moved to monorepo: https://github.com/seankhliao/mono

kluster

Config files for my k8s cluster

secrets

need filter setup, see blog post

(Old) Clusters

Name	Nodes	Notes
cluster28	hetzner AX51	"medea"
cluster27	hetzner AX51	"medea"
cluster26		forever stuck in planning
cluster25	hetzner AX41	deployed with kind (kubeadm gives me headaches with pod connectivity)
cluster24	1x e2-standard-2	misnamed as cluster23
cluster23	3x e2-micro cosd / 0x e2-small preempt scaled 0-3	1.16/rapid nodes get stuck on ImagePull without error
cluster22	1x e2-standard-2	10GiB was not enough for kubelet/node
cluster21	1x e2-standard-2	less monitoring, back to deployments
cluster20	1x e2-standard-2	statefulsets for data
cluster19	1x e2-standard-2	kustomize is rad
cluster18	1x c-1-2G cosd
cluster17	1x c-1-2G cosd	kustomize is okay, i guess
cluster16	1x c-1-2G cosd preempt scaled 0-2 / 1x f1-micro cosd
cluster15		lost to the sands of time
cluster13	1x c-1-2G cosd preempt scaled 0-3 / 1x f1-micro cosd
cluster13	1x c-1-2G cosd preempt scaled 0-3 / 1x f1-micro cosd
cluster12	1x c-1-2G cosd preempt scaled 0-3 / 1x f1-micro cosd
cluster11	1x c-2-2G cosd preempt	node pools added later can't have service accounts set through gui
cluster10	1x c-1-2G cosd preempt scaled 0-3 x 2 zones	autoscaling is weird
cluster9	1x c-1-2G cosd preempt scaled 0-3 / 1x f1-micro cosd	micro pinned @ 100% cpu
cluster8	1x c-1-2G cosd preempt scaled 1-4	isto + stackdriver has high overhead -> choose more cpu
cluster7	1x c-1-2G cosd
cluster6	1x c-1-2G ubunut	mostly stable
cluster5	1x n1-standard-1 ubuntu	memory heavily underutilized
cluster4	4x f1-micro ubuntu	bin-packing nodes to death
cluster3	2x f1-micro ubuntu / 2x f1-micro cosd	alternating dead nodes
cluster2	1x g1-small ubuntu	100% cpu
cluster1	1x f1-micro ubuntu / 2x f1-micro cosd	alternating dead nodes
cluster0	1x g1-small ubuntu	never finished creating system pods