Karim

Posted on Jun 3 • Originally published at deep75.Medium on Jun 3

Une autre manière de faire fonctionner OpenStack au-dessus de Kubernetes avec Atmosphere …

#kubernetes #openstack #docker #phoenixnap

Plusieurs projets de déploiement d’OpenStack qui reposent sur un cluster Kubernetes existent à commencer par OpenStack Helm :

Mais c’est aussi le cas du projet open source Atmosphere du fournisseur Cloud Vexxhost qui fournit un large éventail de fonctionnalités d’infrastructure, notamment des machines virtuelles, Kubernetes, du bare metal, du stockage en bloc et en objet, des équilibreurs de charge en tant que service, et plus encore …

Atmosphere deployment tool | VEXXHOST

Atmosphere fait tourner OpenStack au-dessus de Kubernetes. Cette configuration permet des déploiements, des mises à niveau et des contrôles de santé simples et facile.

VEXXHOST Launches Atmosphere, a New Open Source, OpenStack Deployment Tool

GitHub - vexxhost/atmosphere: Simple & easy private cloud platform featuring VMs, Kubernetes & bare-metal

Je vais simplement suivre le modus operandi proposé par Vexxhost pour un serveur Bare Metal :

Pour cela je pars d’un serveur dans PhoenixNAP sous Ubuntu 22.04 LTS :

Mise à jour de ce dernier et installation des paquets nécessaires au lancement du déploiement d’Atmosphere :

ubuntu@atmosphere:~$ sudo apt-get update && sudo apt-get install git python3-pip -y && sudo pip install poetry

ubuntu@atmosphere:~$ git clone https://github.com/vexxhost/atmosphere.git
Cloning into 'atmosphere'...
remote: Enumerating objects: 25111, done.
remote: Counting objects: 100% (825/825), done.
remote: Compressing objects: 100% (398/398), done.
remote: Total 25111 (delta 377), reused 704 (delta 306), pack-reused 24286
Receiving objects: 100% (25111/25111), 10.78 MiB | 50.17 MiB/s, done.
Resolving deltas: 100% (14299/14299), done.
ubuntu@atmosphere:~$ cd atmosphere/
ubuntu@atmosphere:~/atmosphere$ ls
CHANGELOG.md Jenkinsfile build doc galaxy.yml hack meta playbooks pyproject.toml test-requirements.txt tox.ini
Dockerfile README.md charts flake.lock go.mod images mkdocs.yml plugins release-please-config.json tests zuul.d
Earthfile atmosphere cmd flake.nix go.sum internal molecule poetry.lock roles tools

ubuntu@atmosphere:~/atmosphere$ sudo poetry install --with dev
Creating virtualenv atmosphere-NEvTTHEY-py3.10 in /root/.cache/pypoetry/virtualenvs
Installing dependencies from lock file

Package operations: 89 installs, 0 updates, 0 removals

  - Installing attrs (23.2.0)
  - Installing pycparser (2.22)
  - Installing rpds-py (0.18.0)
  - Installing cffi (1.16.0)
  - Installing markupsafe (2.1.5)
  - Installing mdurl (0.1.2)
  - Installing referencing (0.35.0)
  - Installing cryptography (42.0.5)
  - Installing jinja2 (3.1.3)
  - Installing jsonschema-specifications (2023.12.1)
  - Installing markdown-it-py (3.0.0)
  - Installing packaging (24.0)
  - Installing pbr (6.0.0)
  - Installing pygments (2.17.2)
  - Installing pyyaml (6.0.1)
  - Installing resolvelib (1.0.1)
  - Installing wrapt (1.16.0)
  - Installing ansible-core (2.16.6)
  - Installing bracex (2.4)
  - Installing certifi (2024.2.2)
  - Installing charset-normalizer (3.3.2)
  - Installing click (8.1.7)
  - Installing debtcollector (3.0.0)
  - Installing idna (3.7)
  - Installing iso8601 (2.1.0)
  - Installing jsonschema (4.21.1)
  - Installing netaddr (0.8.0)
  - Installing netifaces (0.11.0)
  - Installing oslo-i18n (6.3.0)
  - Installing pyparsing (3.1.2)
  - Installing rich (13.7.1)
  - Installing subprocess-tee (0.4.1)
  - Installing tzdata (2024.1)
  - Installing urllib3 (2.2.1)
  - Installing ansible-compat (4.1.11)
  - Installing click-help-colors (0.9.4)
  - Installing decorator (5.1.1)
  - Installing distro (1.9.0)
  - Installing enrich (1.2.7)
  - Installing exceptiongroup (1.2.1)
  - Installing iniconfig (2.0.0)
  - Installing jsonpointer (2.4)
  - Installing mccabe (0.7.0)
  - Installing msgpack (1.0.8)
  - Installing os-service-types (1.7.0)
  - Installing oslo-utils (7.1.0)
  - Installing pluggy (1.5.0)
  - Installing pycodestyle (2.9.1)
  - Installing pyflakes (2.5.0)
  - Installing requests (2.31.0)
  - Installing rfc3986 (2.0.0)
  - Installing six (1.16.0)
  - Installing stevedore (5.2.0)
  - Installing tomli (2.0.1)
  - Installing typing-extensions (4.11.0)
  - Installing wcmatch (8.5.1)
  - Installing appdirs (1.4.4)
  - Installing coverage (7.5.0)
  - Installing docker (7.0.0)
  - Installing dogpile-cache (1.3.2)
  - Installing execnet (2.1.1)
  - Installing flake8 (5.0.4)
  - Installing isort (5.13.2)
  - Installing jmespath (1.0.1)
  - Installing jsonpatch (1.33)
  - Installing keystoneauth1 (5.6.0)
  - Installing molecule (6.0.3)
  - Installing munch (4.0.0)
  - Installing oslo-config (9.4.0)
  - Installing oslo-context (5.5.0)
  - Installing oslo-serialization (5.4.0)
  - Installing py (1.11.0)
  - Installing pyinotify (0.9.6)
  - Installing pytest (7.4.4)
  - Installing python-dateutil (2.9.0.post0)
  - Installing regex (2024.4.28)
  - Installing requestsexceptions (1.4.0)
  - Installing selinux (0.3.0)
  - Installing docker-image-py (0.1.12)
  - Installing flake8-isort (4.2.0)
  - Installing molecule-plugins (23.5.3)
  - Installing openstacksdk (0.62.0)
  - Installing oslo-log (5.5.1)
  - Installing pytest-cov (3.0.0)
  - Installing pytest-forked (1.6.0)
  - Installing pytest-mock (3.14.0)
  - Installing pytest-xdist (3.6.1)
  - Installing rjsonnet (0.5.4)
  - Installing ruyaml (0.91.0)

