NetApp Block Storage Back End Guide
A guide to using a NetApp appliance as a Block Storage back end in Red Hat OpenStack Platform 13
Making open source more inclusive
Red Hat is committed to replacing problematic language in our code, documentation, and web properties. We are beginning with these four terms: master, slave, blacklist, and whitelist. Because of the enormity of this endeavor, these changes will be implemented gradually over several upcoming releases. For more details, see our CTO Chris Wright’s message.
Chapter 1. Introduction
This document describes how to use the Director to deploy a NetApp appliance as a back end to the Overcloud’s Block Storage service. The following sections assume that:
- You intend to use only NetApp appliance and drivers for Block Storage back ends
- The OpenStack Overcloud has already been deployed through Director
- The NetApp appliance has already been configured and is ready to be used as a storage repository
- You have the necessary credentials for connecting to the NetApp storage system or proxy server
You have the username and password of an account with elevated privileges. You can use the same account that was created to deploy the Overcloud; in Creating the stack user, a
stackuser is created for this purpose.
When Red Hat OpenStack Platform is deployed through the Director, all major Overcloud settings (in particular, the Block Storage service back end) must be defined and orchestrated through the Director as well. This ensures that the settings will persist through any further Overcloud updates.
For manual instructions on configuring the Block Storage service to use a NetApp appliance as a back end, see Chapter 4. OpenStack Block Storage Service (from the NetApp OpenStack Deployment and Operations Guide). Manually-configured Block Storage settings will need to be re-applied during updates to the Overcloud, as the Director will overwrite any settings it did not orchestrate.
This document explains how to orchestrate your desired NetApp back end configuration to the Overcloud’s Block Storage service. This document will not discuss the different deployment configurations possible with the NetApp back end. Rather, to learn more about the different available NetApp deployment choices, see Theory of Operation & Deployment Choices (from the NetApp OpenStack Deployment and Operations Guide).
Once you are familiar with the resulting back end configuration you want to deploy (and its corresponding settings), refer to this document for instructions on how to orchestrate it through the Director.
At present, the Director only has the integrated components to deploy a single instance of a NetApp back end. As such, this document only describes the deployment of a single back end.
Deploying multiple instances of a NetApp back end requires a custom back end configuration. See the Custom Block Storage Back End Deployment Guide for instructions.
Chapter 2. Process Description
Red Hat OpenStack Platform includes all the drivers required for all NetApp appliances supported by the Block Storage service. In addition, the Director also has the puppet manifests, environment files, and Orchestration templates necessary for integrating the NetApp appliance as a back end to the Overcloud.
Configuring the NetApp appliance as a back end involves editing the default environment file and including it in the Overcloud deployment. This file is available locally on the Undercloud, and can be edited to suit your environment.
After editing this file, invoke it through the Director. Doing so ensures that it will persist through future Overcloud updates. The following sections describe this process in greater detail.
Chapter 3. Define the Back End
This section describes the deployment of a single back end. Deploying multiple instances of a NetApp back end requires a custom back end configuration. See the Custom Block Storage Back End Deployment Guide for instructions.
With a director deployment, the easiest way to define the NetApp appliance as a Block Storage back end is through the integrated NetApp environment file. This file is located in the following path of the undercloud node:
Copy this file to a local path where you can edit and invoke it later. For example, to copy it to
$ cp /usr/share/openstack-tripleo-heat-templates/environments/storage/cinder-netapp-config.yaml ~/templates/
Afterwards, open the copy (
~/templates/cinder-netapp-config.yaml) and edit it as you see fit. The following snippet displays the default contents of this file:
# A heat environment file which can be used to enable a # a Cinder NetApp backend, configured via puppet resource_registry: OS::TripleO::Services::CinderBackendNetApp: ../puppet/services/cinder-backend-netapp.yaml 1 parameter_defaults: 2 CinderEnableNetappBackend: true 3 CinderNetappBackendName: 'tripleo_netapp' CinderNetappLogin: '' CinderNetappPassword: '' CinderNetappServerHostname: '' CinderNetappServerPort: '80' CinderNetappSizeMultiplier: '1.2' CinderNetappStorageFamily: 'ontap_cluster' CinderNetappStorageProtocol: 'nfs' CinderNetappTransportType: 'http' CinderNetappVfiler: '' CinderNetappVolumeList: '' CinderNetappVserver: '' CinderNetappPartnerBackendName: '' CinderNetappNfsShares: '' CinderNetappNfsSharesConfig: '/etc/cinder/shares.conf' CinderNetappNfsMountOptions: '' CinderNetappCopyOffloadToolPath: '' CinderNetappControllerIps: '' CinderNetappSaPassword: '' CinderNetappStoragePools: '' CinderNetappEseriesHostType: 'linux_dm_mp' CinderNetappWebservicePath: '/devmgr/v2'
There are several director heat parameters that control whether an NFS back end or a NetApp NFS Block Storage back end supports a NetApp feature called NAS secure:
Red Hat does not recommend that you enable the feature, because it interferes with normal volume operations. Director disables the feature by default, and Red Hat OpenStack Platform does not support it.
