Closed turnrye closed 3 years ago
Here’s the cluster.yml
that I’m starting with:
# If you intened to deploy Kubernetes in an air-gapped environment,
# please consult the documentation on how to configure custom RKE images.
nodes:
- address: k8s-1.leb.memhamwan.net
role:
- controlplane
- etcd
user: turnrye
- address: k8s-2.leb.memhamwan.net
role:
- worker
user: turnrye
- address: k8s-3.leb.memhamwan.net
role:
- worker
user: turnrye
system_images:
etcd: rancher/coreos-etcd:v3.4.13-rancher1
alpine: rancher/rke-tools:v0.1.65
nginx_proxy: rancher/rke-tools:v0.1.65
cert_downloader: rancher/rke-tools:v0.1.65
kubernetes_services_sidecar: rancher/rke-tools:v0.1.65
kubedns: rancher/k8s-dns-kube-dns:1.15.10
dnsmasq: rancher/k8s-dns-dnsmasq-nanny:1.15.10
kubedns_sidecar: rancher/k8s-dns-sidecar:1.15.10
kubedns_autoscaler: rancher/cluster-proportional-autoscaler:1.8.1
coredns: rancher/coredns-coredns:1.7.0
coredns_autoscaler: rancher/cluster-proportional-autoscaler:1.8.1
nodelocal: rancher/k8s-dns-node-cache:1.15.13
kubernetes: rancher/hyperkube:v1.19.3-rancher1
flannel: rancher/coreos-flannel:v0.13.0-rancher1
flannel_cni: rancher/flannel-cni:v0.3.0-rancher6
calico_node: rancher/calico-node:v3.16.1
calico_cni: rancher/calico-cni:v3.16.1
calico_controllers: rancher/calico-kube-controllers:v3.16.1
calico_ctl: rancher/calico-ctl:v3.16.1
calico_flexvol: rancher/calico-pod2daemon-flexvol:v3.16.1
canal_node: rancher/calico-node:v3.16.1
canal_cni: rancher/calico-cni:v3.16.1
canal_controllers: rancher/calico-kube-controllers:v3.16.1
canal_flannel: rancher/coreos-flannel:v0.13.0-rancher1
canal_flexvol: rancher/calico-pod2daemon-flexvol:v3.16.1
weave_node: weaveworks/weave-kube:2.7.0
weave_cni: weaveworks/weave-npc:2.7.0
pod_infra_container: rancher/pause:3.2
ingress: rancher/nginx-ingress-controller:nginx-0.35.0-rancher1
ingress_backend: rancher/nginx-ingress-controller-defaultbackend:1.5-rancher1
metrics_server: rancher/metrics-server:v0.3.6
windows_pod_infra_container: rancher/kubelet-pause:v0.1.4
ssh_agent_auth: true
Attached are the kube_config_cluster.yml
and cluster.rkestate
, first encrypted with openssl enc -aes-256-cbc -salt
and then zipped to make github happy.
kube_cluster_rke_files.zip
Running through https://rancher.com/docs/rancher/v2.x/en/installation/install-rancher-on-k8s/ now.
I setup a wildcard cname *.k8s.leb.memhamwan.net to use for hostnames. Here's what I configured for now for rancher:
helm install rancher rancher-latest/rancher \
--namespace cattle-system \
--set hostname=rancher.k8s.leb.memhamwan.net \
--set ingress.tls.source=letsEncrypt \
--set letsEncrypt.email=netops@memhamwan.org
Running through https://metallb.universe.tf/installation/#installation-by-manifest now.
metallb config:
apiVersion: v1
kind: ConfigMap
metadata:
namespace: metallb-system
name: config
data:
config: |
address-pools:
- name: default
protocol: layer2
addresses:
- 44.34.128.184-44.34.128.190
I created lb1.k8s.leb - lb7.k8s.leb entries for these in DNS.
https://rancher.k8s.leb.memhamwan.net/ is now up with user "admin" and the standard hamwan pwd.
Let’s provision a 3 node k8s cluster on our esxi host at LEB.