Installing the current project: atmosphere (1.10.4.post186.dev0+779cb921)

Très rapidement, je lance le déploiement qui prend un peu moins d’une heure ici :

ubuntu@atmosphere:~/atmosphere$ sudo poetry run molecule converge -s aio
INFO aio scenario test matrix: dependency, create, prepare, converge
INFO Performing prerun with role_name_check=0...
INFO Running aio > dependency
WARNING Skipping, missing the requirements file.
WARNING Skipping, missing the requirements file.
INFO Running aio > create

PLAY [Wait for user to read warning] *******************************************

TASK [Gathering Facts] *********************************************************
Monday 03 June 2024 12:02:39 +0000 (0:00:00.020) 0:00:00.020 ***********
ok: [localhost]
.
.
.
.
.
PLAY [Configure networking] ****************************************************

TASK [Gathering Facts] *********************************************************
Monday 03 June 2024 12:49:26 +0000 (0:00:00.418) 0:45:34.621 ***********
ok: [instance]

TASK [Add IP address to "br-ex"] ***********************************************
Monday 03 June 2024 12:49:28 +0000 (0:00:02.159) 0:45:36.781 ***********
ok: [instance]

TASK [Set "br-ex" interface to "up"] *******************************************
Monday 03 June 2024 12:49:28 +0000 (0:00:00.153) 0:45:36.934 ***********
ok: [instance]

PLAY RECAP *********************************************************************
instance : ok=669 changed=267 unreachable=0 failed=0 skipped=246 rescued=0 ignored=1   

Monday 03 June 2024 12:49:28 +0000 (0:00:00.304) 0:45:37.239 ***********
===============================================================================
vexxhost.atmosphere.percona_xtradb_cluster : Apply Percona XtraDB cluster - 203.51s
vexxhost.atmosphere.cinder : Deploy Helm chart ------------------------ 157.84s
vexxhost.atmosphere.keycloak : Deploy Helm chart ---------------------- 156.65s
vexxhost.atmosphere.heat : Deploy Helm chart -------------------------- 120.67s
vexxhost.atmosphere.manila : Deploy Helm chart ------------------------ 104.84s
vexxhost.atmosphere.nova : Deploy Helm chart -------------------------- 100.67s
vexxhost.ceph.osd : Install OSDs --------------------------------------- 89.67s
vexxhost.atmosphere.glance : Deploy Helm chart ------------------------- 88.95s
vexxhost.atmosphere.magnum : Deploy Helm chart ------------------------- 88.06s
vexxhost.atmosphere.octavia : Deploy Helm chart ------------------------ 83.51s
vexxhost.atmosphere.keystone : Deploy Helm chart ----------------------- 80.67s
vexxhost.atmosphere.neutron : Deploy Helm chart ------------------------ 74.13s
vexxhost.atmosphere.barbican : Deploy Helm chart ----------------------- 67.48s
vexxhost.ceph.mon : Run Bootstrap coomand ------------------------------ 62.21s
vexxhost.atmosphere.placement : Deploy Helm chart ---------------------- 58.82s
vexxhost.kubernetes.cluster_api : Set node selector for Cluster API components -- 57.15s
vexxhost.atmosphere.glance_image : Download image ---------------------- 53.93s
vexxhost.atmosphere.glance_image : Check if image exists --------------- 50.38s
vexxhost.atmosphere.neutron : Create networks -------------------------- 36.62s
vexxhost.atmosphere.rabbitmq : Deploy cluster -------------------------- 35.65s

