Help settle a dispute

prelag

Weaksauce
Joined
Dec 1, 2007
Messages
77
The employer I work for has about 100 employees.

My co-worker wants me to label (physically, with a label maker) and rename every PC on the network because they have trouble figuring out what machine belongs to which IP in DHCP.

My suggested solution was to add the name of the employee as a description next to the machine name inside of active directory.

Am I in the wrong here? With the amount of turnover/movement we see, labeling and keeping a naming convention for machines out on the floor seems like a waste of time.

Thoughts?
 
Last edited:
The employer I work for has about 100 employees.

My co-worker wants me to label (physically, with a label maker) and rename every PC on the network because they have trouble figuring out what machine belongs to which IP in DHCP.

My suggested solution was to add the name of the employee as a description next to the machine name inside of active directory.

Am I in the wrong here? With the amount of turnover/movement we seem, labeling and keeping a naming convention for machines out of the floor seems like a waste of time.

Thoughts?

Yeah, I dont know about that with turnover/movement rates you stated, it would be time consuming.
Are the computers only used by certain job types or are they all the same group?

When I did apartment complex offices, I labled them Manager, asstmgr, leasing1, leasing 2, etc.
This way when I put the label on it, I had no problem with them telling me which computer.
 
All computers are part of the same group.

In my eyes, why bother with what the PC name is on the floor when you can just add the user's name as a description into Active Directory. Now, instead of walking out onto the floor and looking for a label on a machine, I can open Active Directory and find that exact user.

This seems very very simplistic, so I'm wondering if I'm missing the "bigger" picture here?
 
VLANs are good for this type of thing as you never need to rename them and they can go to any floor and the IP range tells you its location.
 
VLANs are good for this type of thing as you never need to rename them and they can go to any floor and the IP range tells you its location.


Coming from the physical networking realm, I'd second this. For longevity IP addressing and VLAN's is a far more robust model. Again this heavily depends on how the network IP addressing was set up originally. It might have to be tweaked or redone a little, but even done entirely I'd find it to be less work than going machine by machine.
 
We are on a VLAN environment and have IP ranges.

The problem as I understand it, is that when looking at an IP's description in DHCP, it may say gibberish like YMHz1 for the computer name. Now if we need to touch the machine physically, my co-worker says they have no idea how to determine where YMHz1 is on the floor. They would prefer that the name say something like, acc-01, acc-02, etc..

My solution was to add the user's full name next to the computer object YMHz1 in active directory, thus telling my co-worker whose machine it is.

Is this a valid solution?
 
this just sounds mental! Do you not have asset tags that are then used for the hostname?
 
We are on a VLAN environment and have IP ranges.

The problem as I understand it, is that when looking at an IP's description in DHCP, it may say gibberish like YMHz1 for the computer name. Now if we need to touch the machine physically, my co-worker says they have no idea how to determine where YMHz1 is on the floor. They would prefer that the name say something like, acc-01, acc-02, etc..

My solution was to add the user's full name next to the computer object YMHz1 in active directory, thus telling my co-worker whose machine it is.

Is this a valid solution?

the question here is why should anyone have to physically touch any machine on the floor unless another employee reports it physically broken.

if an employee reports a physical problem, then it is a simple "the pc at *insert location name* won't turn on."

label the location, not the pc. cubical number, cnc machine number etc.

all software management and support can and should be done remotely.
 
the question here is why should anyone have to physically touch any machine on the floor unless another employee reports it physically broken.

if an employee reports a physical problem, then it is a simple "the pc at *insert location name* won't turn on."

label the location, not the pc. cubical number, cnc machine number etc.

all software management and support can and should be done remotely.

I agree. The issue was that we had an IP conflict with a machine on the floor and my co-worker wanted to know what user might be affected.

I presented my co-worker with the AD computer object description as a solution and my co-worker completely blew up.

Just wondering if I should stand my ground (new to the IT field, fresh out of school), or go and rename every machine tomorrow morning.

Thanks in advance.
 
I agree. The issue was that we had an IP conflict with a machine on the floor and my co-worker wanted to know what user might be affected.

I presented my co-worker with the AD computer object description as a solution and my co-worker completely blew up.

Just wondering if I should stand my ground (new to the IT field, fresh out of school), or go and rename every machine tomorrow morning.

Thanks in advance.



I say do whatever makes the goddamn boss happy, screw the co-worker. Probably trying to avoid the ass chewing that tends to come in situations like this where someone is being critiqued. Draw up a design and present your argument logically, then take it to the proper individual and defend your case and reasoning. Sometimes you have to remember that you're not in control of the network even if the idea seems flat out stupid. The guy paying you has the final say. I'd just document everything personally in case it ever does blow up in my face.

As far as the IP conflict goes, this could be easily resolved by proper documentation. It could have been someone on your team, or someone before you, but whoever was around from the beginning or before you got there should have had some documentation and hopefully it was kept up to date (sounds like it wasn't if there was).

I'd rather spend my time creating solid documents tying IP(range)/subnet address to which department and where in the building each department is located, how many machines, etc. All written. Looking up affected areas would be easier since you'd know where you'd need to check without any other input from another department.
 
