Objectives

After completing this section, you should be able to describe how to install Red Hat Virtualization Manager (RHV-M) and Red Hat Gluster Storage as a self-hosted, hyperconverged deployment on Red Hat Virtualization Hosts.

Introducing Red Hat Hyperconverged Infrastructure for Virtualization (RHHI-V)

Hyperconverged infrastructure has become increasingly popular for a broad range of use cases. Hyperconverged combines both compute and storage resources, and makes them simultaneously available among all hosts in a single, scalable, virtualization installation known as a pod. Storage is pooled across all hosts in a pod, eliminating the need for a storage area network, and is managed by the same Red Hat Virtualization management software used for standard RHV deployments.

Figure 14.1: The RHHI-V architecture of a single pod

To scale, add more hosts to increase both the compute capacity and the storage space. Because RHHI-V is a software solution instead of a hardware appliance, resource scaling can be configured to match application requirements. Increase only disk space by expanding the storage pools on each host, without adding unnecessary compute capacity. Add more hosts to increase compute capacity, but include only the storage required to implement pool redundancy per cluster.

RHHI-V is a tuned combination of RHEL, Red Hat Virtualization, and Red Hat Gluster Storage, with an Open Virtual Network (OVN) software-defined networking stack, and Red Hat Ansible Automation for provisioning. With OVN, RHHI-V integrates with Red Hat OpenStack Platform and Red Hat OpenShift Container Platform infrastructures for a single hybrid cloud platform. RHHI-V also includes the Virtual Data Optimizer (VDO), which includes these storage enhancements:

RHHI-V is offered as a solution for remote office, back office, and edge computing. However, it is also enjoying rapid acceptance in all data center sizes, as an easily managed and scaled mission-critical workload platform, primarily because RHHI-V deployments can be right-sized.

Installing a RHHI-V Pod

Install Red Hat Virtualization Hosts

The RHV-H images are the same as for a standard installation of RHV 4.3. Download the Hypervisor Image for RHV 4.3 ISO image from the customer portal, and prepare bootable media using the image. Boot each physical system, and then perform a normal installation. Default values are acceptable, but adjust parameters as required for your hardware. Red Hat recommends these configuration choices:

RHV provides a script for performing health checks on RHV hosts, including information from the bootloader, host mount points, storage logical volumes, and the hypervisor thin pool. For additional information, add the --debug option.

[root@hosta ~]# nodectl check
Status: OK
Bootloader ... OK
  Layer boot entries ... OK
  Valid boot entries ... OK
Mount points ... OK
  Separate /var ... OK
  Discard is used ... OK
Basic storage ... OK
  Initialized VG ... OK
  Initialized Thin Pool ... OK
  Initialized LVs ... OK
Thin storage ... OK
  Checking available space in thinpool ... OK
  Checking thinpool auto-extend ... OK
vdsmd ... OK

Enable the Red Hat Virtualization Host software repository on each physical host machine. Register the system first, then enable the RHEL 7 Red Hat Virtualization Host repository for packages and updates.

[root@hosta ~]# subscription-manager repos --enable=rhel-7-server-rhvh-4-rpms

Configure SSH Access

The first host will be the hypervisor for the self-hosted engine running the RHV Manager. The system SSH keys will be used to access all hypervisor hosts, including itself, with passwordless SSH. Repeat this procedure for each network interface, both management and storage, using both IP address and resolved FQDNs to prepopulate the known_hosts public key store file on the engine host. Give the password and accept the public key from the first host:

[root@hosta ~]# ssh root@hosta.example.com
[root@hosta ~]# exit
[root@hosta ~]# ssh root@hostb.example.com
...output omitted...
[root@hostb ~]# exit
[root@hosta ~]# ssh root@hostc.example.com
...output omitted...
[root@hostc ~]# exit

Generate a public/private key pair for the first host root account, so that internal RHV management access will use passwordless, key-based, SSH authentication. Do not use a pass phrase on the key pair, so that it is simple for Ansible to automate deployment and configuration processes using this key.

[root@hosta ~]# ssh-keygen -t rsa
...output omitted...
Enter passphrase (empty for no passphrase): Enter
Enter same passphrase again: Enter
The private key is saved in /root/.ssh/id_rsa. The public key is saved in /root/.ssh/id_rsa.pub.
...output omitted...

Copy the newly created SSH key pair to each host, using both IP addresses and FQDNs. Enter the root password for each host, if prompted.

[root@hosta ~]# ssh-copy-id -i /root/.ssh/id_rsa.pub root@hosta.example.com
...output omitted...
[root@hosta ~]# ssh-copy-id -i /root/.ssh/id_rsa.pub root@hostb.example.com
...output omitted...
[root@hosta ~]# ssh-copy-id -i /root/.ssh/id_rsa.pub root@hostc.example.com
...output omitted...

