "items[?metadata.annotations.\"kubernetes.io/service-account.name\"=='heketi-service-account'].metadata.name|[0]" }
command: "{{ bin_dir }}/kubectl delete secret {{ secrets.stdout|from_json|json_query(storage_query) }}"
when: "storage_query is defined"
ignore_errors: true # noqa ignore-errors
ignore_errors: true

View File

@@ -1,46 +0,0 @@
# Offline deployment
## manage-offline-container-images.sh
Container image collecting script for offline deployment
This script has two features:
(1) Get container images from an environment which is deployed online.
(2) Deploy local container registry and register the container images to the registry.
Step(1) should be done online site as a preparation, then we bring the gotten images
to the target offline environment.
Then we will run step(2) for registering the images to local registry.
Step(1) can be operated with:
```shell
manage-offline-container-images.sh create
```
Step(2) can be operated with:
```shell
manage-offline-container-images.sh register
```
## generate_list.sh
This script generates the list of downloaded files and the list of container images by `roles/download/defaults/main.yml` file.
Run this script will execute `generate_list.yml` playbook in kubespray root directory and generate four files,
all downloaded files url in files.list, all container images in images.list, jinja2 templates in *.template.
```shell
./generate_list.sh
tree temp
temp
├── files.list
├── files.list.template
├── images.list
└── images.list.template
0 directories, 5 files
```
In some cases you may want to update some component version, you can declare version variables in ansible inventory file or group_vars,
then run `./generate_list.sh -i [inventory_file]` to update file.list and images.list.

View File

@@ -1 +0,0 @@
{ "insecure-registries":["HOSTNAME:5000"] }

View File

@@ -1,30 +0,0 @@
#!/bin/bash
set -eo pipefail
CURRENT_DIR=$(cd $(dirname $0); pwd)
TEMP_DIR="${CURRENT_DIR}/temp"
REPO_ROOT_DIR="${CURRENT_DIR%/contrib/offline}"
: ${DOWNLOAD_YML:="roles/download/defaults/main.yml"}
mkdir -p ${TEMP_DIR}
# generate all download files url template
grep 'download_url:' ${REPO_ROOT_DIR}/${DOWNLOAD_YML} \
| sed 's/^.*_url: //g;s/\"//g' > ${TEMP_DIR}/files.list.template
# generate all images list template
sed -n '/^downloads:/,/download_defaults:/p' ${REPO_ROOT_DIR}/${DOWNLOAD_YML} \
| sed -n "s/repo: //p;s/tag: //p" | tr -d ' ' \
| sed 'N;s#\n# #g' | tr ' ' ':' | sed 's/\"//g' > ${TEMP_DIR}/images.list.template
# add kube-* images to images list template
KUBE_IMAGES="kube-apiserver kube-controller-manager kube-scheduler kube-proxy"
for i in $KUBE_IMAGES; do
echo "{{ kube_image_repo }}/$i:{{ kube_version }}" >> ${TEMP_DIR}/images.list.template
done
# run ansible to expand templates
/bin/cp ${CURRENT_DIR}/generate_list.yml ${REPO_ROOT_DIR}
(cd ${REPO_ROOT_DIR} && ansible-playbook $* generate_list.yml && /bin/rm generate_list.yml) || exit 1

View File

@@ -1,20 +0,0 @@
---
- hosts: localhost
become: false
roles:
# Just load default variables from roles.
- role: kubespray-defaults
when: false
- role: download
when: false
tasks:
- name: Generate files.list and images.list files from templates
template:
src: ./contrib/offline/temp/{{ item }}.list.template
dest: ./contrib/offline/temp/{{ item }}.list
mode: 0644
with_items:
- files
- images

View File

@@ -1,170 +0,0 @@
#!/bin/bash
OPTION=$1
CURRENT_DIR=$(cd $(dirname $0); pwd)
TEMP_DIR="${CURRENT_DIR}/temp"
IMAGE_TAR_FILE="${CURRENT_DIR}/container-images.tar.gz"
IMAGE_DIR="${CURRENT_DIR}/container-images"
IMAGE_LIST="${IMAGE_DIR}/container-images.txt"
RETRY_COUNT=5
function create_container_image_tar() {
set -e
IMAGES=$(kubectl describe pods --all-namespaces | grep " Image:" | awk '{print $2}' | sort | uniq)
# NOTE: etcd and pause cannot be seen as pods.
# The pause image is used for --pod-infra-container-image option of kubelet.
EXT_IMAGES=$(kubectl cluster-info dump | egrep "quay.io/coreos/etcd:|k8s.gcr.io/pause:" | sed s@\"@@g)
IMAGES="${IMAGES} ${EXT_IMAGES}"
rm -f ${IMAGE_TAR_FILE}
rm -rf ${IMAGE_DIR}
mkdir ${IMAGE_DIR}
cd ${IMAGE_DIR}
sudo docker pull registry:latest
sudo docker save -o registry-latest.tar registry:latest
for image in ${IMAGES}
do
FILE_NAME="$(echo ${image} | sed s@"/"@"-"@g | sed s/":"/"-"/g)".tar
set +e
for step in $(seq 1 ${RETRY_COUNT})
do
sudo docker pull ${image}
if [ $? -eq 0 ]; then
break
fi
echo "Failed to pull ${image} at step ${step}"
if [ ${step} -eq ${RETRY_COUNT} ]; then
exit 1
fi
done
set -e
sudo docker save -o ${FILE_NAME} ${image}
# NOTE: Here removes the following repo parts from each image
# so that these parts will be replaced with Kubespray.
# - kube_image_repo: "k8s.gcr.io"
# - gcr_image_repo: "gcr.io"
# - docker_image_repo: "docker.io"
# - quay_image_repo: "quay.io"
FIRST_PART=$(echo ${image} | awk -F"/" '{print $1}')
if [ "${FIRST_PART}" = "k8s.gcr.io" ] ||
[ "${FIRST_PART}" = "gcr.io" ] ||
[ "${FIRST_PART}" = "docker.io" ] ||
[ "${FIRST_PART}" = "quay.io" ]; then
image=$(echo ${image} | sed s@"${FIRST_PART}/"@@)
fi
echo "${FILE_NAME} ${image}" >> ${IMAGE_LIST}
done
cd ..
sudo chown ${USER} ${IMAGE_DIR}/*
tar -zcvf ${IMAGE_TAR_FILE} ./container-images
rm -rf ${IMAGE_DIR}
echo ""
echo "${IMAGE_TAR_FILE} is created to contain your container images."
echo "Please keep this file and bring it to your offline environment."
}
function register_container_images() {
if [ ! -f ${IMAGE_TAR_FILE} ]; then
echo "${IMAGE_TAR_FILE} should exist."
exit 1
fi
if [ ! -d ${TEMP_DIR} ]; then
mkdir ${TEMP_DIR}
fi
# To avoid "http: server gave http response to https client" error.
LOCALHOST_NAME=$(hostname)
if [ -d /etc/docker/ ]; then
set -e
# Ubuntu18.04, RHEL7/CentOS7
cp ${CURRENT_DIR}/docker-daemon.json ${TEMP_DIR}/docker-daemon.json
sed -i s@"HOSTNAME"@"${LOCALHOST_NAME}"@ ${TEMP_DIR}/docker-daemon.json
sudo cp ${TEMP_DIR}/docker-daemon.json /etc/docker/daemon.json
elif [ -d /etc/containers/ ]; then
set -e
# RHEL8/CentOS8
cp ${CURRENT_DIR}/registries.conf ${TEMP_DIR}/registries.conf
sed -i s@"HOSTNAME"@"${LOCALHOST_NAME}"@ ${TEMP_DIR}/registries.conf
sudo cp ${TEMP_DIR}/registries.conf /etc/containers/registries.conf
else
echo "docker package(docker-ce, etc.) should be installed"
exit 1
fi
tar -zxvf ${IMAGE_TAR_FILE}
sudo docker load -i ${IMAGE_DIR}/registry-latest.tar
set +e
sudo docker container inspect registry >/dev/null 2>&1
if [ $? -ne 0 ]; then
sudo docker run --restart=always -d -p 5000:5000 --name registry registry:latest
fi
set -e
while read -r line; do
file_name=$(echo ${line} | awk '{print $1}')
raw_image=$(echo ${line} | awk '{print $2}')
new_image="${LOCALHOST_NAME}:5000/${raw_image}"
org_image=$(sudo docker load -i ${IMAGE_DIR}/${file_name} | head -n1 | awk '{print $3}')
image_id=$(sudo docker image inspect ${org_image} | grep "\"Id\":" | awk -F: '{print $3}'| sed s/'\",'//)
if [ -z "${file_name}" ]; then
echo "Failed to get file_name for line ${line}"
exit 1
fi
if [ -z "${raw_image}" ]; then
echo "Failed to get raw_image for line ${line}"
exit 1
fi
if [ -z "${org_image}" ]; then
echo "Failed to get org_image for line ${line}"
exit 1
fi
if [ -z "${image_id}" ]; then
echo "Failed to get image_id for file ${file_name}"
exit 1
fi
sudo docker load -i ${IMAGE_DIR}/${file_name}
sudo docker tag ${image_id} ${new_image}
sudo docker push ${new_image}
done <<< "$(cat ${IMAGE_LIST})"
echo "Succeeded to register container images to local registry."
echo "Please specify ${LOCALHOST_NAME}:5000 for the following options in your inventry:"
echo "- kube_image_repo"
echo "- gcr_image_repo"
echo "- docker_image_repo"
echo "- quay_image_repo"
}
if [ "${OPTION}" == "create" ]; then
create_container_image_tar
elif [ "${OPTION}" == "register" ]; then
register_container_images
else
echo "This script has two features:"
echo "(1) Get container images from an environment which is deployed online."
echo "(2) Deploy local container registry and register the container images to the registry."
echo ""
echo "Step(1) should be done online site as a preparation, then we bring"
echo "the gotten images to the target offline environment."
echo "Then we will run step(2) for registering the images to local registry."
echo ""
echo "${IMAGE_TAR_FILE} is created to contain your container images."
echo "Please keep this file and bring it to your offline environment."
echo ""
echo "Step(1) can be operated with:"
echo " $ ./manage-offline-container-images.sh create"
echo ""
echo "Step(2) can be operated with:"
echo " $ ./manage-offline-container-images.sh register"
echo ""
echo "Please specify 'create' or 'register'."
echo ""
exit 1
fi

