VCP-VVF Support · Exam 2V0-18.25 · VMware vSphere Foundation 9.0

Every testable objective in exam 2V0-18.25 is troubleshooting. Read its blueprint and Sections 1 to 4 are marked as having no testable objectives. Everything sits in Section 5, from a failed VVF deployment to a workflow that will not run. The best preparation is breaking things on purpose and finding out why, so that is what these 100 lab tasks ask you to do.

FirstWhat this exam actually is

Facts from the published exam guide, last updated 15 August 2025:

Source: the official exam guide for 2V0-18.25, published by Broadcom. Read it yourself before you book. Everything below is written against the objectives in that document.

SecondWhere the objectives actually are

VCF Operations is the largest single objective, with six sub-objectives, and compute and clusters follow with four each. Orchestrator is unique to this exam. No other exam in the VCF track asks you to troubleshoot it, so section M is new ground even if you have prepared for another Support exam.

ThirdHow it differs from the VCF Support exam

The VCF Support exam, 2V0-15.25, has the same shape: everything in Section 5, all troubleshooting. Much of the compute and storage content is the same, so this workbook shares those tasks with it. What VVF leaves out is NSX, SDDC Manager, workload domains, certificates and passwords as fleet management, VCF Automation and HCX. What it adds is cluster-level work done directly in vCenter, standard switches alongside distributed ones, and Orchestrator. If you are preparing for the matching Administrator exam too, the VCP-VVF Administrator lab tasks build the environment these faults are created in.

FourthHow to practise a troubleshooting exam

• Create the fault, then diagnose it as if you had not. A fault you built is one where you already know the answer, so you can check your method against it.
• Practise in a lab, never at work. A nested vSphere lab with vSAN covers most of sections D to J. VMware Hands-on Labs help with VCF Operations and the installer. Check each lab runs version 9.0.
• Where you cannot break it, write the checklist. Some tasks, such as a Fibre Channel fault, ask you to write the order you would check things in. That order is what the exam tests.
• Finish with faults somebody else created. Section N is timed, and works best when a colleague breaks things without telling you how.

Workbook100 tasks, in working order

Section A builds a method and a toolkit. Sections B to E cover deployment, conversion, clusters and licences. Sections F to J cover compute, storage and networking. Sections K to M cover VCF Operations, VCF Operations for logs and Orchestrator. Section N is a timed run. Numbers in grey at the right are the objectives.

Before any fault: an order to work in, and the tool for each question.

1. 1Write a troubleshooting method in six steps: scope, when it started, what changed, which layer, evidence, fix.5.8›
2. 2List the tools VVF 9.0 gives you and the question each answers: VCF Operations, VCF Operations for logs, the workbench, vCenter and the ESX host client.5.8›
3. 3Find the logs for vCenter and ESX, and how to reach each.5.5›
4. 4Write the support bundle options for vCenter, ESX and VCF Operations.5.8›
5. 5Write when you would open a case with Broadcom, and what you gather first.5.8›
6. 6Set up a lab where you can break things safely, and record what you may not break at work.5.8›

Objective 5.1. Most failed deployments are a prerequisite somebody missed.

1. 7Remove a reverse DNS record for one appliance and run installer validation. Read the error.5.1›
2. 8Set one host to a different NTP source and run validation. Read the error.5.1›
3. 9Set a VLAN or MTU wrongly in the deployment inputs and run validation. Read the error.5.1›
4. 10Use a password that breaks the complexity rules and record where the installer stops you.5.1›
5. 11Find the installer logs, and the first error in a failed run.5.1›
6. 12Write how you retry a failed deployment step rather than starting again.5.1›

Objective 5.2. Bringing an existing estate into VVF 9.0.

1. 13Write the prerequisites for converting an existing vSphere environment to VVF 9.0.5.2›
2. 14Run the conversion precheck in a lab against an environment that fails one prerequisite, and fix it.5.2›
3. 15Find the logs for a failed conversion and the first error.5.2›
4. 16Write what to check after a conversion to prove it worked.5.2›

Objective 5.3. Create clusters, add and remove clusters and hosts, import.