Le cluster est déployé et je peux vérifier les accès disponibles localement sur ce serveur physique :

root@instance:~# apt install python3-openstackclient -y

root@instance:~# source openrc
root@instance:~# openstack endpoint list
+----------------------------------+-----------+--------------+-----------------+---------+-----------+------------------------------------------------------------------------------+
| ID | Region | Service Name | Service Type | Enabled | Interface | URL |
+----------------------------------+-----------+--------------+-----------------+---------+-----------+------------------------------------------------------------------------------+
| 0584539654844f7f956088e43836ed4c | RegionOne | swift | object-store | True | public | https://object-store.131-153-200-197.nip.io/swift/v1/%(tenant_id)s |
| 069df7eb37584d91a22a95e7d36493d3 | RegionOne | manila | share | True | internal | http://manila-api.openstack.svc.cluster.local:8786/v1 |
| 08b5e9b96ddf4ce79f270f582dc6bf7b | RegionOne | manilav2 | sharev2 | True | public | https://share.131-153-200-197.nip.io/v2 |
| 0b95720b8c0e4de692acf18b769ce8be | RegionOne | heat-cfn | cloudformation | True | admin | http://heat-cfn.openstack.svc.cluster.local:8000/v1 |
| 120958efeb924a0a98bf03ce807c0ace | RegionOne | cinderv3 | volumev3 | True | internal | http://cinder-api.openstack.svc.cluster.local:8776/v3/%(tenant_id)s |
| 1427e413681e4518ac03d768bb44cb60 | RegionOne | placement | placement | True | admin | http://placement-api.openstack.svc.cluster.local:8778/ |
| 1611e222d7b8447596aa44c0965aec1a | RegionOne | cinderv3 | volumev3 | True | public | https://volume.131-153-200-197.nip.io/v3/%(tenant_id)s |
| 19047158ec1446b5b746682bc8d9dd93 | RegionOne | heat-cfn | cloudformation | True | internal | http://heat-cfn.openstack.svc.cluster.local:8000/v1 |
| 1e04bacc8aae44b3818de40e386cf68e | RegionOne | barbican | key-manager | True | admin | http://barbican-api.openstack.svc.cluster.local:9311/ |
| 27d7b0d81a104b54a3dcb8632297707f | RegionOne | keystone | identity | True | admin | http://keystone-api.openstack.svc.cluster.local:5000/ |
| 2c4218a3da784f98890bcb6ac6cb20ae | RegionOne | heat | orchestration | True | admin | http://heat-api.openstack.svc.cluster.local:8004/v1/%(project_id)s |
| 2f19518bd66c421eb6a5e25f7c93c96e | RegionOne | manila | share | True | public | http://manila.openstack.svc.cluster.local/v1 |
| 3578d37453854cfbafdc18f2be4f0a62 | RegionOne | glance | image | True | public | https://image.131-153-200-197.nip.io/ |
| 3a2e106afdbf43689676c85718319f88 | RegionOne | glance | image | True | admin | http://glance-api.openstack.svc.cluster.local:9292/ |
| 3e3240df5df847bc84c204e0b18783f1 | RegionOne | glance | image | True | internal | http://glance-api.openstack.svc.cluster.local:9292/ |
| 3f51c888203c433ea31d2ca67cf3e359 | RegionOne | manilav2 | sharev2 | True | admin | http://manila-api.openstack.svc.cluster.local:8786/v2 |
| 44f08551b541410d8bb28b3a148dca0f | RegionOne | heat | orchestration | True | internal | http://heat-api.openstack.svc.cluster.local:8004/v1/%(project_id)s |
| 6501cca34cd5401d8561522f062ec126 | RegionOne | magnum | container-infra | True | internal | http://magnum-api.openstack.svc.cluster.local:9511/v1 |
| 65c4f687017d4ff4b774a61cd670de52 | RegionOne | barbican | key-manager | True | public | https://key-manager.131-153-200-197.nip.io/ |