I like naming based on criteria that does not change often, such as the computer MAC, SvcTag, SN, Location, or role. You mentioned your organization had high turnover, so how would it make sense to name computers after users? You should already see which computer a user is logged into, and if you have rogue device/IP, it won't show in AD anyways.
Have you asked your more experienced co-worker for their reasoning? As a new hire, no one expects you to know it all, and asking questions instead of challenging conventions will probably be better for your career at this point.
 
The coworker's approach isn't a bad one "for the machine name" if IT at your company control's machine names and user's can't change them. Including the user's name in the computer name also makes it easy for the user to remember their computer name.

I don't think renaming is necessary though. Renaming every computer in the operating system would break every script that anyone has written that uses name. Keep that in mind if you go that route.

A compromise would be to physically label the systems and create a database with contact information. I don't think you want this in Active Directory because that wouldn't be useful for upgrades etc, and creating useful reports off Active Directory isn't the easiest.

I am puzzled by the DHCP reference though. I don't understand how they are getting conflicts if those values are reserved, and including the IP Address seems unnecessary.

A simple naming convention would make this work unnecessary. username, username1, username2, etc for example. Including the physical location is ok if the user's "never" move their systems.

Finally, looking into why you need this troubleshooting info is probably worth more time. :)
 
We name computers with the primary users username in the computer name. Ones that are shared get room number instead.
We are actually moving away from this to an asset tag of some sort due to the computer names not getting updated when they get moved, or a person leaves etc.
For tracking user to computer we use lansweeper which keeps a record of every computer a user logs into.Works great as when a ticket comes in, I just need to know the user with the problem. I search their username in lansweeper and it shows the computers they have logged into by date. Click the computer I want, and then click the link to launch dameware.
SCCM will also do something similar to this.
Whatever you do, minimize the amount of manual changes you have to do. Tag a device once it comes into the building, set the computer name once and never change it.
 
Last edited:
We name computers with the primary users username in the computer name. Ones that are shared get room number instead.
We are actually moving away from this to an asset tag of some sort due to the computer names not getting updated when they get moved, or a person leaves etc.
For tracking user to computer we use lansweeper which keeps a record of every computer a user logs into.Works great as when a ticket comes in, I just need to know the user with the problem. I search their username in lansweeper and it shows the computers they have logged into by date. Click the computer I want, and then click the link to launch dame ware.
SCCM will also do something similar to this.
Whatever you do, minimize the amount of manual changes you have to do. Tag a device once it comes into the building, set the computer name once and never change it.

That sounds good too. Its all about KISS!

Also thanks for the info, I need to check that program out!
 
We name ours location, number. For example "CHS045" "CRH103". Then we add a description in AD of the user, position/location and model of computer. Works well for us.
 
Our naming convention goes by service tag or serial number (hodgepodge of Dell & HP). When a user creates a service ticket or calls in we just ask for the number or tag on the top of the PC and viola, incredibly simple. With an environment of roughly 900 PC's in 22 locations, it has never failed us
 
Static works great if everyone is good at keeping the documentation up to date.
 
correct me if im wrong here but there are ways to determine who is logged into the machine once you know the machine name...

The issue here is that if there is no user logged into the machine. The only solution to that would be to access the c$ share of the computer and check which profiles are stored on your computer.

are you sure you are not running up against a guy who does not like people such as yourself that's fresh out of school and might have a better way of doing things?

I mean come on... blowing up at you for something that was in place before you got there... ugh...

I think your idea will work but there are issues. Frankly there is no full proof way of attaching computer name to user 100% of the time.
 
are you sure you are not running up against a guy who does not like people such as yourself that's fresh out of school and might have a better way of doing things?

He could also be a kid who hasn't seen how things don't work in the real world like they do in the classroom.
 
so if this whole thing started because of an ip conflict that started without malicious intent, you really need to point out that the network is configured incorrectly. IP conflicts don't just "happen" on a network that is configured correctly.

if the network is broken into VLAN's, those vlans should be on a different subnet. At that point DHCP could either be done by a central DHCP server or one on each vlan. Servers (AD, DNS and whatnot) should be on another VLAN. Communication between all these should be routed. Breaking up net network not only prevents IP conflicts, it also isolates trouble if something does happen.
 
This is why you asset tag enduser systems and name their hostname after it.

If you really want a curveball here, dynamic system assignment w/ virtualized desktops. Today I'll have WkstA and then tomorrow I may have WkstZ. So much for your user to system documentation.
 
We do both. All our computers(approx. 1700) are named LLLD-TTTTTTT and the users name or if its a kiosk/multi use pc, it's location is put into the description in AD. We also use vlans; per floor at our HQ and by type in our call centers and distribution centers.

example: tscd-ffdkrl1.domain.com
TSC for Tuscon
D for desktop or L for laptop
ffdkrl1 is the service tag
 
We name every PC. AD description usually has the username if it is a single user machine. Inventory software keeps track of warranty, model, and everything else.

Location-Department-pcnumber

DC-IT-PC01 or DC-IT-LT01

datacenter-itdept-pc#01

Then if there is a problem you know the physical location, department and pc#. Then you can cross reference AD or your inventory system for everything else. Employees change too often to name a PC after them, plus it is messy.

You definitely should standardized naming, hell renaming a computer happens on the same screen as adding it to a domain. Also if you label them all when a user calls they can give you the PCID instead of 'that one in marketing by the vending machines'
 
Back
Top