Protect the Management Domain vCenter and Configure VCF Single Sign-On with VCF Identity Broker
Objectives
- Explain how the VCF 9.0 management domain vCenter (vCenter 9.0, deployed by the VCF Installer) is protected: vSphere HA only, plus file-based backup; vCenter HA and vSphere FT are not supported in VCF
- Verify the vSphere HA protection and the file-based backup configuration of the management domain vCenter and SDDC Manager from VCF Operations
- Configure VCF Single Sign-On from VCF Operations: choose the VCF Identity Broker deployment mode (embedded in the management vCenter, or three-node appliance) and connect an external identity provider (Okta, Ping Identity, Microsoft Entra ID, ADFS, generic SAML 2.0, AD/LDAP or OpenLDAP) with SCIM, JIT or AD/LDAP user and group provisioning
- Connect vCenter, NSX, VCF Operations and VCF Automation to the VCF Identity Broker and assign service roles to the provisioned users and groups in each component
- Validate Single Sign-On, the identity broker health view, the local-account fallback, and a vSphere HA restart of the vCenter appliance; design tenant-scoped RBAC in vCenter and NSX
Prerequisites
A VCF 9.0 instance with the management domain deployed by the VCF Installer and healthy: SDDC Manager, management vCenter 9.0, NSX Manager, VCF Operations and the VCF Operations fleet management appliance. The management vCenter carries a VCF license and is not part of Enhanced Linked Mode (ELM). An external identity provider is reachable from the management network: a Microsoft Entra ID tenant, an Okta or Ping Identity tenant, or an Active Directory domain for AD/LDAP. An SFTP server is reachable for file-based backups.
Prior labs: vcp-admin-03 (VCF 9.0 management model: SDDC Manager, VCF Operations, fleet management), holodeck-02 (if using a nested Holodeck environment)
Required skills:
- VCF 9.0 architecture (VCF Instance, management domain, fleet-level components)
- vSphere HA and vCenter appliance administration (VAMI, file-based backup)
- SAML 2.0 and OpenID Connect concepts; SCIM provisioning
- Active Directory group structure and LDAP bind accounts
- vCenter permissions model and NSX role-based access
Lab Environment
VCF 9.0 management domain: 4 nested ESX hosts in the default management cluster on vSAN; one management vCenter 9.0 (protected by vSphere HA), one three-node NSX Manager cluster, SDDC Manager, VCF Operations, VCF Operations fleet management appliance. Post-lab: VCF Single Sign-On configured for the instance (embedded identity broker in the management vCenter, or a three-node VCF Identity Broker appliance deployed from fleet management), connected to the external identity provider; vCenter, NSX, VCF Operations and VCF Automation configured as SSO clients.
Mermaid source
graph TB IdP[Corporate identity provider: Entra ID / Okta / Ping / ADFS / AD-LDAP] -->|SAML 2.0 or OIDC + SCIM/JIT/LDAP sync| VIDB[VCF Identity Broker: embedded in mgmt vCenter, or 3-node appliance] VIDB --> VC[Management vCenter 9.0] VIDB --> NSX[NSX Manager cluster] VIDB --> OPS[VCF Operations] VIDB --> AUTO[VCF Automation] VC -.vSphere HA.-> ESX1[ESX-01] VC -.-> ESX2[ESX-02] VC -.-> ESX3[ESX-03] VC -.-> ESX4[ESX-04] VC -->|file-based backup| SFTP[SFTP server]
IP Addressing
| Network | Purpose | VLAN |
|---|---|---|
10.1.0.0/20 | Holodeck Site A master CIDR (default), as in holodeck-02: New-HoloDeckNetworkConfig subdivides it into the Site A subnets on VLAN 0 and VLANs 10-25. The VCF 9.0 management components (SDDC Manager, vCenter, NSX, VCF Operations, fleet management, identity broker appliance nodes) use the untagged Untagged-HoL management subnet inside it (gateway 10.1.1.1; e.g. esx-01a 10.1.1.10). Outside Holodeck, substitute your own management subnet and VLAN. | VLANs 0 and 10-25 (Site A default); management on VLAN 0 (untagged Untagged-HoL subnet) |
Credentials
| System | Username | Password |
|---|---|---|