| 695342e32bf84ae48cc0c873227ce1ce | RegionOne | heat | orchestration | True | public | https://orchestration.131-153-200-197.nip.io/v1/%(project_id)s |
| 71af00dd937e49c28827dadbe6d55bbe | RegionOne | swift | object-store | True | internal | http://rook-ceph-rgw-ceph.openstack.svc.cluster.local/swift/v1/%(tenant_id)s |
| 7ada3efd05b64af1ae726d495fe61a6e | RegionOne | cinderv3 | volumev3 | True | admin | http://cinder-api.openstack.svc.cluster.local:8776/v3/%(tenant_id)s |
| 87e74af69cbb4577b4dd29d106aed5fa | RegionOne | magnum | container-infra | True | public | https://container-infra.131-153-200-197.nip.io/v1 |
| 8dae31f3eed946eea30025e56fbfd83a | RegionOne | nova | compute | True | admin | http://nova-api.openstack.svc.cluster.local:8774/v2.1 |
| 8e562226dccc4df0af3c8cdfdec95084 | RegionOne | barbican | key-manager | True | internal | http://barbican-api.openstack.svc.cluster.local:9311/ |
| 8f629f7e748d4c3087bf17aaa0ce2e47 | RegionOne | manila | share | True | admin | http://manila-api.openstack.svc.cluster.local:8786/v1 |
| 91149164a0584145b6344ae1d525457b | RegionOne | keystone | identity | True | public | https://identity.131-153-200-197.nip.io/ |
| a936e3ca42a54b6e97672f20e17c095b | RegionOne | neutron | network | True | admin | http://neutron-server.openstack.svc.cluster.local:9696/ |
| b987ba9af66c4a14ba42d1fd8c1b2285 | RegionOne | placement | placement | True | public | https://placement.131-153-200-197.nip.io/ |
| b9e2e43e90d649d5b6089c7474ba30cd | RegionOne | octavia | load-balancer | True | internal | http://octavia-api.openstack.svc.cluster.local:9876/ |
| bf2839c099f946e39d21a0b6bcad0b87 | RegionOne | neutron | network | True | public | https://network.131-153-200-197.nip.io/ |
| bf959da766554e91972fe1da9cff6d8e | RegionOne | heat-cfn | cloudformation | True | public | https://cloudformation.131-153-200-197.nip.io/v1 |
| d027a62b34ca4eaabc508c04ad8f94e7 | RegionOne | magnum | container-infra | True | admin | http://magnum-api.openstack.svc.cluster.local:9511/v1 |
| d3e857109be14bfabf82e962068b54af | RegionOne | nova | compute | True | public | https://compute.131-153-200-197.nip.io/v2.1 |
| d6985c01aca54ec49e0ec1c728fc271b | RegionOne | nova | compute | True | internal | http://nova-api.openstack.svc.cluster.local:8774/v2.1 |
| d74feae56d034bd7ad421e4588f0731e | RegionOne | octavia | load-balancer | True | admin | http://octavia-api.openstack.svc.cluster.local:9876/ |
| e7bcb22c547e41489cc7d5732bca4a84 | RegionOne | manilav2 | sharev2 | True | internal | http://manila-api.openstack.svc.cluster.local:8786/v2 |
| f005b62e7769497eb54b0c0a3fa3c587 | RegionOne | octavia | load-balancer | True | public | https://load-balancer.131-153-200-197.nip.io/ |
| f40bb006a4984428b581042a78539ef8 | RegionOne | neutron | network | True | internal | http://neutron-server.openstack.svc.cluster.local:9696/ |
| f4d8af60368645c7a5f1d9605afb5494 | RegionOne | placement | placement | True | internal | http://placement-api.openstack.svc.cluster.local:8778/ |
| f5aa8f8ba519480199f224800480e8cb | RegionOne | keystone | identity | True | internal | http://keystone-api.openstack.svc.cluster.local:5000/ |
+----------------------------------+-----------+--------------+-----------------+---------+-----------+------------------------------------------------------------------------------+