1. 17Create a cluster whose hosts do not match the cluster image, read the error, and fix it.5.3›
2. 18Turn on vSAN on a new cluster with the vSAN network missing on one host, and read the result.5.3›
3. 19Add a new cluster and find why VCF Operations shows it with no data or unlicensed.5.3›
4. 20Remove a cluster and record what must be moved or cleaned up first.5.3›
5. 21Add a host with a different ESX build or vendor add-on, read the error, and fix it.5.3›
6. 22Remove a host from a cluster, and record what blocks it when vSAN cannot evacuate the data.5.3›
7. 23Add a host whose vMotion or vSAN network is wrong, and find the fault.5.3›
8. 24Write the checks before importing an existing vSphere environment as a cluster, and the usual failures.5.3›

Objective 5.4. Assignment, entitlements and expiry.

1. 25Assign a licence to the wrong component and read the result.5.4›
2. 26Try to assign more capacity than the licence entitles, and read the error.5.4›
3. 27Read the entitlement view and explain a component shown as unlicensed.5.4›
4. 28Write what the 9.0 documentation says happens when a VVF licence expires.5.4›
5. 29Write how you find which licence is about to expire and who is told.5.4›
6. 30Write the steps to replace an expired licence.5.4›

Objective 5.5. ESX hosts, vCenter, virtual machines and clusters.

1. 31Disconnect a host from vCenter and reconnect it, recording the cause you created.5.5›
2. 32Make a host not responding by stopping its management agents, then restart them.5.5›
3. 33Fill a host scratch or log location and read the symptoms.5.5›
4. 34Break a host time setting and read what fails.5.5›
5. 35Stop a vCenter service and find which feature stopped working.5.5›
6. 36Read vCenter service health from its management interface and restart a failed service.5.5›
7. 37Write what to check when no one can sign in to vCenter.5.5›
8. 38Write how to tell a vCenter database or disk full problem from a service problem.5.5›
9. 39Make a virtual machine fail to power on through lack of reserved capacity, and fix it.5.5›
10. 40Make a virtual machine lose its network by disconnecting the adapter or port group, and fix it.5.5›
11. 41Create a slow virtual machine through a CPU limit, find it, and remove the limit.5.5›
12. 42Fail a vMotion on purpose, for example with a missing port group on the target, and read the error.5.5›
13. 43Make HA admission control block a power-on and explain it.5.5›
14. 44Make DRS fail to balance through an affinity rule, and find the rule.5.5›
15. 45Create an EVC mismatch when adding a host, and fix it.5.5›
16. 46Write how you find what changed on a cluster in the last day.5.5›

Objective 5.6, part one. Create each fault, then diagnose it.

1. 47Break the vSAN network on one host and read the health result and partition.5.6›
2. 48Fail a device and watch the rebuild, recording the delay before it starts.5.6›
3. 49Fill vSAN close to full and read what stops working first.5.6›
4. 50Apply a policy the cluster cannot satisfy and read the compliance error.5.6›
5. 51Find an inaccessible object and explain why.5.6›
6. 52Read the resyncing objects view and explain why data is moving.5.6›
7. 53Run vSAN health checks and work through every failure.5.6›
8. 54Write the order you check a vSAN problem in: network, devices, capacity, policy.5.6›

Objective 5.6, part two.

1. 55Cut the witness off in a lab and read what still works.5.6›
2. 56Write the latency and bandwidth limits between sites and to the witness, from the documentation.5.6›
3. 57Fail one site in a lab and record what restarts and where.5.6›
4. 58Apply a policy without site mirroring to a VM in a stretched cluster and record the risk.5.6›
5. 59Write what breaks when the networks are not present in both sites.5.6›
6. 60Write the checks before adding hosts to a stretched cluster.5.6›

Objective 5.6, part three.

