Holodeck Environment Discovery Playbook
Objectives
- Discover an existing Holodeck VCF deployment systematically with a four-phase method: HoloRouter, Holodeck module and runtime, VCF APIs, then vCenter and NSX
- Use the Holodeck cmdlets (Get-HoloDeckConfig, Get-HoloDeckInstance, Get-HoloDeckSubnet, Get-HoloDeckAppNetwork, Get-HoloDeckDNSConfig, Get-HoloDeckBGPConfig, Get-HoloDeckOverlaySubnet, Get-HoloDeckAppIpPools, Get-HolodeckServiceIPPools) to reconstruct the network, DNS, IP-pool and BGP design without reading the config file
- Authenticate to the SDDC Manager API (POST /v1/tokens, Bearer token), the vSphere Automation API and the NSX Manager API, and use the API Explorer in VCF Operations
- Map the vCenter object hierarchy (datacenter, clusters, hosts, VMs, datastores, networks) and the NSX cluster, transport zones, segments and Tier-0 BGP peering to the HoloRouter
- Produce a deployment architecture record for handoff or VCDX design validation, with every fact tagged by where it was observed
Prerequisites
Access to an existing Holodeck 9.0.2 instance (shared or inherited) with the management domain running: HoloRouter SSH as root, and the documented default credentials for the VCF components (Post Deployment table) or the credentials handed over with the environment.
Required skills:
- PowerShell 7 basics (pwsh on the HoloRouter)
- REST authentication (Basic, Bearer token, session cookie)
- VCF constructs: instance, management domain, workload domain
- vSphere and NSX console navigation
Lab Environment
Unknown by design: the lab starts from an SSH login to the HoloRouter and ends with a documented topology. Expect the documented Site A defaults (10.1.0.0/20, VLANs 0 and 10-25, DNS domain site-a.vcf.lab) unless the deployer used custom values, which Phase 1 will reveal.
Mermaid source
graph LR P1[Phase 1 HoloRouter: Get-HoloDeckSubnet, AppNetwork, DNSConfig, BGPConfig] --> P2[Phase 2 Module and runtime: Get-HoloDeckConfig, Get-HoloDeckInstance, /holodeck-runtime] P2 --> P3[Phase 3 VCF APIs: SDDC Manager /v1, VCF Operations API Explorer] P3 --> P4[Phase 4 vCenter and NSX: inventory, transport zones, Tier-0 BGP] P4 --> DOC[Architecture record with OBSERVED tags]
IP Addressing
| Network | Purpose | VLAN |
|---|---|---|
10.1.0.0/20 | Default Holodeck Site A master CIDR, subdivided by New-HoloDeckNetworkConfig into the Site A subnets (management, vMotion, vSAN, host overlay, edge uplinks, and others) | VLANs 0 and 10-25 (Site A default) |
10.2.0.0/20 | Default Holodeck Site B master CIDR (dual-site only) | VLANs 40-58 (Site B default) |
<external management network> | HoloRouter external management IP (DHCP or static, set at OVA deployment); the only external IP a Holodeck environment needs | External port group |
Credentials
| System | Username | Password |
|---|---|---|
| HoloRouter (SSH and pwsh) | root | Set in the OVA Additional settings at deployment |
| HoloRouter webtop | (no authentication in 9.0.2) | http://<holorouter-ip>:30000; the docs warn not to expose it externally |
| VCF Installer | admin@local | Holodeck documented default (Post Deployment table) |
| VCF Operations | admin | Holodeck documented default (Post Deployment table) |
| Management vCenter | administrator@vsphere.local | Holodeck documented default (Post Deployment table) |
| SDDC Manager | administrator@vsphere.local | Holodeck documented default (Post Deployment table) |
| Management NSX | admin | Holodeck documented default (Post Deployment table) |
| Nested ESX hosts | root | Holodeck documented default (Post Deployment table) |
Tasks
Task 1 Phase 1: HoloRouter discovery
ManageabilityThe HoloRouter holds the network truth of the instance: subnets per VLAN, appliance hostnames and IPs, DNS records, BGP peering to the NSX Tier-0 and the reserved pools. The Get-HoloDeck* cmdlets read it without touching the config file.