- The OS::TripleO::Services::CinderBackendNetApp parameter in the
resource_registrysection refers to a composable service template named
cinder-backend-netapp.yaml. This is the template that the Director should use to load the necessary resources for configuring the back end. By default, the parameter specifies the path to
cinder-backend-netapp.yamlrelatively. As such, update this parameter with the absolute path to the file:
resource_registry: OS::TripleO::Services::CinderBackendNetApp: /usr/share/openstack-tripleo-heat-templates/puppet/services/cinder-backend-netapp.yaml
- The parameter_defaults section contains your back end definition. Specifically, it contains the parameters that director passes to the resources defined in
- The CinderEnableNetappBackend: true line instructs director to use the puppet manifests necessary for the default configuration of a NetApp back end. This includes defining the volume driver that the Block Storage service should use (specifically,
To define your NetApp back end, edit the settings in the parameter_defaults section as you see fit. The following table explains each parameter and lists its corresponding
For more about variables, see the corresponding reference in NetApp OpenStack Docs for your NetApp appliance.
Table 3.1. NetApp universal back end settings
(Required) An arbitrary name to identify the volume back end. The
(Required) Administrative account name used to access the back end or its proxy server. For this parameter, you can use an account with cluster-level administrative permissions (namely,
(Required) The corresponding password of CinderNetappLogin.
(Required) The storage system or proxy server (for E-Series). The value of this option should be the IP address or hostname of either the cluster management logical interface (LIF) or Storage Virtual Machine (SVM) LIF.
(Optional) The TCP port that the Block Storage service should use to communicate with the NetApp back end. If not specified, Data ONTAP drivers will use
(Deprecated) During volume creation, the quantity to be multiplied to the requested volume size to ensure that the NetApp back end has enough space.
(Optional) The storage family type used on the back end device. Use
NOTE: Support for
(Required) The storage protocol to be used. Use either
(Required) Transport protocol to be used for communicating with the back end. Valid options include
The following setting is only valid for clustered Data ONTAP (as in, with CinderNetappStorageFamily set to
Table 3.2. NetApp settings for clustered DATA ONTAP
(Required) Specifies which the name of the SVM where volume provisioning should occur. This refers to a single SVM on the storage cluster.
The following settings are only valid with the E-Series family of devices (as in, with CinderNetappStorageFamily set to
Table 3.3. NetApp settings for E-Series
(Required) A comma-separated list of controller management IPs/hostnames to which provisioning should be restricted.
(Optional) Password to the NetApp E-Series storage array.
(Removed) A comma-separated list of disk pools to which provisioning should be restricted.
Do not edit this parameter, as it now refers to an unavailable driver option.
(Removed) Defines the type of operating system for all initiators that can access a LUN. This information is used when mapping LUNs to individual hosts or groups of hosts.
Do not edit this parameter, as it now refers to an unavailable driver option.
(Optional) Specifies the path to the E-Series proxy application on a proxy server. To determine the full URL for connecting to the proxy application, the driver combines the CinderNetappTransportType, CinderNetappServerHostname, and CinderNetappServerPort port values.
The following settings are only valid when using the NFS protocol (as in, with CinderNetappStorageProtocol set to
nfs). For more information, see NetApp Unified Driver for Clustered Data ONTAP with NFS or NetApp Unified Driver for Data ONTAP operating in 7-Mode with NFS (both from the NetApp OpenStack Deployment and Operations Guide).
Table 3.4. NetApp settings for NFS
(Required) Comma-separated list of Data LIFs exported from the NetApp ONTAP device to be mounted by the Controller nodes. This list gets written to the location defined by CinderNetappNfsSharesConfig. For example:
(Required) Absolute path to the NFS exports file. This file contains a list of available NFS shares to be used as a back end.
(Optional) Comma-separated list of mount options you want to pass to the NFS client. For more information about valid options, see man mount.
(Optional) Specifies the path of the NetApp copy offload tool binary. This binary (available from the NetApp Support portal) must have the
[a] This option is only valid with Clustered Data ONTAP (as in, with CinderNetappStorageFamily set to
Chapter 4. Deploy the Configured Back End
The Director installation uses a non-root user to execute commands, which includes orchestrating the deployment of the Block Storage back end. In Creating a Director Installation User, a user named
stack is created for this purpose. This user is configured with elevated privileges.
Log in as the
stack user to the Undercloud. Then, deploy the NetApp back end (defined in the edited
~/templates/cinder-netapp-config.yaml) by running the following:
$ openstack overcloud deploy --templates -e ~/templates/cinder-netapp-config.yaml
If you passed any extra environment files when you created the overcloud, pass them again here by using the
-e option to avoid making undesired changes to the overcloud. For more information, see Modifying the Overcloud Environment in the Director Installation and Usage guide.
Test the back end after director orchestration is complete. See Chapter 5, Test the Configured Back End.
Chapter 5. Test the Configured Back End
After deploying the back end, test whether you can successfully create volumes on it. Doing so will require loading the necessary environment variables first. These variables are defined in /home/stack/overcloudrc by default.
To load these variables, run the following command as the stack user:
$ source /home/stack/overcloudrc
For more information, see Accessing the Overcloud.
You should now be logged in to the Controller node. From there, you can create a volume type, which can be used to specify the back end you want to use (in this case, the newly-defined back end in Chapter 3, Define the Back End). This is required in an OpenStack deployment where you have other back ends enabled (preferably, also through Director).
To create a volume type named netapp, run:
$ cinder type-create netapp
Next, map this volume type to the back end defined in ]. Given the back end name
tripleo_netapp (as defined through the
CinderNetappBackendName parameter, in xref:edityaml[), run:
$ cinder type-key netapp set volume_backend_name=tripleo_netapp
You should now be able to create a 2GB volume on the newly defined back end by invoking its volume type. To do so, run:
$ cinder create --volume-type netapp 2