| VCF Operations | admin (local) | Set during VCF Installer deployment; used as the local admin for SSO configuration and role assignment |
| Management vCenter 9.0 | administrator@vsphere.local | Set during VCF Installer deployment; remains the local fallback after SSO is configured |
| SDDC Manager | admin@local | Set during VCF Installer deployment; SDDC Manager UI accepts only local accounts after VCF SSO is configured |
| NSX Manager | admin | Set during VCF Installer deployment |
| External identity provider | test users in groups vcf-admins, vcf-readers, vcf-developers | Configured in the identity provider tenant |
Tasks
Task 1 Document and Verify How the Management vCenter Is Protected in VCF 9.0
AvailabilityThe management domain vCenter is deployed by the VCF Installer during bring-up; you do not deploy it yourself. VCF supports exactly one availability method for VCF vCenters: vSphere HA. vCenter HA (active/passive/witness) and vSphere FT are listed as not supported in the VCF 9.0 design guidance (Table 670, Methods for Protecting vCenter). Recovery of a lost appliance is by file-based backup and restore. This task makes you verify the protection that exists instead of building one that is not supported.
Write down the three rows of Table 670 'Methods for Protecting vCenter' from the VCF 9.0 design guidance: vSphere High Availability = Supported in VCF: Yes (you must leave enough cluster capacity for VMs to restart on another host); vCenter High Availability (vCenter HA) = Supported in VCF: No (services fail over to a passive node, no continuous availability, recovery time up to 15 minutes, extra private network, extra resources for passive and witness nodes, and lifecycle management is complicated because the standby VMs must be deleted and recreated for every lifecycle operation); vSphere Fault Tolerance = Supported in VCF: No (the FT vCPU limit would cap the appliance at Medium, and FT needs a dedicated network). Add the design-level alternative the same section gives for protecting the whole management stack rather than one vCenter: multi-rack management vSphere clusters, vSAN stretched clusters, or third-party synchronous replication (metro clusters).
In the vSphere Client of the management vCenter, select the management cluster > Configure > vSphere Availability. Confirm vSphere HA is turned on, note the Host Failure Response (Restart VMs) and the admission control setting. Then open VMs > the vCenter appliance VM > Configure > VM Overrides (or the cluster's VM Overrides list) and record the VM restart priority applied to the vCenter appliance and to the NSX Manager nodes.
In VCF Operations, navigate to Administration > SDDC Manager, select the VCF Instance, open the Backup Settings tab and read Site Settings (the SFTP target: host, port 22, backup directory, SSH fingerprint, encryption passphrase) and SDDC Manager Configurations (the backup schedule). If no SFTP target is configured yet, configure it now with your lab SFTP server and enable the schedule.
Open the management vCenter's appliance management interface (https://<vcenter_fqdn>:5480), go to Backup, and schedule a file-based backup to the same SFTP server. Set the start time so that the vCenter job and the SDDC Manager job start within the same 5-minute window (documented requirement for consistent restore points, Table 948 File-Based Backup Jobs).
Run the SDDC Manager backup once (Backup Settings > SDDC Manager Configurations > Backup Now) and a vCenter backup once (VAMI > Backup > Backup Now). Verify both files appear on the SFTP server.
Validation Gate
Check: Table 670 recorded; vSphere HA confirmed on the management cluster with the restart priorities noted; SDDC Manager and vCenter file-based backups scheduled to the SFTP target within the same 5-minute window and run once successfully.
Expected: The management vCenter's documented protection (vSphere HA + file-based backup) is verified and recorded; no unsupported availability method has been introduced.