View File

@@ -1,8 +0,0 @@
[registries.search]
registries = ['registry.access.redhat.com', 'registry.redhat.io', 'docker.io']
[registries.insecure]
registries = ['HOSTNAME:5000']
[registries.block]
registries = []

View File

@@ -1,2 +0,0 @@
---
disable_service_firewall: false

View File

@@ -1,23 +0,0 @@
---
- block:
- name: List services
service_facts:
- name: Disable service firewalld
systemd:
name: firewalld
state: stopped
enabled: no
when:
"'firewalld.service' in services"
- name: Disable service ufw
systemd:
name: ufw
state: stopped
enabled: no
when:
"'ufw.service' in services"
when:
- disable_service_firewall is defined and disable_service_firewall

View File

@@ -9,8 +9,8 @@ Summary: Ansible modules for installing Kubernetes
Group: System Environment/Libraries
License: ASL 2.0
Url: https://github.com/kubernetes-sigs/kubespray
Source0: https://github.com/kubernetes-sigs/kubespray/archive/%{upstream_version}.tar.gz#/%{name}-%{release}.tar.gz
Url: https://github.com/kubernetes-incubator/kubespray
Source0: https://github.com/kubernetes-incubator/kubespray/archive/%{upstream_version}.tar.gz#/%{name}-%{release}.tar.gz
BuildArch: noarch
BuildRequires: git
@@ -51,7 +51,7 @@ export SKIP_PIP_INSTALL=1
%doc %{_docdir}/%{name}/inventory/sample/hosts.ini
%config %{_sysconfdir}/%{name}/ansible.cfg
%config %{_sysconfdir}/%{name}/inventory/sample/group_vars/all.yml
%config %{_sysconfdir}/%{name}/inventory/sample/group_vars/k8s_cluster.yml
%config %{_sysconfdir}/%{name}/inventory/sample/group_vars/k8s-cluster.yml
%license %{_docdir}/%{name}/LICENSE
%{python2_sitelib}/%{srcname}-%{release}-py%{python2_version}.egg-info
%{_datarootdir}/%{name}/roles/

