Preparing to install an Openshift environment
You must complete the tasks in this topic if you are installing in an Openshift environment before you can install Sametime.
Refer to Planning for Openshift for consideratons related to namespace and video.
-
Deploy in a namespace. You can either create a namespace or use the default namespace. If you deploy Sametime into the default namespace, there are modifications that must be appled to the Helm chart.
Deploy to a new namespace
export NAMESPACE=sametime oc create namespace $NAMESPACE
Deploy to default namespace
-
Edit the values.yaml file.
Comment out the
seLinuxOptions: false
setting for each of the following sections.activity
files
recordings
For example:
activity: fullnameOverride: activity persistence: {} # seLinuxOptions: false
-
-
Deploy the video using one of the following methods.
-
Host Port
This is the default which provides the best performance and scales automatically scalable. This method requires pod-to-node affinity restriction through node labels. A separate video service account is required and the hostnetwork-v2 must be assigned to it.
oc create -f kubernetes/stack/openshift/sametime-hostnetwork-v2.yaml kubectl -n $NAMESPACE create serviceaccount videoUser oc adm policy add-scc-to-user sametime-hostnetwork-v2 -z videoUser -n $NAMESPACE
Edit the values.yaml to reference this video service account:
video: serviceAccount: name: videoUser
-
Using a load balancer is lower performance and has no pod-to-node restrictions. It requires the Kubernetes load balancing infrastructure.
Edit the values.yaml file to enable the loadBalanceVideo setting and reference the Sametime service account.
global: loadBalanceVideo: true ... video: serviceAccount: name: videoUser
-
Using a node port is also lower performance but is restricted to a single node. It requires a no host-network SCC.
Edit the values.yaml file to add the following to the Sametime service account:
global: disableHostNetwork: true ... video: serviceAccount: name: videoUser
-
Continue with the topics for installing in a Kubernetes environment.