Common Errors
Task 2 Configure VCF Single Sign-On: Deployment Mode and Identity Provider
SecurityVCF Single Sign-On is configured from the VCF Operations console (Fleet Management > Identity & Access), not in the individual components. The authentication source is the VCF Identity Broker, which either runs embedded in the management domain vCenter or as a three-node appliance deployed through fleet management. The identity provider is connected with SAML 2.0 or OIDC (or AD/LDAP directly), and users and groups are provisioned with SCIM, JIT or AD/LDAP sync. Passwords are never synced.
Check the prerequisites before you start: the management vCenter is at 9.0 or later with a VCF license, and it is not part of Enhanced Linked Mode (ELM). If it is in ELM (typical after an upgrade from 5.x), deactivate ELM first with cmsso-util break-elm or the SDDC Manager API and resolve the ELM drift; otherwise the vCenter cannot be integrated into VCF SSO. In VCF Operations, confirm your user has the 'All Objects' scope (Administration > Control Panel > Access Control).
In VCF Operations, select Fleet Management > Identity & Access. First time only: review and acknowledge the prerequisite check boxes and click Continue. Click SSO Overview, open the Select a VCF Instance drop-down and select your VCF Instance (Step 1: Select a VCF Instance).
Click Start against Choose deployment mode and select the mode. Decide with the documented criteria: Identity broker (embedded) runs as a service inside the management domain vCenter, needs no deployment, is backed up with vCenter, is protected by vSphere HA, and is recommended for one VCF Instance; its single point of failure is the management vCenter itself. Identity broker appliance is a three-node cluster deployed through Fleet Management > Lifecycle (Add on identity broker, after downloading the binary into Binary Management), each node 8 vCPU and 16 GB RAM, 700 GB storage per cluster, tolerates a single node failure, and one appliance can serve up to five VCF Instances. For this lab choose embedded unless your nested hosts have the capacity for the appliance. Click Next.
Click Start against Configure Identity Provider and, under Choose Identity Provider, select your provider: Okta, Ping Identity, Microsoft Entra ID, Microsoft ADFS, Generic SAML 2.0, Active Directory (AD/LDAP) or OpenLDAP. For Entra ID or Okta, in your identity-provider tenant create an application for VCF first: an OIDC web application (note the client identifier and shared secret, and the tenant's OpenID configuration address) or a SAML 2.0 application. Create three test groups in the identity provider: vcf-admins, vcf-readers, vcf-developers, with at least one test user each.
Click Configure under Configure the Identity Provider. In Authentication Method select OIDC (OpenID Connect) or SAML. For OIDC enter: IDP Display Name, Client Identifier, Shared secret, OpenID Address (the provider's OIDC configuration endpoint); copy the Redirect URI shown by VCF Operations back into the application in your identity provider. Click Next.
In User/Group Provisioning Method choose how users and groups reach the identity broker: SCIM (the identity provider pushes users and groups; you will configure the SCIM application in the provider with the values VCF shows), JIT (users and groups are created at first login from the token or assertion; choose Groups Pre-Provisioning if you want only known groups available for role assignment, or Just-In-Time Group Provisioning to create groups dynamically), or AD/LDAP (users and groups are synced from Active Directory on a schedule, default weekly; you enter the directory details, domain controllers, Base DN and bind account, and the unique identifier that maps the OIDC user to the AD user). For this lab choose JIT with Groups Pre-Provisioning, add your domain, and pre-provision vcf-admins, vcf-readers and vcf-developers. Review the Attributes screen, then Finish.
Validation Gate
Check: Deployment mode chosen for the documented reasons; identity provider configured with OIDC or SAML; provisioning method chosen and the three groups available for role assignment.
Expected: VCF Single Sign-On identity source configured for the VCF Instance from VCF Operations.
Common Errors
Task 3 Connect vCenter, NSX, VCF Operations and VCF Automation to the Identity Broker and Assign Service Roles
SecurityConfiguring the identity source does not grant anyone access. Each component is connected to the identity broker as an SSO client from VCF Operations, and then roles must be assigned in each component to the provisioned users and groups. Without those role assignments, enterprise logins fail even though authentication succeeds.