View File

@@ -1,3 +1,2 @@
*.tfstate*
.terraform.lock.hcl
.terraform

View File

@@ -1,44 +1,40 @@
# Kubernetes on AWS with Terraform
## Kubernetes on AWS with Terraform
## Overview
**Overview:**
This project will create:
* VPC with Public and Private Subnets in # Availability Zones
* Bastion Hosts and NAT Gateways in the Public Subnet
* A dynamic number of masters, etcd, and worker nodes in the Private Subnet
* even distributed over the # of Availability Zones
* AWS ELB in the Public Subnet for accessing the Kubernetes API from the internet
- VPC with Public and Private Subnets in # Availability Zones
- Bastion Hosts and NAT Gateways in the Public Subnet
- A dynamic number of masters, etcd, and worker nodes in the Private Subnet
- even distributed over the # of Availability Zones
- AWS ELB in the Public Subnet for accessing the Kubernetes API from the internet
## Requirements
**Requirements**
- Terraform 0.12.0 or newer
## How to Use
**How to Use:**
- Export the variables for your AWS credentials or edit `credentials.tfvars`:
```commandline
```
export TF_VAR_AWS_ACCESS_KEY_ID="www"
export TF_VAR_AWS_SECRET_ACCESS_KEY ="xxx"
export TF_VAR_AWS_SSH_KEY_NAME="yyy"
export TF_VAR_AWS_DEFAULT_REGION="zzz"
```
- Update `contrib/terraform/aws/terraform.tfvars` with your data. By default, the Terraform scripts use Ubuntu 18.04 LTS (Bionic) as base image. If you want to change this behaviour, see note "Using other distrib than Ubuntu" below.
- Create an AWS EC2 SSH Key
- Run with `terraform apply --var-file="credentials.tfvars"` or `terraform apply` depending if you exported your AWS credentials
Example:
```commandline
terraform apply -var-file=credentials.tfvars
```
- Terraform automatically creates an Ansible Inventory file called `hosts` with the created infrastructure in the directory `inventory`
- Ansible will automatically generate an ssh config file for your bastion hosts. To connect to hosts with ssh using bastion host use generated ssh-bastion.conf.
Ansible automatically detects bastion and changes ssh_args
```commandline
ssh -F ./ssh-bastion.conf user@$ip
```
@@ -46,19 +42,14 @@ ssh -F ./ssh-bastion.conf user@$ip
- Once the infrastructure is created, you can run the kubespray playbooks and supply inventory/hosts with the `-i` flag.
Example (this one assumes you are using Ubuntu)
```commandline
ansible-playbook -i ./inventory/hosts ./cluster.yml -e ansible_user=ubuntu -b --become-user=root --flush-cache
```
***Using other distrib than Ubuntu***
If you want to use another distribution than Ubuntu 18.04 (Bionic) LTS, you can modify the search filters of the 'data "aws_ami" "distro"' in variables.tf.
For example, to use:
- Debian Jessie, replace 'data "aws_ami" "distro"' in variables.tf with
```ini
data "aws_ami" "distro" {
most_recent = true
@@ -74,11 +65,8 @@ data "aws_ami" "distro" {
owners = ["379101102735"]
}
```
- Ubuntu 16.04, replace 'data "aws_ami" "distro"' in variables.tf with
```ini
data "aws_ami" "distro" {
most_recent = true
@@ -94,11 +82,8 @@ data "aws_ami" "distro" {
owners = ["099720109477"]
}
```
- Centos 7, replace 'data "aws_ami" "distro"' in variables.tf with
```ini
data "aws_ami" "distro" {
most_recent = true
@@ -114,49 +99,23 @@ data "aws_ami" "distro" {
owners = ["688023202711"]
}
```
## Connecting to Kubernetes
**Troubleshooting**
You can use the following set of commands to get the kubeconfig file from your newly created cluster. Before running the commands, make sure you are in the project's root folder.
```commandline
# Get the controller's IP address.
CONTROLLER_HOST_NAME=$(cat ./inventory/hosts | grep "\[kube_control_plane\]" -A 1 | tail -n 1)
CONTROLLER_IP=$(cat ./inventory/hosts | grep $CONTROLLER_HOST_NAME | grep ansible_host | cut -d'=' -f2)
# Get the hostname of the load balancer.
LB_HOST=$(cat inventory/hosts | grep apiserver_loadbalancer_domain_name | cut -d'"' -f2)
# Get the controller's SSH fingerprint.
ssh-keygen -R $CONTROLLER_IP > /dev/null 2>&1
ssh-keyscan -H $CONTROLLER_IP >> ~/.ssh/known_hosts 2>/dev/null
# Get the kubeconfig from the controller.
mkdir -p ~/.kube
ssh -F ssh-bastion.conf centos@$CONTROLLER_IP "sudo chmod 644 /etc/kubernetes/admin.conf"
scp -F ssh-bastion.conf centos@$CONTROLLER_IP:/etc/kubernetes/admin.conf ~/.kube/config
sed -i "s^server:.*^server: https://$LB_HOST:6443^" ~/.kube/config
kubectl get nodes
```
## Troubleshooting
### Remaining AWS IAM Instance Profile
***Remaining AWS IAM Instance Profile***:
If the cluster was destroyed without using Terraform it is possible that
the AWS IAM Instance Profiles still remain. To delete them you can use
the `AWS CLI` with the following command:
```commandline
```
aws iam delete-instance-profile --region <region_name> --instance-profile-name <profile_name>
```
### Ansible Inventory doesn't get created
***Ansible Inventory doesn't get created:***
It could happen that Terraform doesn't create an Ansible Inventory file automatically. If this is the case copy the output after `inventory=` and create a file named `hosts`in the directory `inventory` and paste the inventory into the file.
## Architecture
**Architecture**
Pictured is an AWS Infrastructure created with this Terraform project distributed over two Availability Zones.

View File

@@ -20,7 +20,7 @@ module "aws-vpc" {
aws_cluster_name = var.aws_cluster_name
aws_vpc_cidr_block = var.aws_vpc_cidr_block
aws_avail_zones = slice(data.aws_availability_zones.available.names, 0, length(var.aws_cidr_subnets_public) <= length(data.aws_availability_zones.available.names) ? length(var.aws_cidr_subnets_public) : length(data.aws_availability_zones.available.names))
aws_avail_zones = slice(data.aws_availability_zones.available.names, 0, 2)
aws_cidr_subnets_private = var.aws_cidr_subnets_private
aws_cidr_subnets_public = var.aws_cidr_subnets_public
default_tags = var.default_tags
@@ -31,7 +31,7 @@ module "aws-elb" {
aws_cluster_name = var.aws_cluster_name
aws_vpc_id = module.aws-vpc.aws_vpc_id
aws_avail_zones = slice(data.aws_availability_zones.available.names, 0, length(var.aws_cidr_subnets_public) <= length(data.aws_availability_zones.available.names) ? length(var.aws_cidr_subnets_public) : length(data.aws_availability_zones.available.names))
aws_avail_zones = slice(data.aws_availability_zones.available.names, 0, 2)
aws_subnet_ids_public = module.aws-vpc.aws_subnet_ids_public
aws_elb_api_port = var.aws_elb_api_port
k8s_secure_api_port = var.k8s_secure_api_port
@@ -52,20 +52,20 @@ module "aws-iam" {
resource "aws_instance" "bastion-server" {
ami = data.aws_ami.distro.id
instance_type = var.aws_bastion_size
count = var.aws_bastion_num
count = length(var.aws_cidr_subnets_public)
associate_public_ip_address = true
availability_zone = element(slice(data.aws_availability_zones.available.names, 0, length(var.aws_cidr_subnets_public) <= length(data.aws_availability_zones.available.names) ? length(var.aws_cidr_subnets_public) : length(data.aws_availability_zones.available.names)), count.index)
availability_zone = element(slice(data.aws_availability_zones.available.names, 0, 2), count.index)
subnet_id = element(module.aws-vpc.aws_subnet_ids_public, count.index)
vpc_security_group_ids = module.aws-vpc.aws_security_group
key_name = var.AWS_SSH_KEY_NAME
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-bastion-${count.index}"
Cluster = var.aws_cluster_name
Role = "bastion-${var.aws_cluster_name}-${count.index}"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-bastion-${count.index}",
"Cluster", "${var.aws_cluster_name}",
"Role", "bastion-${var.aws_cluster_name}-${count.index}"
))
}
/*
@@ -79,23 +79,19 @@ resource "aws_instance" "k8s-master" {
count = var.aws_kube_master_num
availability_zone = element(slice(data.aws_availability_zones.available.names, 0, length(var.aws_cidr_subnets_public) <= length(data.aws_availability_zones.available.names) ? length(var.aws_cidr_subnets_public) : length(data.aws_availability_zones.available.names)), count.index)
availability_zone = element(slice(data.aws_availability_zones.available.names, 0, 2), count.index)
subnet_id = element(module.aws-vpc.aws_subnet_ids_private, count.index)
vpc_security_group_ids = module.aws-vpc.aws_security_group
root_block_device {
volume_size = var.aws_kube_master_disk_size
}
iam_instance_profile = module.aws-iam.kube_control_plane-profile
iam_instance_profile = module.aws-iam.kube-master-profile
key_name = var.AWS_SSH_KEY_NAME
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-master${count.index}"
"kubernetes.io/cluster/${var.aws_cluster_name}" = "member"
Role = "master"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-master${count.index}",
"kubernetes.io/cluster/${var.aws_cluster_name}", "member",
"Role", "master"
))
}
resource "aws_elb_attachment" "attach_master_nodes" {
@@ -110,22 +106,18 @@ resource "aws_instance" "k8s-etcd" {
count = var.aws_etcd_num
availability_zone = element(slice(data.aws_availability_zones.available.names, 0, length(var.aws_cidr_subnets_public) <= length(data.aws_availability_zones.available.names) ? length(var.aws_cidr_subnets_public) : length(data.aws_availability_zones.available.names)), count.index)
availability_zone = element(slice(data.aws_availability_zones.available.names, 0, 2), count.index)
subnet_id = element(module.aws-vpc.aws_subnet_ids_private, count.index)
vpc_security_group_ids = module.aws-vpc.aws_security_group
root_block_device {
volume_size = var.aws_etcd_disk_size
}
key_name = var.AWS_SSH_KEY_NAME
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-etcd${count.index}"
"kubernetes.io/cluster/${var.aws_cluster_name}" = "member"
Role = "etcd"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-etcd${count.index}",
"kubernetes.io/cluster/${var.aws_cluster_name}", "member",
"Role", "etcd"
))
}
resource "aws_instance" "k8s-worker" {
@@ -134,23 +126,19 @@ resource "aws_instance" "k8s-worker" {
count = var.aws_kube_worker_num
availability_zone = element(slice(data.aws_availability_zones.available.names, 0, length(var.aws_cidr_subnets_public) <= length(data.aws_availability_zones.available.names) ? length(var.aws_cidr_subnets_public) : length(data.aws_availability_zones.available.names)), count.index)
availability_zone = element(slice(data.aws_availability_zones.available.names, 0, 2), count.index)
subnet_id = element(module.aws-vpc.aws_subnet_ids_private, count.index)
vpc_security_group_ids = module.aws-vpc.aws_security_group
root_block_device {
volume_size = var.aws_kube_worker_disk_size
}
iam_instance_profile = module.aws-iam.kube-worker-profile
key_name = var.AWS_SSH_KEY_NAME
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-worker${count.index}"
"kubernetes.io/cluster/${var.aws_cluster_name}" = "member"
Role = "worker"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-worker${count.index}",
"kubernetes.io/cluster/${var.aws_cluster_name}", "member",
"Role", "worker"
))
}
/*
@@ -164,10 +152,10 @@ data "template_file" "inventory" {
public_ip_address_bastion = join("\n", formatlist("bastion ansible_host=%s", aws_instance.bastion-server.*.public_ip))
connection_strings_master = join("\n", formatlist("%s ansible_host=%s", aws_instance.k8s-master.*.private_dns, aws_instance.k8s-master.*.private_ip))
connection_strings_node = join("\n", formatlist("%s ansible_host=%s", aws_instance.k8s-worker.*.private_dns, aws_instance.k8s-worker.*.private_ip))
connection_strings_etcd = join("\n", formatlist("%s ansible_host=%s", aws_instance.k8s-etcd.*.private_dns, aws_instance.k8s-etcd.*.private_ip))
list_master = join("\n", aws_instance.k8s-master.*.private_dns)
list_node = join("\n", aws_instance.k8s-worker.*.private_dns)
connection_strings_etcd = join("\n", formatlist("%s ansible_host=%s", aws_instance.k8s-etcd.*.private_dns, aws_instance.k8s-etcd.*.private_ip))
list_etcd = join("\n", ((var.aws_etcd_num > 0) ? (aws_instance.k8s-etcd.*.private_dns) : (aws_instance.k8s-master.*.private_dns)))
list_etcd = join("\n", aws_instance.k8s-etcd.*.private_dns)
elb_api_fqdn = "apiserver_loadbalancer_domain_name=\"${module.aws-elb.aws_elb_api_fqdn}\""
}
}

