Back in DTW, we have made a commitment to invest in the kubernetes eco-system (which is getting bigger and bigger..)

Every journey starts with small steps and if you are new to the who ‘storage persistency’ issues with Docker, I would highly advice that you first read the post I wrote back in november 2018

https://volumes.blog/2018/12/13/tech-previewing-our-upcoming-xtremio-integration-with-kubernetes-csi-plugin/ read it? Good! so after we have released the VxFlex OS and the XtremIO CSI plugins, it is now the time for PowerMax to get the support as well.

If you think about it, PowerMax is the only array that support both Mainframes and (now), containers. Think of all the mission critical apps it support and now, all the new Devops based apps as well, it can really do it all!

About the support for the PowerMax / CSI initial release:

The CSI Driver adheres to the Container Storage Interface (CSI) specification v1.0 and is compatible with Kubernetes versions 1.13.1, 1.13.2, and 1.13.2 running within a host operating system of Red Hat Enterprise Linux (RHEL) 7.6.

Services Design, Technology, & Supportability Improvements

The CSI Driver for Dell EMC PowerMax v1.0 has the following features

Supports CSI 1.0
Supports Kubernetes version 1.13.1, 1.13.2, and 1.13.3
Supports Red Hat Enterprise Linux 7.6 host operating system
Supports PowerMax – 5978.221.221 (ELM SR)
Persistent Volume (PV) capabilities:
- create
- delete
Dynamic and Static PV provisioning
Volume mount as ext4 or xfs file system on the worker node
Volume prefix for easier LUN identification in Unisphere
HELM charts installer
Access modes:
- SINGLE_NODE_WRITER
- SINGLE_NODE_READER_ONLY

Installation

Driver Installation Prerequisites

Full details are in the Installation Guide but here is a summary of pre-reqs for the driver.

Upstream Kubernetes 1.13.x, with specific feature gates enabled
Downstream distributions, like OpenShift and PKS, have not been tested and may not work at this time.
Docker daemon running and configured with MountFlags=shared on all k8s masters/nodes
Helm and Tiller installed on k8s masters
iscsi-initiator-utils package installed on all the Kubernetes nodes
Make sure that the ISCSI IQNs (initiators) from the Kubernetes nodes are not part of any existing Hosts on the array(s)
A namespace “powermax” should be created prior to the installation. The driver pods, secrets are created in this namespace.
Make sure that all the nodes in the Kubernetes cluster have network connectivity with the U4P instance (used by the driver)

Full details are in the Installation Guide.

Clone the repository from the URL – github.com/dell/csi-powermax
Create a Kubernetes secret – “powermax-creds” – with your U4P’s username and password in the namespace “powermax“
(Optional) Create a Kubernetes secret – “powermax-certs” – with CA cert(s) used to sign U4P’s SSL certificate. If not found, the install script will create an empty secret in the namespace “powermax“
Create your myvalues.yaml file from the values.yaml file and edit some parameters for your installation.These include –
1. The U4P URL (must include the port number as well)
2. clusterPrefix – A unique identifier identifying the Kubernetes cluster (max 3 characters)
3. A list of port group names
4. An optional (white)list of arrays the driver will manage
5. A set of values for the default storage class (SYMID, SRP, ServiceLevel)
Run the “install.powermax” shell script
Customize the installation (if desired) by adding additional Storage Classes to support multiple arrays, storage resource pools, service levels

Using the Driver

See the Installation Document section “Test Examples that use the CSI Driver For Dell EMC PowerMax”

• There are a number of test helm charts and scripts in test/helm that can be used as examples of how to deploy storage using the driver

Create a “test” namespace to hold the tests.
The directory 2vols contains a simple Helm chart that deploys a container with two volumes. To deploy it run “sh starttest.sh 2vols”. To stop it run “sh stoptest.sh”.
There is a sample master/slave Postgres deployment example.

▪ Deploys a master/slave postgres database ▪ You can deploy it using “sh postgres.sh”; however for the script to work correctly you need psql locally installed where you’re running the script from, and the world database from http://pgfoundry.org/projects/dbsamples/
installed at /root/dbsamples-0.1.

▪ If psql and the sample database are installed, the script will install the sample database into the deployed postgres database and run a sample query against it.