Step 4 of the documented flow: in VCF Operations, Fleet Management > Identity & Access, select your VCF Instance and click Start against Configure Components. On the Configure clients page, select the management vCenter and the NSX Manager and click Configure. Tick 'I confirm that I understand the requirement to perform role assignments in order to enable SSO for the selected component(s)', Continue, then Finish Setup and Continue.
Step 5: from the Identity & Access pane, click VCF Management > operations appliance, click Continue under Enable Single Sign-On, acknowledge the role-assignment requirement, select your identity broker from the Identity Broker drop-down and click Configure. Repeat for VCF Management > automation appliance if VCF Automation is deployed.
Step 7, vCenter: log in to the vSphere Client as administrator@vsphere.local, go to Administration > Access Control > Global Permissions (or the object where you want the permission) and add permissions for the provisioned groups: vcf-admins = Administrator at the root with Propagate; vcf-readers = Read-only at the root with Propagate; vcf-developers = a custom role you create under Administration > Access Control > Roles (for example Virtual machine privileges without Datastore.Delete file and Host.Configuration), assigned at the root with Propagate.
Step 7, NSX: log in to NSX Manager as the local admin, go to System > User Management (Users and Roles), add role assignments for the same groups from the identity broker source: vcf-admins = Enterprise Admin, vcf-readers = Auditor, vcf-developers = Network Admin (or a project-scoped role if you use NSX Projects).
Step 7, VCF Operations: log in as the local admin, go to Administration > Control Panel > Access Control, open the User Groups tab, click the ellipsis next to Add and select Import from Source, choose VCF SSO as the source, search for the three groups and Finish. Then select each imported group, Edit, and under Assign Roles and Scope assign a role (Administrator for vcf-admins; Read Only for vcf-readers; a custom or content-level role for vcf-developers) with the appropriate object scope. If VCF Automation is deployed, log in to the Provider Management Portal with the system organization admin, Access Control > Users > Import Users, Source = VCF SSO, and assign roles.
Validation Gate
Check: vCenter and NSX configured as SSO clients; VCF Operations (and VCF Automation) configured for VCF SSO; roles assigned to vcf-admins, vcf-readers and vcf-developers in each component.
Expected: Enterprise users can log in to every connected component with the role their group carries.
Common Errors
Task 4 Validate Single Sign-On, Identity Broker Health, Local Fallback and vSphere HA Recovery
RecoverabilityA production identity design must answer three questions: do users get a single sign-on experience, what happens when the identity broker or its host vCenter is unavailable, and how is the vCenter appliance itself recovered. This task checks each against the documented behaviour: single sign-on with single-component logout, local admin accounts as the fallback, vSphere HA restart for the appliance, and file-based restore as the recovery method.
In a private browser window, open the vSphere Client of the management vCenter and sign in with the vcf-admins test user through the identity provider. In new tabs open the NSX Manager UI and the VCF Operations console. Confirm you are not prompted to authenticate again and that the role in each console matches the group assignment. Then log out of NSX only and confirm the vCenter and VCF Operations sessions remain active.
In VCF Operations, navigate to Inventory > Integrations > Infrastructure Management > Identity Broker health and record the Health (up and running) and Readiness (accepting requests) status of the identity broker. From Fleet Management > Identity & Access > SSO Overview, click the identity broker and view the VCF Instances and components connected to it.
Fallback test: sign out everywhere, then sign in to the vSphere Client as administrator@vsphere.local, to VCF Operations as the local admin, and to SDDC Manager as admin@local. These local accounts remain valid after VCF SSO is configured and are the accounts you use for SSO reset and recovery. Record their password status in VCF Operations > password management (unified password management of the break-glass accounts).
vSphere HA recovery test (lab only): in the vSphere Client, identify the ESX host running the vCenter appliance VM. From the ESX Host Client of that host (https://<esx_fqdn>/ui) or the outer vSphere layer in a nested lab, power off that host abruptly (or stop its VM in the outer vSphere). Wait for vSphere HA to restart the vCenter appliance on another host. Measure the time until the vSphere Client login page responds again and note that enterprise logins (embedded mode) return at the same moment.