View File

@@ -2,9 +2,9 @@ resource "aws_security_group" "aws-elb" {
name = "kubernetes-${var.aws_cluster_name}-securitygroup-elb"
vpc_id = var.aws_vpc_id
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-securitygroup-elb"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-securitygroup-elb"
))
}
resource "aws_security_group_rule" "aws-allow-api-access" {
@@ -42,7 +42,7 @@ resource "aws_elb" "aws-elb-api" {
healthy_threshold = 2
unhealthy_threshold = 2
timeout = 3
target = "HTTPS:${var.k8s_secure_api_port}/healthz"
target = "TCP:${var.k8s_secure_api_port}"
interval = 30
}
@@ -51,7 +51,7 @@ resource "aws_elb" "aws-elb-api" {
connection_draining = true
connection_draining_timeout = 400
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-elb-api"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-elb-api"
))
}

View File

@@ -16,15 +16,15 @@ variable "k8s_secure_api_port" {
variable "aws_avail_zones" {
description = "Availability Zones Used"
type = list(string)
type = "list"
}
variable "aws_subnet_ids_public" {
description = "IDs of Public Subnets"
type = list(string)
type = "list"
}
variable "default_tags" {
description = "Tags for all resources"
type = map(string)
type = "map"
}

View File

@@ -1,6 +1,6 @@
#Add AWS Roles for Kubernetes
resource "aws_iam_role" "kube_control_plane" {
resource "aws_iam_role" "kube-master" {
name = "kubernetes-${var.aws_cluster_name}-master"
assume_role_policy = <<EOF
@@ -40,9 +40,9 @@ EOF
#Add AWS Policies for Kubernetes
resource "aws_iam_role_policy" "kube_control_plane" {
resource "aws_iam_role_policy" "kube-master" {
name = "kubernetes-${var.aws_cluster_name}-master"
role = aws_iam_role.kube_control_plane.id
role = aws_iam_role.kube-master.id
policy = <<EOF
{
@@ -130,9 +130,9 @@ EOF
#Create AWS Instance Profiles
resource "aws_iam_instance_profile" "kube_control_plane" {
resource "aws_iam_instance_profile" "kube-master" {
name = "kube_${var.aws_cluster_name}_master_profile"
role = aws_iam_role.kube_control_plane.name
role = aws_iam_role.kube-master.name
}
resource "aws_iam_instance_profile" "kube-worker" {

View File

@@ -1,5 +1,5 @@
output "kube_control_plane-profile" {
value = aws_iam_instance_profile.kube_control_plane.name
output "kube-master-profile" {
value = aws_iam_instance_profile.kube-master.name
}
output "kube-worker-profile" {

View File

@@ -5,9 +5,9 @@ resource "aws_vpc" "cluster-vpc" {
enable_dns_support = true
enable_dns_hostnames = true
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-vpc"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-vpc"
))
}
resource "aws_eip" "cluster-nat-eip" {
@@ -18,9 +18,9 @@ resource "aws_eip" "cluster-nat-eip" {
resource "aws_internet_gateway" "cluster-vpc-internetgw" {
vpc_id = aws_vpc.cluster-vpc.id
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-internetgw"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-internetgw"
))
}
resource "aws_subnet" "cluster-vpc-subnets-public" {
@@ -29,10 +29,10 @@ resource "aws_subnet" "cluster-vpc-subnets-public" {
availability_zone = element(var.aws_avail_zones, count.index)
cidr_block = element(var.aws_cidr_subnets_public, count.index)
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-${element(var.aws_avail_zones, count.index)}-public"
"kubernetes.io/cluster/${var.aws_cluster_name}" = "member"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-${element(var.aws_avail_zones, count.index)}-public",
"kubernetes.io/cluster/${var.aws_cluster_name}", "member"
))
}
resource "aws_nat_gateway" "cluster-nat-gateway" {
@@ -47,9 +47,9 @@ resource "aws_subnet" "cluster-vpc-subnets-private" {
availability_zone = element(var.aws_avail_zones, count.index)
cidr_block = element(var.aws_cidr_subnets_private, count.index)
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-${element(var.aws_avail_zones, count.index)}-private"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-${element(var.aws_avail_zones, count.index)}-private"
))
}
#Routing in VPC
@@ -64,9 +64,9 @@ resource "aws_route_table" "kubernetes-public" {
gateway_id = aws_internet_gateway.cluster-vpc-internetgw.id
}
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-routetable-public"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-routetable-public"
))
}
resource "aws_route_table" "kubernetes-private" {
@@ -78,9 +78,9 @@ resource "aws_route_table" "kubernetes-private" {
nat_gateway_id = element(aws_nat_gateway.cluster-nat-gateway.*.id, count.index)
}
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-routetable-private-${count.index}"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-routetable-private-${count.index}"
))
}
resource "aws_route_table_association" "kubernetes-public" {
@@ -101,9 +101,9 @@ resource "aws_security_group" "kubernetes" {
name = "kubernetes-${var.aws_cluster_name}-securitygroup"
vpc_id = aws_vpc.cluster-vpc.id
tags = merge(var.default_tags, tomap({
Name = "kubernetes-${var.aws_cluster_name}-securitygroup"
}))
tags = merge(var.default_tags, map(
"Name", "kubernetes-${var.aws_cluster_name}-securitygroup"
))
}
resource "aws_security_group_rule" "allow-all-ingress" {

