Skip to content

Latest commit

 

History

History
68 lines (47 loc) · 2.1 KB

RunCommands.md

File metadata and controls

68 lines (47 loc) · 2.1 KB
Raw

Log of Run Commands to get to cluster state

Fix Two Default Storage Classes (I no longer install local-path)

kubectl patch storageclass local-path -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"false"}}}'

Manually run cronjob

kubectl create job --from=cronjob/<name of cronjob> <name of job>

Fix invalid cert on logs

kubectl get csr --sort-by=.metadata.creationTimestamp

kubectl certificate approve

Delete and readd an OSD

https://gist.github.com/cheethoe/49d9c1d0003e44423e54a060e0b3fbf1

Reformat with sidero

kubie ctx

flux suspend kustomization sidero-configs

kubectl edit cluster cluster-0

spec.paused: true

delete relevant machine

Edit to delete finalization

Delete relevant serverbinding

Restart node to wipe

How to add a node back to cluster

Use the normal k3s-install, but specify k3s_control_token: in kubernetes/k3s.yml

How to replace and readd an OSD back to the cluster

  1. Comment out the node from the code and reconcile that change
  2. Purge the OSD with the web UI or ceph osd purge osd.ID --yes-i-really-mean-it
  3. Delete the job and deployment of the node
  4. Using the toolbox, run ceph auth del osd.ID or you will get an error about the key already existing
  5. Uncomment the node from code, changing as needed, and reconcile that change to re-add the node.

Setup RPI 4 to boot from USB

Guide - Firmware update not needed

Get creds

To login to minio-operator with JWT
kubectl -n minio get secret (kubectl -n minio get serviceaccount console-sa -o jsonpath="{.secrets[0].name}") -o jsonpath="{.data.token}" | base64 --decode

To login to rook-ceph
kubectl -n rook-ceph get secret rook-ceph-dashboard-password -o jsonpath="{['data']['password']}" | base64 --decode && echo

Talos

I had to adjust watchers once. Now in cluster configs.

talosctl patch mc -p '[{ "op": "add", "path": "/machine/sysctls", "value": { "fs.inotify.max_user_watches": 524288, "fs.inotify.max_user_instances": 512 } }]' --immediate