Ainsi que des accès via l’Ingress Controller du cluster Kubernetes sous-jacent :

root@instance:~# kubectl cluster-info
Kubernetes control plane is running at https://10.96.240.10:6443
CoreDNS is running at https://10.96.240.10:6443/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy

To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.
root@instance:~# kubectl get nodes -o wide
NAME STATUS ROLES AGE VERSION INTERNAL-IP EXTERNAL-IP OS-IMAGE KERNEL-VERSION CONTAINER-RUNTIME
instance Ready control-plane 60m v1.28.4 131.153.200.197 <none> Ubuntu 22.04.4 LTS 5.15.0-105-generic containerd://1.7.9
root@instance:~# kubectl get ing -A
NAMESPACE NAME CLASS HOSTS ADDRESS PORTS AGE
auth-system keycloak atmosphere keycloak.131-153-200-197.nip.io 10.98.36.135 80, 443 54m
monitoring kube-prometheus-stack-alertmanager atmosphere alertmanager.131-153-200-197.nip.io 10.98.36.135 80, 443 53m
monitoring kube-prometheus-stack-grafana atmosphere grafana.131-153-200-197.nip.io 10.98.36.135 80, 443 53m
monitoring kube-prometheus-stack-prometheus atmosphere prometheus.131-153-200-197.nip.io 10.98.36.135 80, 443 53m
openstack cloudformation atmosphere cloudformation.131-153-200-197.nip.io 10.98.36.135 80, 443 33m
openstack compute atmosphere compute.131-153-200-197.nip.io 10.98.36.135 80, 443 38m
openstack compute-novnc-proxy atmosphere vnc.131-153-200-197.nip.io 10.98.36.135 80, 443 38m
openstack container-infra atmosphere container-infra.131-153-200-197.nip.io 10.98.36.135 80, 443 27m
openstack container-infra-registry atmosphere container-infra-registry.131-153-200-197.nip.io 10.98.36.135 80, 443 27m
openstack dashboard atmosphere dashboard.131-153-200-197.nip.io 10.98.36.135 80, 443 23m
openstack identity atmosphere identity.131-153-200-197.nip.io 10.98.36.135 80, 443 51m
openstack image atmosphere image.131-153-200-197.nip.io 10.98.36.135 80, 443 46m
openstack key-manager atmosphere key-manager.131-153-200-197.nip.io 10.98.36.135 80, 443 49m
openstack load-balancer atmosphere load-balancer.131-153-200-197.nip.io 10.98.36.135 80, 443 30m
openstack network atmosphere network.131-153-200-197.nip.io 10.98.36.135 80, 443 36m
openstack orchestration atmosphere orchestration.131-153-200-197.nip.io 10.98.36.135 80, 443 33m
openstack placement atmosphere placement.131-153-200-197.nip.io 10.98.36.135 80, 443 41m
openstack rook-ceph-cluster atmosphere object-store.131-153-200-197.nip.io 10.98.36.135 80, 443 48m
openstack sharev2 atmosphere share.131-153-200-197.nip.io 10.98.36.135 80, 443 23m
openstack volumev3 atmosphere volume.131-153-200-197.nip.io 10.98.36.135 80, 443 42m

Le domaine WildCard utilisé ici me donne accès au Dashboard Horizon pour plus de simplicité :

Je charge une image locale d’Ubuntu 24.04 LTS dans le cluster :

root@instance:~# cat openrc 
# Ansible managed: Do NOT edit this file manually!

export OS_IDENTITY_API_VERSION=3

export OS_AUTH_URL="https://identity.131-153-200-197.nip.io/v3"
export OS_AUTH_TYPE=password
export OS_REGION_NAME="RegionOne"
export OS_USER_DOMAIN_NAME=Default
export OS_USERNAME="admin-RegionOne"
export OS_PASSWORD="lzi232PTaHpzoC2HjwSLKepZELQd6ENJ"
export OS_PROJECT_DOMAIN_NAME=Default
export OS_PROJECT_NAME=admin

export OS_CACERT=/usr/local/share/ca-certificates/atmosphere.crt

root@instance:~# wget -c https://cloud-images.ubuntu.com/noble/current/noble-server-cloudimg-amd64.img

Saving to: ‘noble-server-cloudimg-amd64.img’

noble-server-cloudimg-amd64.img 100%[=====================================================================================================>] 454.00M 17.9MB/s in 30s     

2024-06-03 13:22:40 (15.3 MB/s) - ‘noble-server-cloudimg-amd64.img’ saved [476053504/476053504]

root@instance:~# openstack image create --public --container-format bare --disk-format qcow2 --file ~/noble-server-cloudimg-amd64.img ubuntu-24.04
+------------------+------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Field | Value |
+------------------+------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| container_format | bare |
| created_at | 2024-06-03T13:23:12Z |
| disk_format | qcow2 |
| file | /v2/images/81d8eafa-4054-455c-9640-4e83c0566d21/file |
| id | 81d8eafa-4054-455c-9640-4e83c0566d21 |
| min_disk | 0 |
| min_ram | 0 |
| name | ubuntu-24.04 |
| owner | 43321f42e8434f8aa53531bd104e2809 |
| properties | locations='[]', os_hidden='False', owner_specified.openstack.md5='', owner_specified.openstack.object='images/ubuntu-24.04', owner_specified.openstack.sha256='' |
| protected | False |
| schema | /v2/schemas/image |
| status | queued |
| tags | |
| updated_at | 2024-06-03T13:23:12Z |
| visibility | public |
+------------------+------------------------------------------------------------------------------------------------------------------------------------------------------------------+

root@instance:~# openstack image list
+--------------------------------------+---------------------------+--------+
| ID | Name | Status |
+--------------------------------------+---------------------------+--------+
| 04e2f39a-e4ff-426b-aeee-b82acd3bf611 | amphora-x64-haproxy | active |
| 49a3b09d-c191-4c36-8541-31efaffb404d | cirros | active |
| b7e75cdb-37b0-4f2a-badf-cfdee7bca83d | manila-service-image | active |
| 3c5df46c-7015-411c-8009-afa6695672a6 | ubuntu-2204-kube-v1.27.8s | active |
| 81d8eafa-4054-455c-9640-4e83c0566d21 | ubuntu-24.04 | active |
+--------------------------------------+---------------------------+--------+