View File

@@ -8,20 +8,20 @@ variable "aws_cluster_name" {
variable "aws_avail_zones" {
description = "AWS Availability Zones Used"
type = list(string)
type = "list"
}
variable "aws_cidr_subnets_private" {
description = "CIDR Blocks for private subnets in Availability zones"
type = list(string)
type = "list"
}
variable "aws_cidr_subnets_public" {
description = "CIDR Blocks for public subnets in Availability zones"
type = list(string)
type = "list"
}
variable "default_tags" {
description = "Default tags for all resources"
type = map(string)
type = "map"
}

View File

@@ -11,7 +11,7 @@ output "workers" {
}
output "etcd" {
value = join("\n", ((var.aws_etcd_num > 0) ? (aws_instance.k8s-etcd.*.private_ip) : (aws_instance.k8s-master.*.private_ip)))
value = join("\n", aws_instance.k8s-etcd.*.private_ip)
}
output "aws_elb_api_fqdn" {

View File

@@ -9,8 +9,6 @@ aws_cidr_subnets_private = ["10.250.192.0/20", "10.250.208.0/20"]
aws_cidr_subnets_public = ["10.250.224.0/20", "10.250.240.0/20"]
#Bastion Host
aws_bastion_num = 1
aws_bastion_size = "t2.medium"
#Kubernetes Cluster
@@ -19,26 +17,22 @@ aws_kube_master_num = 3
aws_kube_master_size = "t2.medium"
aws_kube_master_disk_size = 50
aws_etcd_num = 3
aws_etcd_size = "t2.medium"
aws_etcd_disk_size = 50
aws_kube_worker_num = 4
aws_kube_worker_size = "t2.medium"
aws_kube_worker_disk_size = 50
#Settings AWS ELB
aws_elb_api_port = 6443
k8s_secure_api_port = 6443
kube_insecure_apiserver_address = "0.0.0.0"
default_tags = {
# Env = "devtest" # Product = "kubernetes"
}

View File

@@ -7,21 +7,22 @@ ${public_ip_address_bastion}
[bastion]
${public_ip_address_bastion}
[kube_control_plane]
[kube-master]
${list_master}
[kube_node]
[kube-node]
${list_node}
[etcd]
${list_etcd}
[calico_rr]
[k8s_cluster:children]
kube_node
kube_control_plane
calico_rr
[k8s-cluster:children]
kube-node
kube-master
[k8s_cluster:vars]
[k8s-cluster:vars]
${elb_api_fqdn}

View File

@@ -6,34 +6,26 @@ aws_vpc_cidr_block = "10.250.192.0/18"
aws_cidr_subnets_private = ["10.250.192.0/20", "10.250.208.0/20"]
aws_cidr_subnets_public = ["10.250.224.0/20", "10.250.240.0/20"]
# single AZ deployment
#aws_cidr_subnets_private = ["10.250.192.0/20"]
#aws_cidr_subnets_public = ["10.250.224.0/20"]
# 3+ AZ deployment
#aws_cidr_subnets_private = ["10.250.192.0/24","10.250.193.0/24","10.250.194.0/24","10.250.195.0/24"]
#aws_cidr_subnets_public = ["10.250.224.0/24","10.250.225.0/24","10.250.226.0/24","10.250.227.0/24"]
#Bastion Host
aws_bastion_num = 1
aws_bastion_size = "t3.small"
aws_bastion_size = "t2.medium"
#Kubernetes Cluster
aws_kube_master_num = 3
aws_kube_master_size = "t3.medium"
aws_kube_master_disk_size = 50
aws_etcd_num = 0
aws_etcd_size = "t3.medium"
aws_etcd_disk_size = 50
aws_kube_master_num = 3
aws_kube_master_size = "t2.medium"
aws_kube_worker_num = 4
aws_kube_worker_size = "t3.medium"
aws_kube_worker_disk_size = 50
aws_etcd_num = 3
aws_etcd_size = "t2.medium"
aws_kube_worker_num = 4
aws_kube_worker_size = "t2.medium"
#Settings AWS ELB
aws_elb_api_port = 6443
k8s_secure_api_port = 6443
aws_elb_api_port = 6443
k8s_secure_api_port = 6443
kube_insecure_apiserver_address = "0.0.0.0"
default_tags = {
# Env = "devtest"

View File

@@ -8,26 +8,25 @@ aws_cidr_subnets_public = ["10.250.224.0/20","10.250.240.0/20"]
aws_avail_zones = ["eu-central-1a","eu-central-1b"]
#Bastion Host
aws_bastion_num = 1
aws_bastion_size = "t3.small"
aws_bastion_ami = "ami-5900cc36"
aws_bastion_size = "t2.small"
#Kubernetes Cluster
aws_kube_master_num = 3
aws_kube_master_size = "t3.medium"
aws_kube_master_disk_size = 50
aws_kube_master_size = "t2.medium"
aws_etcd_num = 3
aws_etcd_size = "t3.medium"
aws_etcd_disk_size = 50
aws_etcd_size = "t2.medium"
aws_kube_worker_num = 4
aws_kube_worker_size = "t3.medium"
aws_kube_worker_disk_size = 50
aws_kube_worker_size = "t2.medium"
aws_cluster_ami = "ami-903df7ff"
#Settings AWS ELB
aws_elb_api_port = 6443
k8s_secure_api_port = 6443
default_tags = { }
inventory_file = "../../../inventory/hosts"
kube_insecure_apiserver_address = 0.0.0.0

View File

@@ -25,7 +25,7 @@ data "aws_ami" "distro" {
filter {
name = "name"
values = ["debian-10-amd64-*"]
values = ["ubuntu/images/hvm-ssd/ubuntu-bionic-18.04-amd64-server-*"]
}
filter {
@@ -33,7 +33,7 @@ data "aws_ami" "distro" {
values = ["hvm"]
}
owners = ["136693071363"] # Debian-10
owners = ["099720109477"] # Canonical
}
//AWS VPC Variables
@@ -44,12 +44,12 @@ variable "aws_vpc_cidr_block" {
variable "aws_cidr_subnets_private" {
description = "CIDR Blocks for private subnets in Availability Zones"
type = list(string)
type = "list"
}
variable "aws_cidr_subnets_public" {
description = "CIDR Blocks for public subnets in Availability Zones"
type = list(string)
type = "list"
}
//AWS EC2 Settings
@@ -63,18 +63,10 @@ variable "aws_bastion_size" {
* The number should be divisable by the number of used
* AWS Availability Zones without an remainder.
*/
variable "aws_bastion_num" {
description = "Number of Bastion Nodes"
}
variable "aws_kube_master_num" {
description = "Number of Kubernetes Master Nodes"
}
variable "aws_kube_master_disk_size" {
description = "Disk size for Kubernetes Master Nodes (in GiB)"
}
variable "aws_kube_master_size" {
description = "Instance size of Kube Master Nodes"
}
@@ -83,10 +75,6 @@ variable "aws_etcd_num" {
description = "Number of etcd Nodes"
}
variable "aws_etcd_disk_size" {
description = "Disk size for etcd Nodes (in GiB)"
}
variable "aws_etcd_size" {
description = "Instance size of etcd Nodes"
}
@@ -95,10 +83,6 @@ variable "aws_kube_worker_num" {
description = "Number of Kubernetes Worker Nodes"
}
variable "aws_kube_worker_disk_size" {
description = "Disk size for Kubernetes Worker Nodes (in GiB)"
}
variable "aws_kube_worker_size" {
description = "Instance size of Kubernetes Worker Nodes"
}
@@ -117,7 +101,7 @@ variable "k8s_secure_api_port" {
variable "default_tags" {
description = "Default tags for all resources"
type = map(string)
type = "map"
}
variable "inventory_file" {

