The recommended way to run StorageOS on an OpenShift 3.10 cluster is to deploy a daemonset with RBAC support.
- Ensure any firewalls permit the appropriate ports
- If your cluster enables SELinux, add the following permissions for each of
the nodes that run StorageOS.
setsebool -P virt_sandbox_use_fusefs on setsebool -P virt_use_fusefs on
-Poption makes the change persistent after reboots.
- Ensure that your docker installation has mount propagation enabled per our mount propagation prerequisites
- Enable the
MountPropagationflag by appending feature gates to the api and controller (you can apply these changes using the Ansible Playbooks)
Note: If you are using atomic installation rather than origin, the location of the yaml config files and service names might change.
Add to the KubernetesMasterConfig section (/etc/origin/master/master-config.yaml):
kubernetesMasterConfig: apiServerArguments: feature-gates: - MountPropagation=true controllerArguments: feature-gates: - MountPropagation=true
Add to the feature-gates to the kubelet arguments (/etc/origin/node/node-config.yaml):
kubeletArguments: feature-gates: - MountPropagation=true
Warning: Restarting OpenShift services can cause downtime in the cluster.
- Restart services in the MasterNode/s
master-restart api master-restart controllers # Restart kubelet systemctl restart atomic-openshift-node.service
- Restart service in all Nodes
# Restart kubelet systemctl restart atomic-openshift-node.service
Install StorageOS operator
Our cluster operator is a Kubernetes native application developed to deploy and configure StorageOS clusters, and assist with maintenance operations. We recommend its use for standard installations.
The operator is a Kubernetes controller that watches the
CRD. Once the controller is ready, a StorageOS cluster definition can be
created. The operator will deploy a StorageOS cluster based on the
configuration specified in the cluster definition.
The StorageOS Cluster Operator can be installed with two options.
- Using Helm
- Standard yaml manifests
(Option 1) Using Helm
helm repo add storageos https://charts.storageos.com helm install storageos/storageoscluster-operator --namespace storageos-operator
The Helm chart can be found in the Charts public repository.
The StorageOS Cluster Operator source code can be found in the cluster-operator repository.
The helm server, tiller, needs privileges to be able to deploy the StorageOS Cluster Operator. You can add the service account to the cluster-admin role for simplicity or create a role that matches the cluster-operator requirements.
(Option 2) Standard yaml manifests
git clone https://github.com/storageos/deploy.git storageos cd storageos/openshift/deploy-storageos/cluster-operator ./deploy-operator.sh
Verify the Cluster Operator Pod
[[email protected]]# oc -n storageos-operator get pod NAME READY STATUS RESTARTS AGE storageoscluster-operator-68678798ff-f28zw 1/1 Running 0 3m
The READY 1/1 indicates that
stosresources can be created.
Create a Secret
Before deploying a StorageOS cluster, create a Secret to define the StorageOS API Username and Password in base64 encoding.
oc create -f - <<END apiVersion: v1 kind: Secret metadata: name: "storageos-api" namespace: "default" labels: app: "storageos" type: "kubernetes.io/storageos" data: # echo -n '<secret>' | base64 apiUsername: c3RvcmFnZW9z apiPassword: c3RvcmFnZW9z END
This example contains a default password, for production installations, use a unique, strong password.
Make sure that the encoding of the credentials doesn’t have special characters such as ‘\n’.
You can define a base64 value by
echo -n "mystring" | base64.
Add scc (security context constraint) for StorageOS
oc adm policy add-scc-to-user privileged system:serviceaccount:storageos:storageos-daemonset-sa
Trigger a StorageOS installation
This is a Cluster Definition example.
oc create -f - <<END apiVersion: "storageos.com/v1alpha1" kind: "StorageOSCluster" metadata: name: "example-storageos" spec: secretRefName: "storageos-api" # Reference the Secret created in the previous step secretRefNamespace: "default" # Namespace of the Secret images: nodeContainer: "storageos/node:1.0.2" # StorageOS version resources: requests: memory: "512Mi" nodeSelectorTerms: - matchExpressions: - key: "node-role.kubernetes.io/compute" operator: In values: - "true" END
specparameters available on the Cluster Operator configuration page.
You can find more examples such as deployments with CSI or deployments referencing a external etcd kv store. store for StorageOS in the Cluster Operator examples page.
If this is your first installation you may wish to follow the StorageOS Volume guide for an example of how to mount a StorageOS volume in a Pod.
There are a variety of flavours, versions and particularities in the container orchestrator scope. Because of this, StorageOS installation procedures aim to be flexible so they can fit different needs depending on the environment, preferences or requirements. The StorageOS cluster operator simplifies the installation by implementing automated install. You can review and adapt the StorageOS install in the following examples. Feel free to extend and modify the publicly available examples.
Installation with Native Drivers (default)
The following github repository hosts installation examples.
git clone https://github.com/storageos/deploy.git storageos cd storageos/openshift/deploy-storageos
You can see various installation examples are available such as
CSI (Container Storage Interface) or
etcd-as-svc. All of them have a
deploy-storageos.sh that serves as a wrapper to trigger the manifest
creation. Follow the according
README.md for each one of them for more
For advanced installations, it is recommended to refer to
will guide the user to deploy an etcd cluster deployed by the official
Kubernetes etcd operator. Once an
external etcd cluster has been created, then use the StorageOS cluster operator
to finish the installation.