Configuration d’une topologie réseau pour un rapide test :

Avec l’image chargée dans Glance, lancement de trois petites instances Ubuntu 24.04 LTS …

root@instance:~# openstack server list
+--------------------------------------+-------+--------+--------------------------------------+--------------+----------+
| ID | Name | Status | Networks | Image | Flavor |
+--------------------------------------+-------+--------+--------------------------------------+--------------+----------+
| 07d6bf6e-edff-4729-9133-31849ea6fe87 | k0s-3 | ACTIVE | network1=10.96.250.216, 11.12.13.185 | ubuntu-24.04 | m1.small |
| afd10cda-f99c-4877-9948-a7f25b5e756a | k0s-1 | ACTIVE | network1=10.96.250.211, 11.12.13.99 | ubuntu-24.04 | m1.small |
| d70230a1-9fdf-40d9-959d-c9960479b4a5 | k0s-2 | ACTIVE | network1=10.96.250.208, 11.12.13.80 | ubuntu-24.04 | m1.small |
+--------------------------------------+-------+--------+--------------------------------------+--------------+----------+

Un cluster kubernetes simple sur la base de ces trois instances est alors possible avec k0sctl.

GitHub - k0sproject/k0sctl: A bootstrapping and management tool for k0s clusters.

root@instance:~# wget -c https://github.com/k0sproject/k0sctl/releases/download/v0.17.8/k0sctl-linux-x64
root@instance:~# mv k0sctl-linux-x64 /usr/local/bin/k0sctl && chmod +x /usr/local/bin/k0sctl 
root@instance:~# k0sctl
NAME:
   k0sctl - k0s cluster management tool

USAGE:
   k0sctl [global options] command [command options] 

COMMANDS:
   version Output k0sctl version
   apply Apply a k0sctl configuration
   kubeconfig Output the admin kubeconfig of the cluster
   init Create a configuration template
   reset Remove traces of k0s from all of the hosts
   backup Take backup of existing clusters state
   config Configuration related sub-commands
   completion  
   help, h Shows a list of commands or help for one command

GLOBAL OPTIONS:
   --debug, -d Enable debug logging (default: false) [$DEBUG]
   --trace Enable trace logging (default: false) [$TRACE]
   --no-redact Do not hide sensitive information in the output (default: false)
   --help, -h show help

root@instance:~# k0sctl init --k0s > k0sctl.yaml

root@instance:~# cat k0sctl.yaml 
apiVersion: k0sctl.k0sproject.io/v1beta1
kind: Cluster
metadata:
  name: k0s-cluster
spec:
  hosts:
  - ssh:
      address: 10.96.250.211
      user: ubuntu
      port: 22
      keyPath: /root/.ssh/id_rsa
    role: controller
  - ssh:
      address: 10.96.250.208
      user: ubuntu
      port: 22
      keyPath: /root/.ssh/id_rsa
    role: worker
  - ssh:
      address: 10.96.250.216
      user: ubuntu
      port: 22
      keyPath: /root/.ssh/id_rsa
    role: worker
  k0s:
    config:
      apiVersion: k0s.k0sproject.io/v1beta1
      kind: Cluster
      metadata:
        name: k0s
      spec:
        api:
          k0sApiPort: 9443
          port: 6443
        installConfig:
          users:
            etcdUser: etcd
            kineUser: kube-apiserver
            konnectivityUser: konnectivity-server
            kubeAPIserverUser: kube-apiserver
            kubeSchedulerUser: kube-scheduler
        konnectivity:
          adminPort: 8133
          agentPort: 8132
        network:
          kubeProxy:
            disabled: false
            mode: iptables
          kuberouter:
            autoMTU: true
            mtu: 0
            peerRouterASNs: ""
            peerRouterIPs: ""
          podCIDR: 10.244.0.0/16
          provider: kuberouter
          serviceCIDR: 10.96.0.0/12
        podSecurityPolicy:
          defaultPolicy: 00-k0s-privileged
        storage:
          type: etcd
        telemetry:
          enabled: true

