# Free 2V0-13.25 practice questions

> 10 free 2V0-13.25 practice questions with the correct answer and an explanation.

Page: https://devopscodelab.com/exams/vmware-2v0-13-25/free-practice-questions/
Practice exams: https://devopscodelab.com/exams/vmware-2v0-13-25/

### Question 1 of 10

An architect identifies a risk: the planned vSAN stretched cluster witness will run in a cloud region whose round-trip latency to both data sites has not been measured. Which mitigation strategy is MOST appropriate to document?

- A. Accept the risk, because the vSAN witness only stores metadata and is not affected by network latency between the sites
- B. Reduce the risk by scheduling a latency test before design sign-off and defining an alternative witness site if limits are exceeded
- C. Transfer the risk by moving the witness responsibility to the cloud provider under its standard infrastructure service agreement
- D. Avoid the risk by replacing the stretched cluster design with two independent clusters and no protection for the workloads

**Answer:** B

**Explanation:** Measuring the latency early and preparing a fallback witness location reduces both the probability and the impact of the risk while keeping the design goal. The witness is sensitive to latency and must meet published round-trip limits, so accepting the risk on that basis is wrong. A provider's generic service agreement does not take ownership of vSAN witness latency, and dropping protection entirely avoids the risk only by failing the availability requirement.

### Question 2 of 10

An architect writes: "The DNS team will create forward and reverse records for all VCF components before the deployment window opens." Nobody from the DNS team has confirmed this yet. Which action BEST handles this item in the design?

- A. Record it as an assumption, link it to a risk of deployment delay, and assign an owner to validate it before the window opens
- B. Record it as a constraint and design the deployment to tolerate missing DNS records by using IP addresses for every component
- C. Record it as a requirement owned by the architect and remove it once the VCF Installer validation passes during the deployment
- D. Record it as a design decision with a justification, because the architect has chosen to rely on the DNS team for name resolution

**Answer:** A

**Explanation:** An unconfirmed belief that the design depends on is an assumption, and good practice is to tie it to the risk it creates and assign someone to validate it early. VCF components require valid forward and reverse DNS, so designing around IP addresses is not a valid workaround. It is not a requirement the architect owns, and it is not a design decision because it describes another team's expected action rather than a choice between alternatives.

### Question 3 of 10

A hospital group has just refreshed its vSphere 8 hosts with vSAN and the infrastructure team refuses to rebuild them, yet the CIO wants that environment to become the base of a VCF 9.0 deployment with VCF lifecycle management. What should the architect propose?

- A. Rebuild the hosts with the VCF Installer, because VCF 9.0 can only deploy greenfield management domains
- B. Migrate every VM to a newly deployed greenfield VCF Instance using HCX bulk migration, then decommission the vSphere 8 hosts and the old vCenter
- C. Use the VCF Installer to converge the existing vCenter environment into a VCF management domain, after upgrading it to a supported version
- D. Deploy SDDC Manager as a standalone appliance beside the existing vCenter and register each host with it manually

**Answer:** C

**Explanation:** VCF 9.0 can converge existing vSphere infrastructure: the VCF Installer takes an existing vCenter environment that meets version and configuration prerequisites and turns it into a management domain, deploying the remaining components. VCF 9.0 is not limited to greenfield deployments. A standalone SDDC Manager with manually registered hosts is not a supported way to bring infrastructure under VCF, and a full HCX migration is possible but is not required by the requirement to avoid redeployment.

### Question 4 of 10

A design requires the NSX host overlay network to carry Geneve-encapsulated traffic between ESX hosts. Which physical network setting should the architect specify for the host overlay VLAN?

- A. An MTU of 1500 bytes end to end, because Geneve fragments packets automatically when they exceed the path MTU
- B. An MTU of 1600 bytes on the ToR switches and 1500 bytes on the hosts, so that the switches absorb the encapsulation
- C. An MTU of at least 1700 bytes end to end, with 9000 bytes recommended, on every switch carrying the overlay VLAN
- D. An MTU of 9000 bytes on the ESX host uplinks only, because the physical switches forward frames regardless of size

**Answer:** C

**Explanation:** Geneve adds an encapsulation header, so the overlay path needs a larger MTU end to end; NSX guidance is at least 1700 bytes with 9000 bytes recommended for VCF designs. Overlay traffic is not meant to be fragmented, so 1500 bytes causes drops. Setting jumbo frames only on host uplinks fails because switches drop oversized frames, and the hosts themselves must also use the larger MTU on their TEP interfaces.

### Question 5 of 10

A customer requires that vSAN traffic in a workload domain is never affected by vMotion bursts, but only two 25 GbE uplinks exist per host. Which physical design decision BEST meets this requirement?

- A. Configure a dedicated physical uplink for vMotion only, leaving vSAN to share the second uplink with all other traffic
- B. Disable vMotion on the cluster during business hours, and allow migrations only during the weekly maintenance window
- C. Enable Network I/O Control on the vSphere Distributed Switch with higher shares for vSAN traffic than for vMotion
- D. Place vSAN and vMotion on the same port group with an explicit failover order using both uplinks as active

**Answer:** C

**Explanation:** Network I/O Control allocates bandwidth by traffic type during contention, so giving vSAN higher shares protects it from vMotion bursts while both uplinks stay available for redundancy. A shared port group gives no prioritization, dedicating one uplink to vMotion removes redundancy for it and still lets vSAN contend with other traffic, and disabling vMotion hurts manageability and availability.

