This Gist explains how to deploy the Habana Labs operator in an OpenShift cluster. The first step is to deploy an OpenShift 4.11 cluster. Simply follow the documentation.
The habanalabs module needs to load a firmware, and the driver container will copy it to /var/lib/firmware on the node. We need to tell the node kernel to lookup that folder for firmwares, as it's not a default path. This is done by applying a MachineConfig to all worker nodes. The following command line will do it.
The RHPDS catalog item you provisioned is a single-node OpenShift environment that is backed by an Amazon P-type EC2 instance which has 1 NVIDIA GPU. It is a 100% vanilla/standard OpenShift Container Platform 3.10 installation. Post-install, there were a few additional things done consuming Ansible content from the https://github.com/redhat-performance/openshift-psap
| #!/bin/bash | |
| # This script allocates all primary shards that are unassigned to a a given node using | |
| # the openshift binary. The binary must be in the path and the user executing the script | |
| # must have access to logging project. The inputs are: | |
| # pod An Elasticsearch pod name | |
| # node An node Elasticsearch cluster which should be any one of the DC's. We could | |
| # probably infer this from the pod name | |
| pod=$1 | |
| node=$2 |
This gist is mostly based on the dnsmasq appendix from the openshift-training repo. However, I updated it and included fixes for the many gotchas I found along the way.
This is useful for folks who want to set up a DNS service as part of the cluster itself, either because they cannot easily change their DNS setup outside of the cluster, or just because they want to keep the cluster setup self-contained.