root@instance:~# k0sctl apply --config k0sctl.yaml 

⠀⣿⣿⡇⠀⠀⢀⣴⣾⣿⠟⠁⢸⣿⣿⣿⣿⣿⣿⣿⡿⠛⠁⠀⢸⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⠀█████████ █████████ ███
⠀⣿⣿⡇⣠⣶⣿⡿⠋⠀⠀⠀⢸⣿⡇⠀⠀⠀⣠⠀⠀⢀⣠⡆⢸⣿⣿⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀███ ███ ███
⠀⣿⣿⣿⣿⣟⠋⠀⠀⠀⠀⠀⢸⣿⡇⠀⢰⣾⣿⠀⠀⣿⣿⡇⢸⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⠀███ ███ ███
⠀⣿⣿⡏⠻⣿⣷⣤⡀⠀⠀⠀⠸⠛⠁⠀⠸⠋⠁⠀⠀⣿⣿⡇⠈⠉⠉⠉⠉⠉⠉⠉⠉⢹⣿⣿⠀███ ███ ███
⠀⣿⣿⡇⠀⠀⠙⢿⣿⣦⣀⠀⠀⠀⣠⣶⣶⣶⣶⣶⣶⣿⣿⡇⢰⣶⣶⣶⣶⣶⣶⣶⣶⣾⣿⣿⠀█████████ ███ ██████████
k0sctl v0.17.8 Copyright 2023, k0sctl authors.
Anonymized telemetry of usage will be sent to the authors.
By continuing to use k0sctl you agree to these terms:
https://k0sproject.io/licenses/eula
INFO ==> Running phase: Set k0s version  
INFO Looking up latest stable k0s version         
INFO Using k0s version v1.30.1+k0s.0              
INFO ==> Running phase: Connect to hosts 
INFO [ssh] 10.96.250.216:22: connected            
INFO [ssh] 10.96.250.208:22: connected            
INFO [ssh] 10.96.250.211:22: connected            
INFO ==> Running phase: Detect host operating systems 
INFO [ssh] 10.96.250.216:22: is running Ubuntu 24.04 LTS 
INFO [ssh] 10.96.250.208:22: is running Ubuntu 24.04 LTS 
INFO [ssh] 10.96.250.211:22: is running Ubuntu 24.04 LTS 
INFO ==> Running phase: Acquire exclusive host lock 
INFO ==> Running phase: Prepare hosts    
INFO ==> Running phase: Gather host facts 
INFO [ssh] 10.96.250.216:22: using k0s-3 as hostname 
INFO [ssh] 10.96.250.211:22: using k0s-1 as hostname 
INFO [ssh] 10.96.250.208:22: using k0s-2 as hostname 
INFO [ssh] 10.96.250.211:22: discovered ens3 as private interface 
INFO [ssh] 10.96.250.208:22: discovered ens3 as private interface 
INFO [ssh] 10.96.250.216:22: discovered ens3 as private interface 
INFO [ssh] 10.96.250.211:22: discovered 11.12.13.99 as private address 
INFO [ssh] 10.96.250.208:22: discovered 11.12.13.80 as private address 
INFO [ssh] 10.96.250.216:22: discovered 11.12.13.185 as private address 
INFO ==> Running phase: Validate hosts   
INFO ==> Running phase: Validate facts   
INFO ==> Running phase: Download k0s on hosts 
INFO [ssh] 10.96.250.216:22: downloading k0s v1.30.1+k0s.0 
INFO [ssh] 10.96.250.211:22: downloading k0s v1.30.1+k0s.0 
INFO [ssh] 10.96.250.208:22: downloading k0s v1.30.1+k0s.0 
INFO ==> Running phase: Install k0s binaries on hosts 
INFO [ssh] 10.96.250.211:22: validating configuration 
INFO ==> Running phase: Configure k0s    
INFO [ssh] 10.96.250.211:22: installing new configuration 
INFO ==> Running phase: Initialize the k0s cluster 
INFO [ssh] 10.96.250.211:22: installing k0s controller 
INFO [ssh] 10.96.250.211:22: waiting for the k0s service to start 
INFO [ssh] 10.96.250.211:22: waiting for kubernetes api to respond 
INFO ==> Running phase: Install workers  
INFO [ssh] 10.96.250.216:22: validating api connection to https://11.12.13.99:6443 
INFO [ssh] 10.96.250.208:22: validating api connection to https://11.12.13.99:6443 
INFO [ssh] 10.96.250.211:22: generating a join token for worker 1 
INFO [ssh] 10.96.250.211:22: generating a join token for worker 2 
INFO [ssh] 10.96.250.216:22: writing join token   
INFO [ssh] 10.96.250.208:22: writing join token   
INFO [ssh] 10.96.250.216:22: installing k0s worker 
INFO [ssh] 10.96.250.208:22: installing k0s worker 
INFO [ssh] 10.96.250.216:22: starting service     
INFO [ssh] 10.96.250.216:22: waiting for node to become ready 
INFO [ssh] 10.96.250.208:22: starting service     
INFO [ssh] 10.96.250.208:22: waiting for node to become ready 
INFO ==> Running phase: Release exclusive host lock 
INFO ==> Running phase: Disconnect from hosts 
INFO ==> Finished in 42s                 
INFO k0s cluster version v1.30.1+k0s.0 is now installed 
INFO Tip: To access the cluster you can now fetch the admin kubeconfig using: 
INFO k0sctl kubeconfig