### Question 6 of 10

A logical design must decide where the fleet-level VCF Operations and VCF Automation appliances run in a fleet with two VCF Instances. Which placement is the standard design?

- A. In the management domain of the first VCF Instance, next to its other management appliances
- B. On physical servers outside VCF, so that their availability is independent of either management domain
- C. Split between the two instances, with VCF Operations in the first and VCF Automation in the second
- D. In a dedicated VI workload domain of the second VCF Instance, so that the first instance keeps more capacity

**Answer:** A

**Explanation:** Fleet-level management components are deployed in the management domain of the first VCF Instance, where they are protected and lifecycled with the other management components. Placing them in a workload domain mixes management and tenant resources, splitting them across instances adds dependencies without a stated benefit, and physical servers outside VCF are not a supported deployment target for these appliances.

### Question 7 of 10

A stretched VI workload domain cluster spans two availability zones. The customer wants north-south routing to continue if one zone fails. Which NSX Edge design meets this requirement?

- A. Place both Edge nodes in zone A, but peer them with the ToR switches of both zones over uplink VLANs stretched to zone B
- B. Deploy Edge nodes so that both zones host Edges that can carry traffic, with BGP peering to the ToR switches in each zone
- C. Place both Edge nodes in availability zone B so that they are isolated from the busier preferred site
- D. Place both Edge nodes in availability zone A, the preferred site, and rely on vSphere HA to restart them in zone B

**Answer:** B

**Explanation:** Placing Edge nodes in both zones, each with BGP peering to its local ToR switches, lets the surviving zone keep routing immediately after a zone failure. Keeping all Edges in zone A means losing that zone stops routing until vSphere HA restarts them elsewhere, and peering them with zone B switches does not help when the Edge VMs themselves are down. Isolating the Edges in zone B simply moves the same single-zone dependency to the other site.

### Question 8 of 10

A security requirement states that administrators of the private cloud must sign in once with their corporate identity provider credentials and multi-factor authentication to reach vCenter, NSX and VCF Operations. Which VCF 9.0 design decision supports this?

- A. Enable lockdown mode on all ESX hosts so that administrators can authenticate only through the vCenter instances
- B. Configure local SSO users on each vCenter instance and synchronize their passwords with a scheduled script
- C. Deploy VCF Identity Broker and federate it with the corporate identity provider for single sign-on across components
- D. Join every ESX host to Active Directory and grant administrators direct root access through AD group membership

**Answer:** C

**Explanation:** VCF Identity Broker federates VCF management components with an external identity provider, giving administrators single sign-on with the provider's multi-factor authentication. Local SSO users with synchronized passwords bypass the corporate identity provider and its MFA. Joining hosts to Active Directory does not give single sign-on to NSX or VCF Operations, and lockdown mode hardens host access but does not provide federated sign-in.

### Question 9 of 10

A security requirement says east-west traffic between application tiers must be filtered even when the VMs share the same subnet. Which design decision meets this requirement?

- A. Use the NSX Distributed Firewall with rules based on security groups for each application tier
- B. Configure the Tier-0 gateway firewall with rules between the IP addresses of each application tier
- C. Place each tier on a separate VLAN and filter traffic between VLANs on the physical core firewall
- D. Enable vSphere VM encryption on the application VMs so that intercepted traffic cannot be read

**Answer:** A

**Explanation:** The NSX Distributed Firewall enforces rules at each VM's virtual NIC, so it filters traffic even between VMs on the same subnet, and group-based rules follow the workloads. Separate VLANs with a core firewall require readdressing and do not filter traffic within one subnet, the Tier-0 gateway firewall only sees traffic that is routed through it, and VM encryption protects data at rest rather than filtering traffic.

### Question 10 of 10

An architect designs an HCX deployment to migrate workloads from a legacy vSphere site into VCF. Which appliances are deployed as part of an HCX service mesh to migrate VMs and keep their IP addresses?

- A. The NSX Edge cluster for migration traffic and the NSX Manager cluster for layer-2 extension between sites
- B. The HCX Interconnect appliance for migration traffic and the HCX Network Extension appliance for layer-2 extension
- C. The vSphere Replication appliance for migration traffic and a VPN gateway VM for extending layer-2 networks
- D. The VCF Installer appliance for migration traffic and SDDC Manager for creating destination network segments

**Answer:** B

**Explanation:** An HCX service mesh deploys the Interconnect appliance for migration and replication traffic and the Network Extension appliance to stretch layer-2 networks so that VMs keep their IP addresses. NSX Edges and Managers are not the HCX data path, the standalone vSphere Replication appliance and VPN VMs are not HCX service mesh components, and the VCF Installer and SDDC Manager do not migrate workloads.

Ready for a timed run? Take the free 15-question 2V0-13.25 mini exam: https://devopscodelab.com/app/exams/vmware-2v0-13-25/mini/

DevOpsCode Lab is not affiliated with VMware. 2V0-13.25 and related names are trademarks of their owners. We write every question ourselves; these are not real exam questions.
