1

I'm discussing with my coworkers when it is appropriate to change a computer name in Active Directory. Our current naming structure is along the lines of . As computers are rebuilt they are either renamed or have an inaccurate departmental listing.

My thinking is that a workstation ID should be like a server ID - once created it never changes and whatever human needed info is stored in the description / extended AD attributes. This would assist in tracking long lived problems in event logs, trending, etc.

Is there anything along the line of a best practice in this regard? I spent a great deal of time researching this and found little more than instructions on changing names and resolving AD membership issues. I would love some official documentation to mull over.

2
  • 2
    There is no "right" way to answer this question, as every organization has its own standards regarding workstation (and server) names.
    – Massimo
    Jan 12, 2012 at 20:37
  • Agreed; there are generally best practices however. I'm interested in if such documentation exists in this case. Jan 12, 2012 at 20:43

2 Answers 2

6

I have found that anything meaningful you put in the name, at some point, becomes a misnomer.

So over time my and my colleagues developed our own "best practices".

These days we just stick with PC-xxxxxx (where xxxxxx is a 6-digit number, with leading zero's) for any PC which is not a server.

SW-xxxx for Windows servers, SU-xxxxx for Unix servers. SS-xxxxxx for storage devices which covers NAS, SAN, disk-packs and tape-libraries.

L2-xxxxx for any layer 2 switch. L3-xxxxx for Layer 3 switches and routers.

AP-xxxxx for Access points and NW-xxxxx for any other network device which is part of the infra-structure (think WLAN controller, Leased line modem, etc.)

Printers/multifunctionals get PR-xxxxxx. MO-xxxxx is for monitors (above $300 purchase value we need to track those)

Anything not convered by the above is simply not labeled if the purchase value was below $300 (we consider it an expendable) or is labeled OS-xxxxxx (other stuff) if it was more expensive or if it is a networked device. (Regardless what it is. If it lives on the LAN the ID-tag and the mac-address get stored in our tracking database, together with the owner of the device, so we know who to hit over the head if it is causing problems.)

The device ID is printed on 2 stickers with the name in plain-text and a simple (full ascii) Code39 barcode with the same value. 1 sticker goes on the front the device. The other somewhere out of sight on the bottom. People have a tendency to peel-off stickers (they find them an eye-sore), especially on laptops and monitors, so we put the second one "out-of-sight". In case of laptops under the battery is a nice spot.

We are currently considering switching to QR-codes for the barcode as we can put more info in those (purchase date, end of warranty, mac-addresses) and they can be easily read with a smart-phone.

1

I would say that the naming scheme should be sufficiently generic that change of use or ownership should not affect the accuracy of it's name.

My former employer had a simple convention of ITS followed by an incremental 7 digit number.

Keep it generic and you will find that you never need to rename a workstation again.

Not the answer you're looking for? Browse other questions tagged .