View File

@@ -1,154 +0,0 @@
# Kubernetes on Exoscale with Terraform
Provision a Kubernetes cluster on [Exoscale](https://www.exoscale.com/) using Terraform and Kubespray
## Overview
The setup looks like following
```text
Kubernetes cluster
+-----------------------+
+---------------+ | +--------------+ |
| | | | +--------------+ |
| API server LB +---------> | | | |
| | | | | Master/etcd | |
+---------------+ | | | node(s) | |
| +-+ | |
| +--------------+ |
| ^ |
| | |
| v |
+---------------+ | +--------------+ |
| | | | +--------------+ |
| Ingress LB +---------> | | | |
| | | | | Worker | |
+---------------+ | | | node(s) | |
| +-+ | |
| +--------------+ |
+-----------------------+
```
## Requirements
* Terraform 0.13.0 or newer
*0.12 also works if you modify the provider block to include version and remove all `versions.tf` files*
## Quickstart
NOTE: *Assumes you are at the root of the kubespray repo*
Copy the sample inventory for your cluster and copy the default terraform variables.
```bash
CLUSTER=my-exoscale-cluster
cp -r inventory/sample inventory/$CLUSTER
cp contrib/terraform/exoscale/default.tfvars inventory/$CLUSTER/
cd inventory/$CLUSTER
```
Edit `default.tfvars` to match your setup. You MUST, at the very least, change `ssh_public_keys`.
```bash
# Ensure $EDITOR points to your favorite editor, e.g., vim, emacs, VS Code, etc.
$EDITOR default.tfvars
```
For authentication you can use the credentials file `~/.cloudstack.ini` or `./cloudstack.ini`.
The file should look like something like this:
```ini
[cloudstack]
key = <API key>
secret = <API secret>
```
Follow the [Exoscale IAM Quick-start](https://community.exoscale.com/documentation/iam/quick-start/) to learn how to generate API keys.
### Encrypted credentials
To have the credentials encrypted at rest, you can use [sops](https://github.com/mozilla/sops) and only decrypt the credentials at runtime.
```bash
cat << EOF > cloudstack.ini
[cloudstack]
key =
secret =
EOF
sops --encrypt --in-place --pgp <PGP key fingerprint> cloudstack.ini
sops cloudstack.ini
```
Run terraform to create the infrastructure
```bash
terraform init ../../contrib/terraform/exoscale
terraform apply -var-file default.tfvars ../../contrib/terraform/exoscale
```
If your cloudstack credentials file is encrypted using sops, run the following:
```bash
terraform init ../../contrib/terraform/exoscale
sops exec-file -no-fifo cloudstack.ini 'CLOUDSTACK_CONFIG={} terraform apply -var-file default.tfvars ../../contrib/terraform/exoscale'
```
You should now have a inventory file named `inventory.ini` that you can use with kubespray.
You can now copy your inventory file and use it with kubespray to set up a cluster.
You can type `terraform output` to find out the IP addresses of the nodes, as well as control-plane and data-plane load-balancer.
It is a good idea to check that you have basic SSH connectivity to the nodes. You can do that by:
```bash
ansible -i inventory.ini -m ping all
```
Example to use this with the default sample inventory:
```bash
ansible-playbook -i inventory.ini ../../cluster.yml -b -v
```
## Teardown
The Kubernetes cluster cannot create any load-balancers or disks, hence, teardown is as simple as Terraform destroy:
```bash
terraform destroy -var-file default.tfvars ../../contrib/terraform/exoscale
```
## Variables
### Required
* `ssh_public_keys`: List of public SSH keys to install on all machines
* `zone`: The zone where to run the cluster
* `machines`: Machines to provision. Key of this object will be used as the name of the machine
* `node_type`: The role of this node *(master|worker)*
* `size`: The size to use
* `boot_disk`: The boot disk to use
* `image_name`: Name of the image
* `root_partition_size`: Size *(in GB)* for the root partition
* `ceph_partition_size`: Size *(in GB)* for the partition for rook to use as ceph storage. *(Set to 0 to disable)*
* `node_local_partition_size`: Size *(in GB)* for the partition for node-local-storage. *(Set to 0 to disable)*
* `ssh_whitelist`: List of IP ranges (CIDR) that will be allowed to ssh to the nodes
* `api_server_whitelist`: List of IP ranges (CIDR) that will be allowed to connect to the API server
* `nodeport_whitelist`: List of IP ranges (CIDR) that will be allowed to connect to the kubernetes nodes on port 30000-32767 (kubernetes nodeports)
### Optional
* `prefix`: Prefix to use for all resources, required to be unique for all clusters in the same project *(Defaults to `default`)*
An example variables file can be found `default.tfvars`
## Known limitations
### Only single disk
Since Exoscale doesn't support additional disks to be mounted onto an instance, this script has the ability to create partitions for [Rook](https://rook.io/) and [node-local-storage](https://kubernetes.io/docs/concepts/storage/volumes/#local).
### No Kubernetes API
The current solution doesn't use the [Exoscale Kubernetes cloud controller](https://github.com/exoscale/exoscale-cloud-controller-manager).
This means that we need to set up a HTTP(S) loadbalancer in front of all workers and set the Ingress controller to DaemonSet mode.

View File

@@ -1,65 +0,0 @@
prefix = "default"
zone = "ch-gva-2"
inventory_file = "inventory.ini"
ssh_public_keys = [
# Put your public SSH key here
"ssh-rsa I-did-not-read-the-docs",
"ssh-rsa I-did-not-read-the-docs 2",
]
machines = {
"master-0" : {
"node_type" : "master",
"size" : "Medium",
"boot_disk" : {
"image_name" : "Linux Ubuntu 20.04 LTS 64-bit",
"root_partition_size" : 50,
"node_local_partition_size" : 0,
"ceph_partition_size" : 0
}
},
"worker-0" : {
"node_type" : "worker",
"size" : "Large",
"boot_disk" : {
"image_name" : "Linux Ubuntu 20.04 LTS 64-bit",
"root_partition_size" : 50,
"node_local_partition_size" : 0,
"ceph_partition_size" : 0
}
},
"worker-1" : {
"node_type" : "worker",
"size" : "Large",
"boot_disk" : {
"image_name" : "Linux Ubuntu 20.04 LTS 64-bit",
"root_partition_size" : 50,
"node_local_partition_size" : 0,
"ceph_partition_size" : 0
}
},
"worker-2" : {
"node_type" : "worker",
"size" : "Large",
"boot_disk" : {
"image_name" : "Linux Ubuntu 20.04 LTS 64-bit",
"root_partition_size" : 50,
"node_local_partition_size" : 0,
"ceph_partition_size" : 0
}
}
}
nodeport_whitelist = [
"0.0.0.0/0"
]
ssh_whitelist = [
"0.0.0.0/0"
]
api_server_whitelist = [
"0.0.0.0/0"
]