Recovery method walkthrough (document, do not execute): read 'Restore a vCenter Instance from a File-Based Backup'. Record the method: download the vCenter ISO for the same build, run the UI installer and choose Restore; Stage 1 deploys a new appliance (for the management vCenter, on a healthy ESX host in the management cluster) from the SFTP backup location; Stage 2 completes the file-based restore with the encryption password. Note the rule that the management domain vCenter must be restored before a workload domain vCenter.
Validation Gate
Check: SSO verified with role enforcement and single component logout; identity broker health and SSO Overview recorded; local fallback accounts verified; vSphere HA restart of the vCenter appliance observed and timed; file-based restore method documented.
Expected: Authentication and vCenter availability behaviour match the VCF 9.0 documentation and are recorded with measured values.
Common Errors
Task 5 Design Tenant-Scoped Access Control in vCenter and NSX on Top of VCF SSO
SecurityVCF SSO centralizes who a user is. What each team may do is still decided per component, by vCenter permissions on inventory objects and by NSX roles. This task builds a two-team model where each team manages only its own folder and network objects, all authenticated through the same identity provider.
In the identity provider, create two groups, team-a-admins and team-b-admins, each with one test user. Make them available to the identity broker: with Groups Pre-Provisioning, add them in VCF Operations > Fleet Management > Identity & Access > Identity Source > Edit (Pre-provisioning Domains and Groups); with SCIM, push them from the identity provider; with AD/LDAP, add them to the synced groups and trigger an on-demand sync.
In vCenter, create two VM folders, Tenant-A and Tenant-B (VMs and Templates > New Folder). Create a custom role Tenant-Admin under Administration > Access Control > Roles with the Virtual machine privilege groups (Interaction, Configuration, Inventory, Snapshot management) and Resource > Assign virtual machine to resource pool, without Datastore or Host privileges.
Assign permissions on the folders, not at the root: Tenant-A folder > Permissions > Add: group team-a-admins, role Tenant-Admin, Propagate to children; Tenant-B folder likewise for team-b-admins. Make sure neither group has a permission at the vCenter root or in Global Permissions; a root-level permission with Propagate would grant the whole inventory.
Log in as the team-a user through SSO: the inventory shows only Tenant-A and its VMs (parent objects appear read-only as a path), Tenant-B is not visible, and host actions are refused. Repeat for the team-b user.
In NSX, scope network administration per team. If your NSX deployment uses Projects, create Project-A and Project-B and assign team-a-admins and team-b-admins the Project Admin role on their project only; otherwise assign Network Admin to one team and document that NSX roles without Projects are global, so tenant separation then relies on operational process and DFW grouping rather than on RBAC scope.
Draw the access-control architecture: identity provider groups -> VCF Identity Broker (embedded or appliance) -> component role assignments (vCenter folder permissions, NSX project roles, VCF Operations roles and scope). Mark where tenant A's authority ends and tenant B's begins, and list the trust assumptions (both tenants trust the same identity provider and the same identity broker; a compromise of either affects both).
Validation Gate
Check: Two team groups provisioned; folder-scoped vCenter permissions enforced; NSX per-project roles assigned or the gap documented; architecture diagram with trust assumptions.
Expected: Tenant-scoped RBAC implemented on top of VCF SSO and defensible.
Common Errors
Final Validation
The management vCenter's documented protection (vSphere HA plus file-based backup; vCenter HA and vSphere FT not supported in VCF) is verified and recorded. VCF Single Sign-On is configured from VCF Operations with a chosen identity broker deployment mode and an external identity provider; vCenter, NSX, VCF Operations and VCF Automation are SSO clients with service roles assigned. SSO, identity broker health, local fallback accounts and a vSphere HA restart of the vCenter appliance are validated, and tenant-scoped RBAC is implemented.