SSH to the HoloRouter as root. Record the appliance facts: hostname, uptime, external IP (ip -br addr), Photon release (cat /etc/photon-release), and whether the webtop is enabled (port 30000 listening; note that in 9.0.2 it has no authentication).
ssh root@<holorouter-ip>
hostnamectl; uptime; ip -br addr; cat /etc/photon-release; ss -ltn | grep 30000Start pwsh and confirm the Holodeck module: Get-Command -Module Holodeck. Record the module version (Get-Module Holodeck | select Version) so the architecture record states which Holodeck release built the instance.
pwsh
Get-Command -Module Holodeck | select Name
Get-Module Holodeck | select Name,VersionReconstruct the network plan: Get-HoloDeckSubnet -Site a | ft -AutoSize (name, VLAN, subnet, gateway per network); if Site B exists repeat with -Site b. Compare with the documented defaults (10.1.0.0/20, VLANs 0 and 10-25; 10.2.0.0/20, VLANs 40-58) to detect custom CIDRs or VLAN ranges.
Get-HoloDeckSubnet -Site a | ft -AutoSize
Get-HoloDeckSubnet -Site b | ft -AutoSizeMap the appliances: Get-HoloDeckAppNetwork -Site a lists hostname, IP and FQDN for every nested host and VCF component. Confirm DNS with Get-HoloDeckDNSConfig (all records) and spot-check one FQDN with -FQDN esx-01a.site-a.vcf.lab. Note any additional records added with Set-HoloDeckDNSConfig (Harbor, jump hosts) as evidence of post-deployment work.
Get-HoloDeckAppNetwork -Site a | ft -AutoSize
Get-HoloDeckDNSConfig
Get-HoloDeckDNSConfig -FQDN esx-01a.site-a.vcf.labRecord the routing design: Get-HoloDeckBGPConfig -Site a (HoloRouter AS, edge AS numbers, BGP password presence) and the reserved pools: Get-HoloDeckOverlaySubnet -Site a, Get-HoloDeckAppIpPools -Site a, Get-HolodeckServiceIPPools -Site a. If BGP shows peers, an NSX Edge cluster was deployed; if the overlay subnets are reserved but no peers exist, note it as an open question for Phase 4.
Get-HoloDeckBGPConfig -Site a
Get-HoloDeckOverlaySubnet -Site a
Get-HoloDeckAppIpPools -Site a
Get-HolodeckServiceIPPools -Site aValidation Gate
Check: Appliance facts, module version, subnet plan (default or custom), appliance inventory, DNS records, BGP and pool design all recorded from the HoloRouter.
Expected: All items present in the lab notebook: Appliance facts, module version, subnet plan (default or custom), appliance inventory, DNS records, BGP and pool design all recorded from the HoloRouter.
Common Errors
Task 2 Phase 2: Holodeck module and runtime discovery
ManageabilityThe config files, the instance output and the runtime tree tell you how the instance was built: which VCF version, which options (management-only, edge cluster, Automation, Supervisor), which depot, and what the deployment state says. Read, do not edit.
List the configs and read their descriptions: Get-HoloDeckConfig. One config per deployment is the documented rule, so more than one config means more than one deployment was attempted or exists (dual site, redeploys, a 5.2 pod). Import the one that matches the running instance with Import-HoloDeckConfig -ConfigId <id> and confirm with $config.
Get-HoloDeckConfig
Import-HoloDeckConfig -ConfigId <config-id>
$configRead the instance record: Get-HoloDeckInstance -InstanceID <id> (the InstanceID is usually visible in the config description or the deployer's notes; try the common ones such as holo). Record VCF version, deployed components, networking and deployment state. Then locate the output.json it reads under /holodeck-runtime and compare.