View File

@@ -1,49 +0,0 @@
provider "exoscale" {}
module "kubernetes" {
source = "./modules/kubernetes-cluster"
prefix = var.prefix
machines = var.machines
ssh_public_keys = var.ssh_public_keys
ssh_whitelist = var.ssh_whitelist
api_server_whitelist = var.api_server_whitelist
nodeport_whitelist = var.nodeport_whitelist
}
#
# Generate ansible inventory
#
data "template_file" "inventory" {
template = file("${path.module}/templates/inventory.tpl")
vars = {
connection_strings_master = join("\n", formatlist("%s ansible_user=ubuntu ansible_host=%s ip=%s etcd_member_name=etcd%d",
keys(module.kubernetes.master_ip_addresses),
values(module.kubernetes.master_ip_addresses).*.public_ip,
values(module.kubernetes.master_ip_addresses).*.private_ip,
range(1, length(module.kubernetes.master_ip_addresses) + 1)))
connection_strings_worker = join("\n", formatlist("%s ansible_user=ubuntu ansible_host=%s ip=%s",
keys(module.kubernetes.worker_ip_addresses),
values(module.kubernetes.worker_ip_addresses).*.public_ip,
values(module.kubernetes.worker_ip_addresses).*.private_ip))
list_master = join("\n", keys(module.kubernetes.master_ip_addresses))
list_worker = join("\n", keys(module.kubernetes.worker_ip_addresses))
api_lb_ip_address = module.kubernetes.control_plane_lb_ip_address
}
}
resource "null_resource" "inventories" {
provisioner "local-exec" {
command = "echo '${data.template_file.inventory.rendered}' > ${var.inventory_file}"
}
triggers = {
template = data.template_file.inventory.rendered
}
}

View File

@@ -1,193 +0,0 @@
data "exoscale_compute_template" "os_image" {
for_each = var.machines
zone = var.zone
name = each.value.boot_disk.image_name
}
data "exoscale_compute" "master_nodes" {
for_each = exoscale_compute.master
id = each.value.id
# Since private IP address is not assigned until the nics are created we need this
depends_on = [exoscale_nic.master_private_network_nic]
}
data "exoscale_compute" "worker_nodes" {
for_each = exoscale_compute.worker
id = each.value.id
# Since private IP address is not assigned until the nics are created we need this
depends_on = [exoscale_nic.worker_private_network_nic]
}
resource "exoscale_network" "private_network" {
zone = var.zone
name = "${var.prefix}-network"
start_ip = cidrhost(var.private_network_cidr, 1)
# cidr -1 = Broadcast address
# cidr -2 = DHCP server address (exoscale specific)
end_ip = cidrhost(var.private_network_cidr, -3)
netmask = cidrnetmask(var.private_network_cidr)
}
resource "exoscale_compute" "master" {
for_each = {
for name, machine in var.machines :
name => machine
if machine.node_type == "master"
}
display_name = "${var.prefix}-${each.key}"
template_id = data.exoscale_compute_template.os_image[each.key].id
size = each.value.size
disk_size = each.value.boot_disk.root_partition_size + each.value.boot_disk.node_local_partition_size + each.value.boot_disk.ceph_partition_size
state = "Running"
zone = var.zone
security_groups = [exoscale_security_group.master_sg.name]
user_data = templatefile(
"${path.module}/templates/cloud-init.tmpl",
{
eip_ip_address = exoscale_ipaddress.ingress_controller_lb.ip_address
node_local_partition_size = each.value.boot_disk.node_local_partition_size
ceph_partition_size = each.value.boot_disk.ceph_partition_size
root_partition_size = each.value.boot_disk.root_partition_size
node_type = "master"
ssh_public_keys = var.ssh_public_keys
}
)
}
resource "exoscale_compute" "worker" {
for_each = {
for name, machine in var.machines :
name => machine
if machine.node_type == "worker"
}
display_name = "${var.prefix}-${each.key}"
template_id = data.exoscale_compute_template.os_image[each.key].id
size = each.value.size
disk_size = each.value.boot_disk.root_partition_size + each.value.boot_disk.node_local_partition_size + each.value.boot_disk.ceph_partition_size
state = "Running"
zone = var.zone
security_groups = [exoscale_security_group.worker_sg.name]
user_data = templatefile(
"${path.module}/templates/cloud-init.tmpl",
{
eip_ip_address = exoscale_ipaddress.ingress_controller_lb.ip_address
node_local_partition_size = each.value.boot_disk.node_local_partition_size
ceph_partition_size = each.value.boot_disk.ceph_partition_size
root_partition_size = each.value.boot_disk.root_partition_size
node_type = "worker"
ssh_public_keys = var.ssh_public_keys
}
)
}
resource "exoscale_nic" "master_private_network_nic" {
for_each = exoscale_compute.master
compute_id = each.value.id
network_id = exoscale_network.private_network.id
}
resource "exoscale_nic" "worker_private_network_nic" {
for_each = exoscale_compute.worker
compute_id = each.value.id
network_id = exoscale_network.private_network.id
}
resource "exoscale_security_group" "master_sg" {
name = "${var.prefix}-master-sg"
description = "Security group for Kubernetes masters"
}
resource "exoscale_security_group_rules" "master_sg_rules" {
security_group_id = exoscale_security_group.master_sg.id
# SSH
ingress {
protocol = "TCP"
cidr_list = var.ssh_whitelist
ports = ["22"]
}
# Kubernetes API
ingress {
protocol = "TCP"
cidr_list = var.api_server_whitelist
ports = ["6443"]
}
}
resource "exoscale_security_group" "worker_sg" {
name = "${var.prefix}-worker-sg"
description = "security group for kubernetes worker nodes"
}
resource "exoscale_security_group_rules" "worker_sg_rules" {
security_group_id = exoscale_security_group.worker_sg.id
# SSH
ingress {
protocol = "TCP"
cidr_list = var.ssh_whitelist
ports = ["22"]
}
# HTTP(S)
ingress {
protocol = "TCP"
cidr_list = ["0.0.0.0/0"]
ports = ["80", "443"]
}
# Kubernetes Nodeport
ingress {
protocol = "TCP"
cidr_list = var.nodeport_whitelist
ports = ["30000-32767"]
}
}
resource "exoscale_ipaddress" "ingress_controller_lb" {
zone = var.zone
healthcheck_mode = "http"
healthcheck_port = 80
healthcheck_path = "/healthz"
healthcheck_interval = 10
healthcheck_timeout = 2
healthcheck_strikes_ok = 2
healthcheck_strikes_fail = 3
}
resource "exoscale_secondary_ipaddress" "ingress_controller_lb" {
for_each = exoscale_compute.worker
compute_id = each.value.id
ip_address = exoscale_ipaddress.ingress_controller_lb.ip_address
}
resource "exoscale_ipaddress" "control_plane_lb" {
zone = var.zone
healthcheck_mode = "tcp"
healthcheck_port = 6443
healthcheck_interval = 10
healthcheck_timeout = 2
healthcheck_strikes_ok = 2
healthcheck_strikes_fail = 3
}
resource "exoscale_secondary_ipaddress" "control_plane_lb" {
for_each = exoscale_compute.master
compute_id = each.value.id
ip_address = exoscale_ipaddress.control_plane_lb.ip_address
}

