File Server designs/templates and auditing

1) Do you have a standard file server build for file servers in your environment? Is that consistent for all servers serving a specific role? Our IT have published a file server design document, which to me indicates that is going to be a standard for all new file servers utilised. Is this common?

2) Is the file server, before actual user files are added to do, referred to anything in particular, i.e. it’s basically the bare bones of the config but with no data as yet.
3) Will a standard file server build that’s intended to be the benchmark for all future file servers have a specific “name” in windows environments?

4) And will the file server in design phase typically be built well in advance of data being added to it and it going “live”. We want to audit the config of the template copy, but have no real idea how long before IT sections will build it, prior to it going live.

5) Are there any issues in auditing the config of the server in a test environment before it goes live, i.e. configs that cant be applied to a server in a non live environment that will only be applied once it goes into a live domain? I.e. is it worth auditing a file server in build phase in a non live environment, or better to wait until it goes into the live domain?
Who is Participating?
DavidConnect With a Mentor PresidentCommented:
No, no, no, no, and YES!!

You need to realize that a "file server" has matured from being a dedicated piece of equipment to a service that is run locally, remotely, and/or both at the same time on any number of similar or dissimilar physical or even virtual computers or even appliances.

Virtualization, clustering, appliances, and clouds have changed the dynamics such that this entire survey is about 10 years obsolete.
Rich RumbleConnect With a Mentor Security SamuraiCommented:
We, and most others, use Server Images, VM Images are very popular. Our users get a standard image, our GC's, our DNS, our FileServers all have a standard image for each. After they are in place (imaged) you go through Sysprep, naming the box, joining the domain, and some other settings.
The basic steps are, build a server, it can be generic, or specific, like a DNS server. Once it's setup, clean up what you can, delete temp files, unjoin from the domain if joined, defrag, then run sysprep and power down. Image the server using Ghost, or copy the VM's file. When you need a new VM, copy the file to where you need it, power on, go through sysprep prompts.
I don't know why I could do something in a lab or test environment I couldn't in the live side.
Our servers performance degrades when we use any AV product with real-time scanning, so the "server" images get AV but it's a scheduled scan that runs "after hours" or during really low traffic times. In the test environment we could run AV in real-time, but in the live we can't.
I should clarify our server builds, to save time, we have a very generic FULL server build, and then we use add/remove programs and other utilities to strip out whats not needed for each role, then after the server is built for it's images purpose, we run sysprep and take an image or copy the vm file. This will save you 1-2 hours of putting in the CD/DVD and building through the windows prompts and wizards (formatting the drive, naming, joining the domain, time zone, network settings, display settings, services...)
All Courses

From novice to tech pro — start learning today.