✓ Table 670 recorded and vSphere HA verified on the management cluster → vSphere HA on; restart priorities for vCenter and NSX Manager noted; no vCenter HA or FT
✓ File-based backups configured → SDDC Manager and vCenter backups to SFTP within the same 5-minute window; NSX hourly backup present; one run of each succeeded
✓ VCF SSO identity source configured from VCF Operations → Deployment mode chosen with reasons; identity provider connected (OIDC or SAML); groups provisioned
✓ Components connected and roles assigned → vCenter, NSX, VCF Operations (and VCF Automation) configured as SSO clients; three groups hold roles in each
✓ SSO, health, fallback and HA recovery validated → SSO across consoles with single component logout; identity broker healthy and ready; local accounts work; vSphere HA restarted the vCenter appliance and the time was recorded
✓ Tenant-scoped RBAC → Each team sees only its folder; NSX project roles or the documented gap
Cleanup / Restore
Snapshot: vcp-admin-04-complete
• Record the identity broker deployment mode, identity provider, provisioning method and the groups with their roles per component in your lab notebook
• Keep the SFTP backup target and the encryption passphrase in your password manager
• Power the failed lab host back on and confirm it rejoins the management cluster; clear any vSphere HA alarms
• Keep the access-control architecture diagram for the design defence
Design Reflection (VCDX)
A panel will test two things here. First, that you know what VCF supports for vCenter availability: vSphere HA only, with vCenter HA and vSphere FT explicitly not supported (Table 670), and that recovery is by file-based restore of a new appliance, management domain first. Proposing vCenter HA in a VCF design is a defence-ending error.
Second, that you can place the identity broker deliberately: embedded (simplest, one per instance, backed up and protected with vCenter, but the management vCenter becomes the authentication single point of failure) versus the three-node appliance (independent of vCenter maintenance, tolerates a node failure, up to five instances per broker, more to operate). Then the SSO topology: fleet-wide (one broker, large SSO scope, large blast radius), cross-instance (several brokers each serving a set of instances; VCF Operations and VCF Automation connect to one of them), or single-instance.
Be ready to explain why identity source management moves to VCF Operations, why ELM must be deactivated first, how the SDDC Manager UI is affected, and how PowerCLI keeps working.
Requirements
- The management vCenter must be protected with vSphere HA and file-based backups to an SFTP target (SDDC Manager and vCenter jobs starting within the same 5-minute window)
- Enterprise users must authenticate once to vCenter, NSX, VCF Operations and VCF Automation through the corporate identity provider
- Identity source configuration must be managed centrally in VCF Operations
- Role assignments must exist in every component for each provisioned group before SSO goes live
- Local administrator accounts must remain available and their passwords managed as break-glass accounts
Constraints
- vCenter HA and vSphere FT are not supported for VCF vCenters (Table 670)
- VCF SSO requires vCenter 9.0 or later with a VCF license and no Enhanced Linked Mode
- The embedded identity broker serves one VCF Instance; the appliance serves up to five
- Identity broker appliance: three nodes, 8 vCPU and 16 GB RAM each, 700 GB storage per cluster
- SDDC Manager UI accepts only local accounts after VCF SSO is configured
- Changing the identity provider or the deployment mode requires a reset of VCF SSO and re-adding users to components
Assumptions
- The VCF Instance was deployed by the VCF Installer with a management domain and VCF Operations (vcp-admin-03)
- An external identity provider tenant or an Active Directory domain is reachable from the management network
- An SFTP server is available for file-based backups
- The nested lab has spare capacity to lose one management host for the vSphere HA test
Risks
- Embedded identity broker: a management vCenter outage removes enterprise authentication for all connected components. Impact: operators fall back to local accounts. Mitigation: vSphere HA, restart priority, documented local-account procedure; appliance mode where uptime requirements demand it
- Identity provider outage or compromise. Impact: no enterprise logins, or forged access to every connected component. Mitigation: local break-glass accounts under unified password management, short token lifetimes, audit of role assignments
- Backup jobs out of alignment. Impact: inconsistent restore point between SDDC Manager and vCenter. Mitigation: same 5-minute start window, monitored SFTP capacity
- Root-level permissions granted to tenant groups. Impact: cross-tenant visibility. Mitigation: folder-scoped permissions only, periodic permission review
Self-Assessment Discussion Prompts
- Why does VCF support only vSphere HA for its vCenters, and how do you meet a demanding RTO for the management stack without vCenter HA?