Get-HoloDeckInstance -InstanceID holo
find /holodeck-runtime -name output.jsonWalk the runtime tree without editing anything: ls /holodeck-runtime (templates, config, bin, logs, specs); ls /holodeck-runtime/bin/* to see which VCF versions have binaries staged (9.0.2.0, 9.0.1.0, 5.2.2 and so on); ls -lt /holodeck-runtime/logs | head for the most recent deployment logs and their dates.
ls /holodeck-runtime
ls /holodeck-runtime/bin/*
ls -lt /holodeck-runtime/logs | headFrom the newest deployment log, extract the New-HoloDeckInstance options used: -Version, -ManagementOnly or full stack, -NsxEdgeClusterMgmtDomain, -DeployVcfAutomation, -DeploySupervisorMgmtDomain, -vSANMode, -CIDR, -VLANRangeStart, -DNSDomain, -DepotType, -Site. Look for Update-HoloDeckInstance runs (-AdditionalCluster, -AddVcfAutomationAllAppsOrg) and New-HoloDeckESXiNodes runs as Day 2 evidence.
grep -iE 'New-HoloDeckInstance|Update-HoloDeckInstance|New-HoloDeckESXiNodes|ManagementOnly|vSANMode|DepotType|NsxEdge|VcfAutomation|Supervisor' /holodeck-runtime/logs/<newest-log> | head -40Record the depot used (online with a build token or the Offline Depot appliance; the log shows which), the License Later start date (deployment completion date; 90 days for 9.0, 60 for 5.2) and compute the expiry. Write Phase 2 findings as OBSERVED (log or cmdlet) or ASSUMED (inferred from absence).
Validation Gate
Check: Configs and live instance identified; output.json and cmdlet agree; staged versions, deployment options, Day 2 operations, depot and licence window recorded with evidence tags.
Expected: All items present in the lab notebook: staged versions, deployment options, Day 2 operations, depot and licence window recorded with evidence tags.
Common Errors
Task 3 Phase 3: VCF API discovery
ManageabilitySDDC Manager's API is the authoritative inventory of the VCF instance: domains, clusters, hosts, credentials, bundles and tasks. In VCF 9.0 the API Explorer lives in VCF Operations (Developer Center > APIs & SDKs > API Explorer) and the token endpoint is POST /v1/tokens.
From the HoloRouter (bash or pwsh) obtain a token from SDDC Manager with the documented call: POST https://sddcmanager-a.site-a.vcf.lab/v1/tokens with the admin@local JSON body. Extract accessToken into TOKEN. Do not write the password into your notes.
TOKEN=$(curl -sk https://sddcmanager-a.site-a.vcf.lab/v1/tokens -X POST -H 'Content-Type: application/json' -d '{"username":"admin@local","password":"<admin@local password>"}' | python3 -c 'import sys,json;print(json.load(sys.stdin)["accessToken"])')
echo ${TOKEN:0:12}...Query the instance inventory: GET /v1/domains (domain names, type MANAGEMENT or VI, status, vCenter and NSX cluster references), GET /v1/clusters, GET /v1/hosts (fqdn, status ASSIGNED or UNASSIGNED_USEABLE, domain). Pretty-print and record counts and names.
H="Authorization: Bearer $TOKEN"
curl -sk -H "$H" https://sddcmanager-a.site-a.vcf.lab/v1/domains | python3 -m json.tool | head -80
curl -sk -H "$H" https://sddcmanager-a.site-a.vcf.lab/v1/clusters | python3 -m json.tool | head -60
curl -sk -H "$H" https://sddcmanager-a.site-a.vcf.lab/v1/hosts | python3 -m json.tool | grep -E '"fqdn"|"status"'Read the credential inventory as the documentation does for recovery: GET /v1/credentials?resourceType=VCENTER&accountType=USER, ?resourceType=PSC, ?resourceType=ESXI. Record the resource names and account names only; mask the values in your record.