Configure Red Hat Gluster Storage

Before the hosted engine can be installed, the attached storage devices on all hosts must be configured for Red Hat Gluster Storage use. All host-attached disks, other than the just installed RHV-H system disk, must be empty and have no existing partitions or labels. Log in to the Web Console on the first host, choose the Virtualization Hosted Engine option, and then click the Start button for the Hyperconverged installation.

Figure 14.2: The Gluster Configuration window opens

Red Hat Gluster Storage is deployed using a wizard. RHHI-V pods are always installed in multiples of three hypervisor hosts. The initial installation is always accomplished by installing the first three hosts, and then adding hyperconverged hosts later in additional groups of three hosts.

Specify the storage network FQDNs for each of the three hyperconverged hosts. The first hyperconverged host, from which the SSH key pair was generated and distributed, must always be entered as Host1, since that host is the engine host and will run deployment tasks. If you are using arbitrated replicated volumes instead of standard replication, as recommended, list the host that will have the arbiter brick as Host3.

Figure 14.3: Specify hyperconverged hosts

The storage FQDNs were added in the previous step. On the next page, add the management network IP address or FQDN for each of the additional hosts in the RHHI-V pod. These addresses are required for these hosts to be automatically added to RHV-M as part of this deployment.

Figure 14.4: Specify management network FQDNs

The non-system storage disks will now be assigned to Gluster. Specify the names of the volumes to create. There must be at least 3 volumes. As needed, you can create multiple vmstore and data volumes, but always keep virtual machine system disk images separate from all other data disks, to facilitate more efficient back up and disaster recovery procedures.

Figure 14.5: Specify the volumes needed

Each storage disk becomes a brick. You previously mapped the volume type, brick and arbiter layout, raid levels, and naming, as preparation for entering the required fields on this screen.

Figure 14.6: Specify the brick layout

After reviewing and approving the Review, click the Deploy button to begin the Gluster deployment. When the window displays the Successfully deployed Gluster message, you can install the hosted engine. Click the Continue to Hosted Engine Deployment button to proceed with the installation.

Deploy the Hosted Engine

When the RHV Manager is installed as self-hosted, the hosted engine deployment will deploy the RHV-M appliance image as a virtual machine on the first hypervisor host. The embedded RHV-M setup creates a Default data center, and a Default cluster with your three physical hosts as members, and then enables Red Hat Gluster Storage functionality on each. All cluster hosts are configured to use the virtual-host tuned profile.

The RHV-M hosted engine is deployed using a wizard. First, specify the configuration information to use as the deployment creates the RHV-M virtual machine. You must replace the prepopulated MAC address with a valid and unique value. The password entered here will be set for the root account of this virtual machine.

Figure 14.7: Specify the hosted engine virtual machine

Enter the password for the admin account in the Administration Portal. You can also specify notification behavior here. After reviewing and approving the Prepare VM page, click the Prepare VM button to deploy the hosted engine virtual machine. When the window displays the Execution completed successfully message, you can continue.

Figure 14.8: Specify the management parameters

Specify the primary host and the location of the engine volume created during the Gluster deployment. It is critical that the storage location and mount options are properly configured in order for the engine volume to be used efficiently. After reviewing and approving the Finish page, click the Finish Deployment button to complete the hosted engine configuration. This configuration takes approximately 30 minutes. When the window displays the Hosted Engine deployment complete message, verify the deployment.

Figure 14.9: Specify the hosted engine storage

Log in to the Administration Portal at http://engine.example.com/ovirt-engine, using the admin account, and then inspect the dashboard for hosts, storage domains, and virtual machines.

Figure 14.10: Verify the hosted engine deployment in the Dashboard

Configure the Gluster Storage Domain

The final task is to add the previously configured Gluster storage as a RHV storage domain. A storage domain requires a dedicated storage logical network attached to each cluster host.

When the network is attached to all hosts, verify the network health. In the Network Interfaces tab, inspect the state of the host network. If any network interface displays an Out of sync state, or is missing an IP Address, click the Refresh Capabilities button to synchronize the configuration.

The standard RHHI-V pod deployment is complete. If additional hosts beyond the first three are needed, they can be added now through the Administration Portal. Each host must have already been installed with the RHV-H operating system, configured for passwordless SSH access, and configured for Gluster storage, the same as the original three hosts. On the Hosts page, open the New Host window, clear the automatic firewall configuration, and select the hosted engine Deploy action. Finally, attach the gluster network to complete the new host deployment tasks.

Use the Administration Portal to verify the full RHHI-V pod installation. Inspect the hosts, storage domains, clusters, logical networks, and virtual machines.