1. 61Break iSCSI port binding or a target address, read the error, and fix it.5.6›
2. 62Remove a host from an NFS export permission and read the mount error.5.6›
3. 63Write how an FC zoning or masking error shows up on the host.5.6›
4. 64Write the difference between all paths down and permanent device loss.5.6›
5. 65Read the storage paths for a device and find a dead path.5.6›
6. 66Write what to collect for the storage team when a datastore disappears.5.6›

Objective 5.7. In VVF the network stops at the switch: there is no NSX.

1. 67Mismatch the MTU on one host and prove it with vmkping.5.7›
2. 68Put an uplink on the wrong VLAN and find it.5.7›
3. 69Read the distributed switch health check and fix a finding.5.7›
4. 70Rename a standard switch port group on one host and record what breaks for vMotion and HA.5.7›
5. 71Lose host management while migrating a VMkernel adapter to a distributed switch, and record how rollback saves it.5.7›
6. 72Block traffic with a standard switch security setting, such as MAC address changes, and find it.5.7›
7. 73Set a wrong failover order on a standard switch and find why traffic fails when one uplink goes down.5.7›

Objective 5.8, part one: when the monitoring itself is broken.

1. 74Stop a collector and find the objects with no data.5.8›
2. 75Break an adapter credential and read the collection error.5.8›
3. 76Read VCF Operations cluster status and a node that is down.5.8›
4. 77Break log forwarding from a host and find the gap in VCF Operations for logs.5.8›
5. 78Use VCF Operations for logs to find the first error before an incident.5.8›
6. 79Use VCF Operations to find a component in a bad state.5.8›

Objective 5.8, part two: the workbench, Log Assist, and your own dashboards, reports and alerts.

1. 80Open the troubleshooting workbench for a slow virtual machine and read its findings.5.8›
2. 81Use the workbench to find a change that happened just before an incident.5.8›
3. 82Generate a log bundle for a component from VCF Operations.5.8›
4. 83Upload a log bundle to a support case with Log Assist, or record each step if you have no case.5.8›
5. 84Create a dashboard that shows the health of one problem area.5.8›
6. 85Create a report that proves an issue is fixed.5.8›
7. 86Create an alert that would have caught an issue earlier.5.8›
8. 87Write which of the three you would build first in a live incident.5.8›

Objective 5.9. Configuration faults, then workflow faults.

1. 88Break the vCenter connection in Orchestrator, for example with a wrong credential, and read the error.5.9›
2. 89Make Orchestrator distrust a certificate it needs, and fix the trust.5.9›
3. 90Break the authentication setup and record what users see when they sign in.5.9›
4. 91Run a workflow with a wrong input and read the failed run, its logs and variables.5.9›
5. 92Find a workflow run that is stuck or waiting and explain its state.5.9›
6. 93Write the checks when an action works by hand but fails when called from a workflow.5.9›

Do this section twice in the last week, with faults somebody else created.

1. 94A host shows not responding. Find the cause. 10 minutes.5.5›
2. 95A virtual machine will not power on. Find the cause. 10 minutes.5.5›
3. 96A vSAN object is inaccessible. Find the cause. 15 minutes.5.6›
4. 97A datastore has disappeared from one host. Find the cause. 10 minutes.5.6›
5. 98vMotion fails between two hosts. Find the network cause. 10 minutes.5.7›
6. 99VCF Operations shows no data for a cluster. Find the cause. 10 minutes.5.8›
7. 100An Orchestrator workflow fails. Find the cause. 10 minutes.5.9›

How to use itFour weeks, if that is what you have

Exam details are from the published Broadcom exam guide for 2V0-18.25, last updated 15 August 2025, and were checked on 5 October 2026. Broadcom revises blueprints regularly, so confirm the current version before you book. Tasks here are my own, written from the published objectives; they are not exam content.