View File

@@ -1,31 +0,0 @@
output "master_ip_addresses" {
value = {
for key, instance in exoscale_compute.master :
instance.name => {
"private_ip" = contains(keys(data.exoscale_compute.master_nodes), key) ? data.exoscale_compute.master_nodes[key].private_network_ip_addresses[0] : ""
"public_ip" = exoscale_compute.master[key].ip_address
}
}
}
output "worker_ip_addresses" {
value = {
for key, instance in exoscale_compute.worker :
instance.name => {
"private_ip" = contains(keys(data.exoscale_compute.worker_nodes), key) ? data.exoscale_compute.worker_nodes[key].private_network_ip_addresses[0] : ""
"public_ip" = exoscale_compute.worker[key].ip_address
}
}
}
output "cluster_private_network_cidr" {
value = var.private_network_cidr
}
output "ingress_controller_lb_ip_address" {
value = exoscale_ipaddress.ingress_controller_lb.ip_address
}
output "control_plane_lb_ip_address" {
value = exoscale_ipaddress.control_plane_lb.ip_address
}

View File

@@ -1,52 +0,0 @@
#cloud-config
%{ if ceph_partition_size > 0 || node_local_partition_size > 0}
bootcmd:
- [ cloud-init-per, once, move-second-header, sgdisk, --move-second-header, /dev/vda ]
%{ if node_local_partition_size > 0 }
# Create partition for node local storage
- [ cloud-init-per, once, create-node-local-part, parted, --script, /dev/vda, 'mkpart extended ext4 ${root_partition_size}GB %{ if ceph_partition_size == 0 }-1%{ else }${root_partition_size + node_local_partition_size}GB%{ endif }' ]
- [ cloud-init-per, once, create-fs-node-local-part, mkfs.ext4, /dev/vda2 ]
%{ endif }
%{ if ceph_partition_size > 0 }
# Create partition for rook to use for ceph
- [ cloud-init-per, once, create-ceph-part, parted, --script, /dev/vda, 'mkpart extended ${root_partition_size + node_local_partition_size}GB -1' ]
%{ endif }
%{ endif }
ssh_authorized_keys:
%{ for ssh_public_key in ssh_public_keys ~}
- ${ssh_public_key}
%{ endfor ~}
write_files:
- path: /etc/netplan/eth1.yaml
content: |
network:
version: 2
ethernets:
eth1:
dhcp4: true
%{ if node_type == "worker" }
# TODO: When a VM is seen as healthy and is added to the EIP loadbalancer
# pool it no longer can send traffic back to itself via the EIP IP
# address.
# Remove this if it ever gets solved.
- path: /etc/netplan/20-eip-fix.yaml
content: |
network:
version: 2
ethernets:
"lo:0":
match:
name: lo
dhcp4: false
addresses:
- ${eip_ip_address}/32
%{ endif }
runcmd:
- netplan apply
%{ if node_local_partition_size > 0 }
- mkdir -p /mnt/disks/node-local-storage
- chown nobody:nogroup /mnt/disks/node-local-storage
- mount /dev/vda2 /mnt/disks/node-local-storage
%{ endif }

View File

@@ -1,42 +0,0 @@
variable "zone" {
type = string
# This is currently the only zone that is supposed to be supporting
# so called "managed private networks".
# See: https://www.exoscale.com/syslog/introducing-managed-private-networks
default = "ch-gva-2"
}
variable "prefix" {}
variable "machines" {
type = map(object({
node_type = string
size = string
boot_disk = object({
image_name = string
root_partition_size = number
ceph_partition_size = number
node_local_partition_size = number
})
}))
}
variable "ssh_public_keys" {
type = list(string)
}
variable "ssh_whitelist" {
type = list(string)
}
variable "api_server_whitelist" {
type = list(string)
}
variable "nodeport_whitelist" {
type = list(string)
}
variable "private_network_cidr" {
default = "172.0.10.0/24"
}

View File

@@ -1,9 +0,0 @@
terraform {
required_providers {
exoscale = {
source = "exoscale/exoscale"
version = ">= 0.21"
}
}
required_version = ">= 0.13"
}

View File

@@ -1,15 +0,0 @@
output "master_ips" {
value = module.kubernetes.master_ip_addresses
}
output "worker_ips" {
value = module.kubernetes.worker_ip_addresses
}
output "ingress_controller_lb_ip_address" {
value = module.kubernetes.ingress_controller_lb_ip_address
}
output "control_plane_lb_ip_address" {
value = module.kubernetes.control_plane_lb_ip_address
}

View File

@@ -1,65 +0,0 @@
prefix = "default"
zone = "ch-gva-2"
inventory_file = "inventory.ini"
ssh_public_keys = [
# Put your public SSH key here
"ssh-rsa I-did-not-read-the-docs",
"ssh-rsa I-did-not-read-the-docs 2",
]
machines = {
"master-0" : {
"node_type" : "master",
"size" : "Small",
"boot_disk" : {
"image_name" : "Linux Ubuntu 20.04 LTS 64-bit",
"root_partition_size" : 50,
"node_local_partition_size" : 0,
"ceph_partition_size" : 0
}
},
"worker-0" : {
"node_type" : "worker",
"size" : "Large",
"boot_disk" : {
"image_name" : "Linux Ubuntu 20.04 LTS 64-bit",
"root_partition_size" : 50,
"node_local_partition_size" : 0,
"ceph_partition_size" : 0
}
},
"worker-1" : {
"node_type" : "worker",
"size" : "Large",
"boot_disk" : {
"image_name" : "Linux Ubuntu 20.04 LTS 64-bit",
"root_partition_size" : 50,
"node_local_partition_size" : 0,
"ceph_partition_size" : 0
}
},
"worker-2" : {
"node_type" : "worker",
"size" : "Large",
"boot_disk" : {
"image_name" : "Linux Ubuntu 20.04 LTS 64-bit",
"root_partition_size" : 50,
"node_local_partition_size" : 0,
"ceph_partition_size" : 0
}
}
}
nodeport_whitelist = [
"0.0.0.0/0"
]
ssh_whitelist = [
"0.0.0.0/0"
]
api_server_whitelist = [
"0.0.0.0/0"
]

View File

@@ -1 +0,0 @@
../../../../inventory/sample/group_vars

View File

@@ -1,19 +0,0 @@
[all]
${connection_strings_master}
${connection_strings_worker}
[kube_control_plane]
${list_master}
[kube_control_plane:vars]
supplementary_addresses_in_ssl_keys = [ "${api_lb_ip_address}" ]
[etcd]
${list_master}
[kube_node]
${list_worker}
[k8s_cluster:children]
kube_control_plane
kube_node

View File

@@ -1,46 +0,0 @@
variable "zone" {
description = "The zone where to run the cluster"
}
variable "prefix" {
description = "Prefix for resource names"
default = "default"
}
variable "machines" {
description = "Cluster machines"
type = map(object({
node_type = string
size = string
boot_disk = object({
image_name = string
root_partition_size = number
ceph_partition_size = number
node_local_partition_size = number
})
}))
}
variable "ssh_public_keys" {
description = "List of public SSH keys which are injected into the VMs."
type = list(string)
}
variable "ssh_whitelist" {
description = "List of IP ranges (CIDR) to whitelist for ssh"
type = list(string)
}
variable "api_server_whitelist" {
description = "List of IP ranges (CIDR) to whitelist for kubernetes api server"
type = list(string)
}
variable "nodeport_whitelist" {
description = "List of IP ranges (CIDR) to whitelist for kubernetes nodeports"
type = list(string)
}
variable "inventory_file" {
description = "Where to store the generated inventory file"
}

Some files were not shown because too many files have changed in this diff Show More