1Y0-A15 · Question #1
Scenario: Nether Tech Inc. is planning to implement Provisioning Services for deployment of images to their XenApp servers and XenDesktop workstations. The architect has created the design for the…
The correct answer is B. TCP Large Send Offload will be enabled on the NICs. See the full explanation below for the reasoning.
Question
Scenario: Nether Tech Inc. is planning to implement Provisioning Services for deployment of images to their XenApp servers and XenDesktop workstations. The architect has created the design for the Provisioning Services environment, making the following design decisions for the Provisioning Services environment:
One Provisioning Services farm will be created: PVSFARM. Two Provisioning Services servers will be installed and configured for High Availability: PVS01 and PVS02. One site will be created: Nether Tech Inc. Two Device Collections will be configured:
XenApp and XenDesktop. A central store will be defined and located on the SAN; all vDisks will be added to this store. The local hard drive of the XenApp servers will be used to store the write cache. TCP Large Send Offload will be enabled on the NICs. A specified user account will be used as Provisioning Services service account. Each Provisioning Services server will have four NICs; two teamed and connected with the "General use subnet" (per the design document) and the other two NICs, disabled. An Active Directory group policy object (GPO) will be configured and linked to the XenApp and XenDesktop organizational units to disable Active Directory Machine Account Password Management. Online and offline applications will be used to minimize changes to the vDisk. Offline applications will be pre-deployed to desktops and XenApp servers for faster access to the applications by users. According to the design decisions from the architect, which decision does NOT follow best practices?
Options
- AOne site will be created.
- BTCP Large Send Offload will be enabled on the NICs.
- CThe local hard drive of theXenApp servers is used to store the write cache.
- DOnline and offline applications will be used to minimize changes to thevDisk.
- EOffline applications will be pre-deployed to desktops andXenApp servers for faster access to the
How the community answered
(40 responses)- A3% (1)
- B83% (33)
- D5% (2)
- E10% (4)
Community Discussion
6
Mateus R.Feb 19, 2026TCP Large Send Offload sounds like a good idea on paper, the same way putting a turbocharger on a garden hose sounds like it would move more water. In networking, LSO lets the CPU hand off the job of splitting large data chunks into smaller packets to the NIC itself. The problem is that Provisioning Services relies on a very precise, timing-sensitive stream of data to boot target devices over the network, and LSO can scramble that timing by reordering or delaying packets in ways that confuse the PVS streaming protocol. Citrix best practice documentation explicitly calls out TCP Large Send Offload as something you should disable on PVS server NICs, not enable, so option B is your answer here.
Mei-Ling H.Feb 28, 2026The answer is B. Citrix best practice for Provisioning Services specifically calls for TCP Large Send Offload to be DISABLED on the NICs, not enabled. LSO can cause packet segmentation issues with the PVS streaming protocol, leading to instability and dropped connections during vDisk streaming. The word "enabled" in option B is exactly the flag to catch here, because the direction is reversed from what the best practice guide recommends. The other options, local write cache on the hard drive, a single site, offline apps pre-deployed, and using online plus offline apps to limit vDisk changes, all align with standard PVS deployment guidance.
Giselle K.Mar 10, 2026TCP Large Send Offload causes streaming issues, so B is right, though C tempts people who forget XenApp local cache is actually fine.
Brenda K.Feb 11, 2026Clock check first: this one looks like a flag-and-revisit on the surface because of all the scenario text, but once you strip the noise it is actually a quick win if you know your NIC settings cold. The trap is that TCP Large Send Offload sounds like a performance boost, and your instinct is to say "more throughput, good" but PVS streaming is extremely sensitive to NIC offloading features because they can cause packet fragmentation and corruption on the streaming network, which is why Citrix explicitly calls out disabling LSO as a required step in PVS NIC hardening. Options C and D are solid practices, local write cache on the XenApp server disk is a supported and often recommended placement, and layering online plus offline apps is the right way to keep vDisks stable. My question for the group: does anyone know offhand whether the guidance to disable LSO applies equally to both the streaming NIC and the management NIC, or only to the NIC carrying the PVS streaming traffic? Getting that right in the lab is where I see people lose points on scenario follow-ups.
Ingrid P.Feb 3, 2026One site per farm is the design anti-pattern here, splits HA incorrectly.
Mateus R.Feb 4, 2026Ingrid, I hear you but B is actually the right call here, because grouping all nodes from the same physical location into one site is exactly how the cluster knows to keep quorum and workloads local during a network partition. One site per farm would scatter the HA logic and leave the cluster blind to which machines are physically close together, which is the problem B solves.