We are currently recommending that you deploy ceph using kolla-ansible or ceph-deploy.
This list of instructions is currently incomplete.
completely remove the following option from /etc/kolla-kubernetes/kolla-kubernetes.yml
keyring: /etc/ceph/ceph.client.admin.keyring
set the user option in the storage_ceph to ‘kolla’ in /etc/kolla-kubernetes/kolla-kubernetes.yml and pool = kollavolumes
Upload ceph.conf and admin key generated from the external ceph while orchestration_engine=ANSIBLE
kubectl create configmap ceph-conf --namespace=kolla \
--from-file=ceph.conf=/etc/kolla/ceph-osd/ceph.conf
kubectl create secret generic ceph-client-admin-keyring --namespace=kolla\
--from-file=data=/etc/kolla/ceph-osd/ceph.client.admin.keyring
Before any pv’s are created, do the following
kollakube res create pod ceph-rbd
kollakube res create pod ceph-admin
watch kubectl get pods --namespace=kolla
Wait for ceph-admin to come up.
Create a pool a user:
Create disks for ‘rabbitmq’ and ‘mariadb’ like so
cmd="rbd create --pool kollavolumes --image-feature layering --size 10240"
cmd="$cmd mariadb; rbd map --pool kollavolumes mariadb; #format it and unmount/unmap..."
kubectl exec -it ceph-admin -- /bin/bash -xec "$cmd"
It is very half baked, intended only for testing. Please don’t store anything you care about in it as we will guarantee it will loose your data.
Except where otherwise noted, this document is licensed under Creative Commons Attribution 3.0 License. See all OpenStack Legal Documents.