Kubernetes via k0s est actif et accessible via son fichier kubeconfigspécifique :

root@instance:~# k0sctl kubeconfig > kubeconfig
root@instance:~# kubectl --kubeconfig=kubeconfig cluster-info
Kubernetes control plane is running at https://10.96.250.211:6443
CoreDNS is running at https://10.96.250.211:6443/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy

To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.
root@instance:~# kubectl --kubeconfig=kubeconfig get nodes -o wide
NAME STATUS ROLES AGE VERSION INTERNAL-IP EXTERNAL-IP OS-IMAGE KERNEL-VERSION CONTAINER-RUNTIME
k0s-2 Ready <none> 2m41s v1.30.1+k0s 11.12.13.80 <none> Ubuntu 24.04 LTS 6.8.0-31-generic containerd://1.7.17
k0s-3 Ready <none> 2m41s v1.30.1+k0s 11.12.13.185 <none> Ubuntu 24.04 LTS 6.8.0-31-generic containerd://1.7.17

root@instance:~# kubectl --kubeconfig=kubeconfig get po,svc -A
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-system pod/coredns-6997b8f8bd-957bd 1/1 Running 0 2m41s
kube-system pod/coredns-6997b8f8bd-ddr5t 1/1 Running 0 2m41s
kube-system pod/konnectivity-agent-2bxgl 1/1 Running 0 2m51s
kube-system pod/konnectivity-agent-4gfsw 1/1 Running 0 2m51s
kube-system pod/kube-proxy-2cq5w 1/1 Running 0 2m51s
kube-system pod/kube-proxy-m6rnv 1/1 Running 0 2m51s
kube-system pod/kube-router-p9s4t 1/1 Running 0 2m51s
kube-system pod/kube-router-qhcp4 1/1 Running 0 2m51s
kube-system pod/metrics-server-5cd4986bbc-rf4wc 1/1 Running 0 2m57s

NAMESPACE NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
default service/kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 3m15s
kube-system service/kube-dns ClusterIP 10.96.0.10 <none> 53/UDP,53/TCP,9153/TCP 3m5s
kube-system service/metrics-server ClusterIP 10.97.91.41 <none> 443/TCP 3m1s

Le démonstrateur FC en une commande (habituelle maintenant) :

root@instance:~# cat test.yaml 
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: fcdemo3
  labels:
    app: fcdemo3
spec:
  replicas: 4
  selector:
    matchLabels:
      app: fcdemo3
  template:
    metadata:
      labels:
        app: fcdemo3
    spec:
      containers:
      - name: fcdemo3
        image: mcas/franceconnect-demo2:latest
        ports:
        - containerPort: 3000
---
apiVersion: v1
kind: Service
metadata:
  name: fcdemo-service
spec:
  type: ClusterIP
  selector:
    app: fcdemo3
  ports:
  - protocol: TCP
    port: 80
    targetPort: 3000

root@instance:~# kubectl --kubeconfig=kubeconfig apply -f test.yaml
deployment.apps/fcdemo3 created
service/fcdemo-service created

root@instance:~# kubectl --kubeconfig=kubeconfig get po,svc
NAME READY STATUS RESTARTS AGE
pod/fcdemo3-85f6bd87c-7jvpk 1/1 Running 0 13s
pod/fcdemo3-85f6bd87c-btv9m 1/1 Running 0 13s
pod/fcdemo3-85f6bd87c-nlw9x 1/1 Running 0 13s
pod/fcdemo3-85f6bd87c-v66bs 1/1 Running 0 13s

NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
service/fcdemo-service ClusterIP 10.98.123.24 <none> 80/TCP 13s
service/kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 8m19s

Redirection de port ici pour accès local à ce dernier …

root@instance:~# kubectl --kubeconfig=kubeconfig port-forward service/fcdemo-service 12222:80 --address='0.0.0.0'
Forwarding from 0.0.0.0:12222 -> 3000

Sachant qu’Atmosphere disposait de Magnum pour déployer un cluster Kubernetes intrinsèque …

Pour conclure, Atmosphere est un projet pour un déploiement simplifié d’un cluster Openstack conteneurisé dans un cluster Kubernetes de base (en reprenant OpenStack Helm) et que l’on retrouve également dans la nouvelle mouture de Canonical MicroStack par exemple …

OpenStack for the edge, micro clouds and developers

À suivre !

DEV Community

Une autre manière de faire fonctionner OpenStack au-dessus de Kubernetes avec Atmosphere …

Top comments (0)

Read next

Self-Host a Joplin Sync Server in Proxmox

Cool SHIT you can do with docker

Effortless Web Hosting: Build and Run an NGINX Server with Docker

Prometheus vs CloudWatch for Cloud Native Applications (Updated in 2024)