- Embedded or appliance identity broker: which would you choose for a single-instance platform, and what changes your answer for a five-instance fleet?
- Fleet-wide, cross-instance or single-instance SSO topology: how do you trade blast radius against operational overhead and SSO scope?
- What breaks for operators after VCF SSO is configured (SDDC Manager UI, identity sources in vCenter, PowerCLI), and how do you prepare them?
- SCIM, JIT or AD/LDAP provisioning: which fits an organisation that already governs group membership in Active Directory, and why?
- How do you demonstrate to an auditor that tenant A cannot see tenant B when both authenticate through the same identity broker?
Extensions
Deploy the Identity Broker Appliance and Change the Deployment Mode
Download the VCF Identity Broker binary into Binary Management, deploy the three-node appliance from Fleet Management > Lifecycle, then reset VCF SSO (full reset) and reconfigure in appliance mode. Record how long the reset and re-assignment of roles takes: this is the migration cost the design guidance warns about.
harderConnect a Second VCF Instance to the Existing SSO Configuration
With a second instance available, use Connect to an existing SSO configuration in Identity & Access and configure its components against the first instance's identity broker. Discuss the cross-instance model and which VCF Operations console manages the identity source.
harderRestore the Management vCenter from File-Based Backup
In a disposable lab, destroy the management vCenter appliance and restore it with the vCenter ISO installer's Restore path (Stage 1 new appliance on a healthy management host, Stage 2 restore from SFTP). Confirm VCF SSO (embedded mode) and the SDDC Manager inventory return.
harderPowerCLI with VCF Single Sign-On
Generate an OIDC client under Fleet Management > Identity & Access > VCF Other Components and connect to vCenter with New-OAuthSecurityContext, New-VISamlSecurityContext and Connect-VIServer -SamlSecurityContext.
sameReferences
- VMware Cloud Foundation 9.0 documentation: Identity and Access Management (VCF Single Sign-On, Deployment Modes of the VCF Identity Broker, Configure a New VCF Single Sign-On for a VCF Instance, Steps 1-7)Tier 1 — Official
Procedures in Tasks 2-4. In the VCF 9.0.1 documentation PDF: pp.2837-2918. - VMware Cloud Foundation 9.0 design: vCenter design, Table 670 Methods for Protecting vCenterTier 1 — Official
vSphere HA supported; vCenter HA and vSphere FT not supported. VCF 9.0.1 PDF p.1092. Task 1. - VMware Cloud Foundation 9.0 design: VCF Identity Broker Detailed Design and VCF Single Sign-On Detailed Design (embedded and appliance models; fleet-wide, cross-instance and single-instance SSO)Tier 1 — Official
VCF 9.0.1 PDF pp.1129-1138. Design reflection. - VMware Cloud Foundation 9.0: File-Based Backups for SDDC Manager, NSX Manager and vCenter; Restore a vCenter Instance from a File-Based BackupTier 1 — Official
Table 948 backup jobs and the restore method. VCF 9.0.1 PDF pp.3006-3007, 3015. Tasks 1 and 4. - VMware Cloud Foundation 9.0: Deploying VCF Identity Broker (appliance mode from fleet management)Tier 1 — Official
VCF 9.0.1 PDF pp.1300-1302. Extension 1. - Microsoft Entra ID / Okta / Ping Identity application setup for OIDC and SAMLTier 2 — Supporting source
Identity-provider side configuration for Task 2 (Okta: developer.okta.com; Ping: docs.pingidentity.com).