Quick Start Guide - Non-Production
This guide will provide step by step instructions on how to install Ondat onto your cluster, with the Ondat helm chart, for a non-production environment.
⚠️ This guide is for a non-production installation. Please follow the other installation guides for a production-ready installation of Ondat
Prerequisites
Ondat requires certain kernel modules to function. In particular it requires Linux-IO, an open-source implementation of the SCSI target, on all nodes that will execute Ondat (usually the workers). More information can be found here
This guide assumes you already have a Kubernetes cluster, with at least 3 worker nodes.
This guide works on the following Kubernetes distributions
- Vanilla Kubernetes
- Rancher
Step 1 - Install Ondat Helm Charts
Add the Ondat chart repository to Helm:
helm repo add ondat https://ondat.github.io/charts
helm repo update
Step 2 - Install Local Path Provisioner
Etcd requires a storage class before Ondat can be started. For non-production environments the Local Path Provisioner storage class can be used.
kubectl apply --filename="https://raw.githubusercontent.com/rancher/local-path-provisioner/v0.0.21/deploy/local-path-storage.yaml"
The following command can be used to ensure the Local Path Provisioner was successfully deployed.
> kubectl get pod,storageclass --namespace=local-path-storage
NAME READY STATUS RESTARTS AGE
pod/local-path-provisioner-c4d687f4c-bxmjt 1/1 Running 0 3h10m
NAME PROVISIONER RECLAIMPOLICY VOLUMEBINDINGMODE ALLOWVOLUMEEXPANSION AGE
storageclass.storage.k8s.io/local-path rancher.io/local-path Delete WaitForFirstConsumer false 3h10m
Step 3 - Customise and install the helm chart
A few changes need to be made to the default helm chart values, so it can run in smaller non-production sized clusters.
⚠️ Make sure to set the values of
ONDAT_USERNAMEandONDAT_PASSWORD
export ONDAT_USERNAME="changeme"
export ONDAT_PASSWORD="changeme"
helm install ondat ondat/ondat --create-namespace --namespace storageos \
--set ondat-operator.cluster.admin.username=${ONDAT_USERNAME},\
ondat-operator.cluster.admin.password=${ONDAT_PASSWORD},\
etcd-cluster-operator.cluster.replicas=3,\
etcd-cluster-operator.cluster.storageclass=local-path,\
etcd-cluster-operator.cluster.storage=6Gi,\
etcd-cluster-operator.cluster.resources.requests.cpu=100m,\
etcd-cluster-operator.cluster.resources.requests.memory=300Mi
Step 4 - Wait until the storageos storage class has been created
It can take a few seconds for the storageos storage class to be created. This needs to happen before it can be set as the default storage class.
The following command can be used to ensure the storageos storage class was successfully deployed.
> kubectl -n storageos get storageclass -w
NAME PROVISIONER RECLAIMPOLICY VOLUMEBINDINGMODE ALLOWVOLUMEEXPANSION AGE
local-path rancher.io/local-path Delete WaitForFirstConsumer false 3h4m
storageos csi.storageos.com Delete Immediate true 75m
Step 5 - Set the default storage class
Set the storageos storage class as the default.
kubectl patch storageclass local-path -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"false"}}}'
kubectl patch storageclass storageos -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"true"}}}'
Step 6 - Ensure everything is running
It can take a couple of minutes for the cluster to converge. A ready cluster will look like this:
> kubectl -n storageos get pods,storageclass,svc
NAME READY STATUS RESTARTS AGE
pod/ondat-controller-manager-5f7f669b4-fqj57 1/1 Running 0 73m
pod/ondat-controller-manager-5f7f669b4-jmbfk 1/1 Running 0 73m
pod/ondat-ondat-operator-675c4c9c88-9jgrv 2/2 Running 0 73m
pod/ondat-proxy-6887466cd5-vlkq6 1/1 Running 0 73m
pod/storageos-api-manager-6677c95579-qdght 1/1 Running 0 71m
pod/storageos-api-manager-6677c95579-rzwqr 1/1 Running 1 (71m ago) 71m
pod/storageos-csi-helper-6fbb8cc7d9-h767k 4/4 Running 0 71m
pod/storageos-node-fmtrd 3/3 Running 3 (72m ago) 72m
pod/storageos-node-h54xp 3/3 Running 3 (72m ago) 72m
pod/storageos-node-jm5th 3/3 Running 3 (72m ago) 72m
pod/storageos-scheduler-664886b7b-8jrv2 1/1 Running 0 72m
NAME PROVISIONER RECLAIMPOLICY VOLUMEBINDINGMODE ALLOWVOLUMEEXPANSION AGE
storageclass.storage.k8s.io/local-path rancher.io/local-path Delete WaitForFirstConsumer false 3h1m
storageclass.storage.k8s.io/storageos csi.storageos.com Delete Immediate true 72m
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
service/ondat-proxy ClusterIP 10.100.222.57 <none> 80/TCP 73m
service/storageos ClusterIP 10.102.114.114 <none> 5705/TCP 72m
service/storageos-api-manager-metrics ClusterIP 10.105.234.54 <none> 8080/TCP 71m
service/storageos-operator ClusterIP 10.109.66.235 <none> 8443/TCP 73m
service/storageos-operator-webhook ClusterIP 10.103.117.39 <none> 443/TCP 73m
service/storageos-webhook ClusterIP 10.108.126.126 <none> 443/TCP 71m
> kubectl -n storageos-etcd get pods,pdb
NAME READY STATUS RESTARTS AGE
pod/storageos-etcd-0-jbmq6 1/1 Running 0 73m
pod/storageos-etcd-1-nhbk6 1/1 Running 0 72m
pod/storageos-etcd-2-n9b9w 1/1 Running 0 72m
NAME MIN AVAILABLE MAX UNAVAILABLE ALLOWED DISRUPTIONS AGE
poddisruptionbudget.policy/storageos-etcd 2 N/A 1 73m
Applying a Licence to the Cluster
⚠️ Newly installed Ondat clusters must be licensed within 24 hours. Our Community Edition tier supports up to 1 TiB of provisioned storage.
To obtain a licence, follow the instructions on the licensing operations page.