curl -sk -H "$H" 'https://sddcmanager-a.site-a.vcf.lab/v1/credentials?resourceType=VCENTER&accountType=USER' | python3 -m json.tool | grep -E '"resourceName"|"username"'
curl -sk -H "$H" 'https://sddcmanager-a.site-a.vcf.lab/v1/credentials?resourceType=ESXI' | python3 -m json.tool | grep -E '"resourceName"|"username"'Record lifecycle evidence: GET /v1/tasks (recent tasks: bring-up, host commissioning, domain creation, upgrades) and GET /v1/bundles or the release endpoints to see which releases are available or applied. Check GET /v1/sddc-managers for the SDDC Manager appliance record and version and compare with the BOM for the instance's release.
curl -sk -H "$H" 'https://sddcmanager-a.site-a.vcf.lab/v1/tasks' | python3 -m json.tool | grep -E '"name"|"status"' | head -40
curl -sk -H "$H" https://sddcmanager-a.site-a.vcf.lab/v1/sddc-managers | python3 -m json.toolOpen VCF Operations (ops-a) and navigate to Developer Center > APIs & SDKs > API Explorer. Select the VCF instance, expand Hosts and run GET /v1/hosts with Status UNASSIGNED_USEABLE to see spare hosts the UI method exposes; compare with your curl output. Record which UI surfaces exist on this version (Fleet Management, Inventory, Lifecycle) as evidence of 9.0 rather than 5.2.
Write the Phase 3 summary: instance release, domains and their vCenters and NSX clusters, host counts by status, managed account inventory (names only), last significant tasks, and any drift between the HoloRouter view (Phase 1) and the SDDC Manager view (for example hosts built by New-HoloDeckESXiNodes but never commissioned).
Validation Gate
Check: Token obtained; domains, clusters, hosts, credentials (masked), tasks and version recorded; API Explorer used in VCF Operations; drift between HoloRouter and SDDC Manager views listed.
Expected: All items present in the lab notebook: drift between HoloRouter and SDDC Manager views listed.
Common Errors
Task 4 Phase 4: vCenter and NSX deep dive
AvailabilityThe VCF inventory says what exists; vCenter and NSX say how it is built. Map the object hierarchy and the network fabric, then confirm the Phase 1 BGP hypothesis against the NSX Tier-0 configuration.
Authenticate to the management vCenter's vSphere Automation API: POST https://vc-mgmt-a.site-a.vcf.lab/api/session with Basic auth (administrator@vsphere.local) returns a session ID; send it as the vmware-api-session-id header on later calls (vSphere Automation API, developer.broadcom.com).
SID=$(curl -sk -X POST -u 'administrator@vsphere.local:<password>' https://vc-mgmt-a.site-a.vcf.lab/api/session | tr -d '"')
curl -sk -H "vmware-api-session-id: $SID" https://vc-mgmt-a.site-a.vcf.lab/api/vcenter/datacenter | python3 -m json.toolMap the hierarchy: /api/vcenter/cluster, /api/vcenter/host (connection and power state), /api/vcenter/vm (names: the VCF appliances live in the management cluster), /api/vcenter/datastore (vSAN datastore name and capacity), /api/vcenter/network (distributed port groups per Holodeck VLAN). Record counts and names in a tree.
for o in cluster host vm datastore network; do echo "== $o"; curl -sk -H "vmware-api-session-id: $SID" https://vc-mgmt-a.site-a.vcf.lab/api/vcenter/$o | python3 -m json.tool | grep -E '"name"|"connection_state"|"power_state"|"capacity"' ; doneIn the vSphere Client confirm the cluster services (vSphere HA on, DRS mode, vSAN OSA or ESA, vSphere Lifecycle Manager image), the distributed switch MTU (9000 expected behind the HoloRouter) and the port group VLAN IDs against the Phase 1 subnet table. Note the vCenter version and build and compare with the BOM.
