A platform team has a hardened Windows Server VM that was generalized after patching, agent installation, and baseline configuration. They must deploy the same build to development, test, and production subscriptions, and they want a controlled way to publish newer versions later without rebuilding the image each time. What should they create first?

An availability set for the source VM so new virtual machines inherit the same configuration.

An availability set improves fault and update domain placement, but it does not package an operating system build for reuse. It cannot be used to stamp out identical VMs across subscriptions or version images over time.

A managed disk snapshot of the OS disk so new VMs can be created from the captured state.

A snapshot captures a disk point in time, but it is not the best choice for a governed, versioned image lifecycle. It is also focused on disk recovery rather than standardized multi-subscription VM deployment.

A proximity placement group so all future VMs land close to the current server.

A proximity placement group reduces latency by keeping resources physically close, but it does not define a reusable VM image. It addresses placement, not image management or standardized server builds.

What does this AZ-104 question test?

Static NAT maps one inside address to one outside address.

What is the correct answer to this question?

The correct answer is: An Azure Compute Gallery image version based on the generalized VM image. — Azure Compute Gallery is designed for reusable, managed image distribution. Because the source VM is already generalized, the team can capture it as a gallery image version and then deploy that version to multiple subscriptions and regions. The gallery also lets administrators publish new versions later without recreating the entire image process from scratch. That combination of consistency, versioning, and broad reuse is what the scenario requires. Why others are wrong: An availability set only provides fault-domain and update-domain separation. A snapshot is useful for recovery or ad hoc cloning, but it is not a managed distribution mechanism. A proximity placement group is for reducing network latency, not for image lifecycle management. None of those options provides the controlled, reusable image publishing workflow requested.

What should I do if I get this AZ-104 question wrong?

Then try more questions from the same exam bank and focus on understanding why the wrong options are tempting.