Prerequisites for Disaster Recovery for Dell

Introduction

This section covers what to confirm and plan for before deploying Superna Disaster Recovery Edition: platform and release compatibility, system requirements for the Eyeglass appliance, licensing, and network port requirements. For a conceptual introduction to Disaster Recovery and the terminology used throughout this documentation, see the Disaster Recovery overview; for failover types, topologies, and planning depth, see the Design Guide.

Release Compatibility

Platform Support

info

Platform and version compatibility is maintained on a single page. See Platform Support for the supported OneFS, ECS, hypervisor, and operating system versions.

note

Disaster Recovery is not supported on Dell ECS platforms.

Operating System and Hypervisor Support

Supported Operating Systems: openSUSE Leap, SUSE Linux Enterprise Server Supported Hypervisors: VMware, Hyper-V, Nutanix

System Requirements

Virtualization Platforms

• VMware vSphere: Version 6.0 ESX host or higher is required.
• Microsoft Hyper-V: Appliance deployment requires support for the VHDX format.
• Nutanix AHV: A deployment procedure is available for Superna appliances, but Nutanix AHV is not a formally supported platform — contact Superna Support before deploying to Nutanix AHV in production.
• VMware vCenter: vCenter 7 and vCenter 8 are supported.

info

For Hyper-V-specific deployment steps, see Deploy on Hyper-V in the Installation Guide.

System Resources (Eyeglass Appliance)

• CPU: 4 vCPU
• Memory: 24 GB RAM (may need to be upgraded based on scalability requirements — see Scalability Limits below)
• Disk Space:
  • Operating System Partition: 58 GB
  • Additional Disk Space: 140 GB
  • Total Disk Size in VMware: 198 GB

Scalability Limits

Eyeglass Scalability Limits and Appliance Memory Minimum Requirements

Scaling Limit Area	Tested Scaling Limits	Notes
Number of Managed Clusters (1 appliance)	Manages up to 22 clusters	Contact Support for RAM requirements
SyncIQ Policies Across All Clusters	Supports > 100 policies with 64 GB RAM; > 200 policies with 84 GB RAM
Access Zones	Handles > 10 zones with 32 GB RAM; > 30 zones with 64 GB RAM; > 50 zones with 84 GB RAM	Requires 32–84 GB of RAM
Failover Job Limitations	Supports 100 policies in a single failover	Requires 64 GB RAM
Total Object Count (shares + exports + quotas)	< 5,000 objects: 16 GB RAM; 5,000–10,000: 32–48 GB RAM; > 10,000: 64 GB RAM; > 20,000: 84 GB RAM
Clusters Added to the Appliance	4 clusters: 32 GB RAM; 4–8 clusters: 64 GB RAM; > 10 clusters: 84 GB RAM
Performance Auditor	Requires a minimum of 32 GB RAM when licensed	Minimum 32 GB RAM
Concurrent Administrators (3 or more)	Adds 8 GB RAM per additional logged-in administrator

info

Each software release may adjust memory requirements, and alarm code SCA0094 recommends memory for your specific configuration. The recommendation from SCA0094 takes priority over this table.

Network Requirements

• Latency: The latency between the administration PC and the Eyeglass VM GUI must be less than 15 ms (measured by ping).

Authentication and User Management

• Active Directory (AD) Authentication: An AD authentication provider must be available in the system zone for Role-Based Access Control (RBAC) and SID-to-user resolution API requirements.

Browser Requirements

• Supported Browser: Google Chrome is required. The browser must support WebSockets.
• Unsupported Browsers: Internet Explorer is not supported.
• Cookie Settings: Third-party cookies must not be disabled, as they are required for authentication sessions and file downloads.

Licensing

Disaster Recovery uses one of two license key types per PowerScale cluster:

• DR-enabled keys provide full failover and failback automation for the cluster, plus DR state information (failover status, RPO status, compliance) for replicating cluster pairs.
• Event-monitoring-only (probe) keys provide event monitoring for a cluster without DR automation — for example, for a cold/DR-only cluster you want visibility into but do not plan to fail over.

See Add Probe License Keys for the full comparison, and License Keys for the retrieval and application procedure.

Network Ports

The Eyeglass appliance requires specific firewall ports open to the PowerScale/ECS clusters it manages, to DNS/NTP/SMTP infrastructure, and between the appliance and administrator browsers.

Port	Protocol	Source → Destination	Description
DNS 53 UDP	DNS	Appliance → DNS server (or GroupNet, if DNS is configured on all clusters)	Required for multiple validations needed for failover and Failover Readiness
TLS 443	TCP TLS 1.2	Appliance → Internet	DR Remote Monitoring / Phone Home log upload for support and health checks
NTP 123	UDP	Appliance → NTP server	Use the same NTP source as the clusters; disable VMware host VM time sync
SMTP 25	TCP	Appliance → Mail server	Email of Eyeglass alarms
HTTP 80	TCP	Appliance → openSUSE mirror repositories	Weekly security patch downloads
HTTPS 8080	TCP TLS 1.2	Appliance → PowerScale/ECS cluster	REST API, authenticated using the service account
SSH 22	AES	Appliance → PowerScale/ECS cluster	SSH access for some CLI commands
HTTPS 443	TCP TLS 1.2 AES	Admin browser → Appliance	Secures client-to-browser access
HTTPS 2011/2012 Websocket	TCP TLS 1.2 AES	Admin browser → Appliance	Real-time appliance-to-browser updates
SSH 22	TCP AES	Admin browser → Appliance	Secure shell access
SMB 445 (proxy login only)	TCP	Appliance → PowerScale/ECS	Authenticates to AD through the cluster using standard SMB authentication for role-based login proxy
Dual DNS Delegation	UDP	Appliance port 53 UDP → GroupNet DNS servers	Validates Dual DNS delegation is configured correctly
ICMP	ICMP	Appliance → PowerScale/ECS	If disabled, or if PMTUD is unsupported, OneFS defaults to a 536-byte MTU, typically degrading performance

See the Installation Guide for the complete table with additional detail.

See Also

• Design Guide – Failover types, supported topologies, and failover planning guidance.
• Installation Guide – Deploy the Eyeglass appliance and add clusters once these prerequisites are confirmed.
• Platform Support – Supported OneFS, ECS, hypervisor, and operating system versions.