Authenticate to NSX Manager: POST https://nsx-mgmt-a.site-a.vcf.lab/api/session/create with j_username and j_password form fields returns a JSESSIONID cookie and an X-XSRF-TOKEN header; reuse both. Check GET /api/v1/cluster/status (cluster STABLE) and GET /api/v1/cluster/nodes (record the node count you observe; the Holodeck documentation does not state it).
curl -sk -c nsx.cookie -D nsx.hdr -X POST https://nsx-mgmt-a.site-a.vcf.lab/api/session/create -d 'j_username=admin' -d 'j_password=<password>'
XSRF=$(grep -i x-xsrf-token nsx.hdr | awk '{print $2}' | tr -d '\r')
curl -sk -b nsx.cookie -H "X-XSRF-TOKEN: $XSRF" https://nsx-mgmt-a.site-a.vcf.lab/api/v1/cluster/status | python3 -m json.tool | head -30
curl -sk -b nsx.cookie -H "X-XSRF-TOKEN: $XSRF" https://nsx-mgmt-a.site-a.vcf.lab/api/v1/cluster/nodes | python3 -m json.tool | grep -E '"fqdn"|"display_name"'Map the fabric in the NSX UI or Policy API: System > Fabric > Hosts (transport node profile, host overlay TEP pool matching Get-HoloDeckAppIpPools), Networking > Transport Zones, Networking > Segments (compare with Get-HoloDeckOverlaySubnet), Networking > Tier-0 Gateways (if present: BGP neighbours should be the HoloRouter addresses and AS numbers from Get-HoloDeckBGPConfig; status Established).
Record the NSX version and build (System > Appliances or GET /api/v1/node/version) against the BOM, and whether the Tier-0 is active-standby or active-active, how many edge nodes exist, and which segments carry workloads.
Assemble the deployment architecture record: one page per phase with a table of facts, each tagged OBSERVED (cmdlet, API or UI and the date) or ASSUMED (needs a test), plus an open-questions list for the deployer. Add the Holodeck module version, the VCF release and the BOM comparison. This is the handoff document and your VCDX design-validation evidence.
Validation Gate
Check: vCenter hierarchy, cluster services, vDS and VLAN mapping, NSX cluster, fabric, segments and Tier-0 BGP recorded; builds compared with the BOM; architecture record complete.
Expected: All items present in the lab notebook: architecture record complete.
Common Errors
Final Validation
An inherited Holodeck 9.0.2 instance is fully characterised across four phases, from the HoloRouter network truth through the Holodeck runtime, the SDDC Manager and VCF Operations APIs, to the vCenter hierarchy and the NSX fabric, with a handoff record whose every fact carries an evidence tag.
✓ Phase 1 → Subnets, appliances, DNS, BGP and pools recorded; default or custom plan identified
✓ Phase 2 → Live config and instance identified; options, Day 2 operations, depot and licence window reconstructed
✓ Phase 3 → Token flow, domains, hosts, credentials (masked), tasks, version; API Explorer used
✓ Phase 4 → vCenter tree, cluster services, NSX cluster, fabric and Tier-0 state; BOM comparison
✓ Record → Architecture record with OBSERVED/ASSUMED tags and open questions
Cleanup / Restore
Snapshot: holodeck-08-complete
• Delete nsx.cookie and nsx.hdr and clear TOKEN and SID from the shell
• Remove any credential values from the notebook; keep names only
• Hand the architecture record to the environment owner
Design Reflection (VCDX)
Discovery is the skill behind every VCDX design review question that starts with 'how do you know': you must be able to say where a fact was observed and what you assumed. The four-phase order matters because each layer validates the one above it: the HoloRouter's intended plan, the Holodeck runtime's build options, SDDC Manager's inventory and vCenter and NSX's actual configuration. Drift between layers is exactly what a reviewer probes.
