Cisco Intersight provides an installation wizard to install, configure, and deploy Cisco HyperFlex clusters — HyperFlex Edge, FI-attached, and HyperFlex Datacenter without Fabric Interconnect. The wizard constructs a pre-configuration definition of your cluster called an HyperFlex Cluster Profile. This definition is a logical representation of the HyperFlex nodes in your HyperFlex cluster and includes:

• Security—credentials for HyperFlex cluster such as controller VM password, Hypervisor username, and password.

• Configuration—server requirements, firmware, etc.

• Connectivity—upstream network, virtual network, etc.

HyperFlex Cluster Profiles are built on policies which administrator defined sets of rules and operating characteristics such as the node identity, interfaces, and network connectivity. Every active node in your HyperFlex cluster must be associated with an HyperFlex Cluster Profile.

After gathering the node configuration settings to build the HyperFlex Cluster Profile, the installation wizard will validate and deploy the HyperFlex Cluster Profile in your HyperFlex cluster. You can clone a successfully deployed HyperFlex Cluster Profile, and then use that copy as the basis to create a new cluster. For instructions on cloning HyperFlex cluster profiles, see Cloning HyperFlex Cluster Profiles.

Non Pre-configured Cisco HyperFlex Systems

Note

Beginning in April 2024, HyperFlex servers are shipped from the factory without VMware ESXi preinstalled. It is imperative that the ESXi hypervisor is installed before starting the HyperFlex Installation. For instructions to manually prepare factory shipped servers for the Cisco HyperFlex install, seeCisco HyperFlex install, see Cisco HyperFlex Systems Installation Guide for VMware ESXi.

Navigate to CONFIGURE > Policies > Create Policy > HyperFlex Cluster. HyperFlex Policies in Cisco Intersight provide different configurations including Auto Support, External Storage (such as FC and iSCSI), security, network configuration and more. A policy that is once configured can be assigned to any number of servers in order to provide a configuration baseline.

HyperFlex Policies can also be cloned by using the Policy Clone wizard with properties that are similar to the existing policies. The clone policy action is available on both the policies list and detailed views.

All HyperFlex policies can be shared by HyperFlex Edge and HyperFlex with Fabric Interconnect, unless otherwise mentioned below:

• Auto Support Policy—Auto Support is the alert notification service provided through HyperFlex Data Platform in Intersight. If enabled, notifications are sent to designated email addresses or email aliases that you want to receive the notifications. Typically, Auto Support is configured during HyperFlex storage cluster creation by configuring the SMTP mail server and adding email recipients.

• Backup Configuration Policy—The backup policy consists of the Edge cluster datastores being protected, the backup interval, snapshot retention value, and the backup target cluster. All virtual machines residing in the protected Edge cluster datastore will be automatically protected by the backup policy. This includes VMs created in the protected datastore, as well as VMs migrated into the protected datastore.

• DNS, NTP, and Timezone Policy—Configures DNS, NTP, and Timezone on all servers. DNS and NTP servers should reside outside of the HyperFlex storage cluster. Use an internally-hosted NTP server to provide a reliable source for the time.

• External FC Storage Policy—Enables the use of external FC Storage and configures the WWxN name and the associated VSAN details. This policy is not applicable to HyperFlex Edge clusters.

• External iSCSI Storage Policy—Enables the use of external iSCSI Storage and configures the associated VLAN details. This policy is not applicable to HyperFlex Edge clusters.

• HTTP Proxy—Specifies the HTTP proxy settings to be used by the HyperFlex installation process and the HyperFlex Storage Controller VMs. This policy is required when the internet access of your servers including CIMC and HyperFlex Storage Controller VMs is secured by an HTTP proxy.

• Network Configuration Policy—Configures the VLAN and KVM for the management network in Fabric Interconnects; and Uplink Speed, VLAN, and Jumbo Frames for the management network in Edge clusters. The VLAN must have access to Intersight. This policy cannot be shared by HyperFlex Edge and HyperFlex with Fabric Interconnect clusters.

• Node IP Ranges Policy—Configures the management IP ranges for hypervisors and controller VMs. The data IPs are automatically assigned in a /24 subnet in the range 169.254.x.2 to 169.254.239.254.

• Replication Network Configuration Policy—The replication network policy consists of replication VLAN, gateway, subnet mask, bandwidth, MTU, and replication IP address range parameters. The replication network policy is unique for each Edge cluster configured to use N:1 Replication.

• Security Policy—Configures ESXi and Controller VM password for the HyperFlex cluster. This policy presents an option to update the Hypervisor password in Intersight, if you have not already changed it on the Hypervisor.

• Storage Configuration Policy—Configures the options for VDI Optimization (for hybrid HyperFlex systems). For HyperFlex with Fabric Interconnect and DC-No-FI, this policy provides the option to enable Logical Availability Zones as part of the Cluster Deployment workflow. Logical Availability Zones are not supported on HyperFlex Edge clusters.

  Note

  Logical Availability Zones are automatic partitions of the physical cluster into multiple logical zones. They are created to avoid multiple node and component failures on large clusters, and to increase cluster resiliency. HyperFlex Data Platform intelligently places a copy of data in every zone. When there is a node failure in a single zone, it does not cause the entire cluster to fail because the other zones contain data replicas. Logical Availability Zones can only be enabled with clusters that are 8 converged nodes or larger. The LAZ option in the Storage Configuration policy is recommended for clusters greater than 8 nodes.

• vCenter Policy—An optional policy during installation of the HyperFlex cluster. However, post-installation, you must register the cluster to vCenter to ensure that the cluster functions smoothly. This policy is optional to the installation of the HyperFlex cluster.

• DC-No-FI Policy—Cisco HyperFlex Datacenter without Fabric Interconnect (DC-No-FI) brings the simplicity of hyperconvergence to data center deployments without the requirement of connecting the converged nodes to Cisco Fabric Interconnect.