Requirements
- Complete inventory of an inherited instance
- Evidence-tagged architecture record
- No modification of the discovered environment
Constraints
- Read-only: no config edits, no API writes
- Credential values never recorded
- API Explorer location differs between 9.0 (VCF Operations) and 5.2 (SDDC Manager)
Assumptions
- Documented default credentials or a handover list
- Management domain healthy
- HoloRouter SSH available
Risks
- Trusting the documentation over the instance — Impact: wrong plan recorded — Mitigation: cmdlet output wins
- Credential leakage into notes — Impact: lab compromise — Mitigation: names only, masked values
- Stale tokens mid-discovery — Impact: partial data — Mitigation: re-authenticate per phase
Self-Assessment Discussion Prompts
- Which layer would you trust when the HoloRouter subnet table and the vDS port group VLANs disagree, and why?
- What in the SDDC Manager task history tells you about the environment's operational maturity?
- How would you adapt the playbook for a VCF 5.2.2 instance built with Cloud Builder?
- Which of your ASSUMED facts would you test first before presenting this architecture?
Extensions
Scripted discovery
Turn Phases 1-3 into a pwsh script on the HoloRouter that writes a JSON architecture record; run it against the dual-site pod from holodeck-06.
harderWorkload domain discovery
Repeat Phase 4 against vc-wld01-a and nsx-wld01-a on a full-stack instance and record the shared-versus-dedicated NSX decision.
same5.2.2 variant
Run the playbook against the Pod 2 instance from holodeck-07 and record the API and UI differences.
sameDrift report
Deliberately add a nested host with New-HoloDeckESXiNodes without commissioning it and show how each phase reveals (or misses) it.
same⚠ Known Pitfalls
References
- Holodeck 9.0.2 documentation (index): Environment Overview, Holorouter Overview, Concepts, Pre-requisites, Deployment, During Deployment, Post Deployment, Day 2 Operations, How ToTier 1 — Official
Primary source for every HoloRouter and toolkit step in this lab. Fetched 1 Oct 2026. - Holodeck 9.0.2 command referenceTier 1 — Official
New-HoloDeckInstance, Update-HoloDeckInstance, Get-HoloDeckInstance, New/Get/Import-HoloDeckConfig, Reset-HoloDeckState, New-HoloDeckNetworkConfig, Set-/Reset-HoloRouter, Get-HoloDeckSubnet/AppNetwork/BGPConfig/DNSConfig/OverlaySubnet/AppIpPools, Get-HolodeckServiceIPPools. - Holodeck 9.0.2 FAQ and troubleshootingTier 1 — Official
Pre-check errors (binary not found, no trunk port group, NTP not running), ESX build SSH wait, memory tiering, nested vSAN ESA. - Holodeck 9.0.2 downloads and release notes (known issues)Tier 1 — Official
HoloRouter OVA and Offline Depot OVA; release notes at /release_notes/ carry the Known Issues per release. - Holodeck GitHub repository: issues and discussionsTier 2 — Supporting source
Where to report bugs; community troubleshooting. The releases page carries no binaries (OVA downloads are on the Broadcom Support Portal). - VMware Cloud Foundation 9.0 documentation: SDDC Manager API (tokens, credentials, restore tasks), API Explorer in VCF Operations, NSX Manager cluster status API, VCF 9.0 BOMTier 1 — Official
VCF 9.0.1 PDF p.3012 (POST /v1/tokens with admin@local; Bearer token), pp.3014-3015 (GET /v1/credentials?resourceType=VCENTER|PSC|ESXI), p.1513 (VCF Operations > Developer Center > APIs & SDKs > API Explorer; GET /v1/hosts with status UNASSIGNED_USEABLE), p.2538 (NSX /api/session/create), p.2767 (GET /api/v1/cluster/status, /api/v1/cluster/nodes), pp.79 and 224 (VCF 9.0 BOM: vCenter 9.0 build 24755230, ESX 24755229, NSX 9.0.0.0 build 24733065). - vSphere Automation API (REST): session and vcenter inventory endpointsTier 1 — Official
POST /api/session returns the session ID used in the vmware-api-session-id header; GET /api/vcenter/{datacenter,cluster,host,vm,datastore,network}. Task 4.