Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Basically I want to ask question about configurable addressing scheme for LAN interface hardware. I have read about it from a book, some main points are given by:

a configurable addressing scheme provides a mechanism that a customer can use to set a physical address.The mechanism can be manual (the switches that must be set when the interface is first installed), or an electronic memory such as an EPROM that can be downloded from the computer(what does this means). Most hardware needs to be configured only once- configuration is usually done when the hardware is first installed.

Question: Suppose a network administrator configures the LAN interface hardware (assigns the address) when he installs it. Now later on if he needs to change the physical address of the device can he change it? Or in this addressing scheme the hardware can only be configured once and we can not reconfigure it later on.

share|improve this question

Back in the elder days of the 80's and early 90's, when such addresses were actually configurable for some network network protocols, the addresses were set by a variety of means. In some cases DIP switches on the network card were used to configure the significant bits; if a change was needed the Computer Technician had to pop open the chassis and change the address by hand. If an entire network needed to change addresses, a configuration party was organized by the IT staff and a bunch of people hit every network station by hand. Generally overnight or on a weekend, which could get spendy if overtime was involved.

In modern days, the network address (at least for Ethernet, I can't speak to other networking protocols) is set in BIOS on the device with the ability to be overridden by the driver software. It is a very rare case where there is a real need to change the factory-shipped address. The case Brian alluded to, two cards from the same manufacturer sharing the same address, is one such rare case. The other case is if the MAC address is encoded into an access control list somewhere, and it is easier to change the address on a replacement device than it is to change it in the ACL list.

share|improve this answer
While I'm too young to have participated in a "configuration party", I'm guessing that all the pizza and beer in the world won't make you feel like you had an awesome time at this event.... Much like the "computer lab cleanup party" I recently had, or the "desktop imaging and rollout party" I'm gearing up for ;) – Holocryptic Jul 26 '10 at 21:03
Just as fun as you imagine. What made it a party was the relaxed dress code. – sysadmin1138 Jul 26 '10 at 22:12
+1 for shorts and sandals – Holocryptic Jul 26 '10 at 22:28

It is possible to manufacture devices that are programmable only once. Most allow multiple configurations though.

In the case of the hardware MAC address, it's a very bad idea to fiddle with it. Each manufacturer has a block (range) that they are assigned and they allocate from. If you set a device randomly, you could end up with 2 devices on the network with the same address. That's a bad thing since you can't tell them apart and it causes networking problems. I've seen this happen before when 2 devices from the same manufacturer that we had (and were NOT programmable) had the same address...

share|improve this answer
I had that happen once as well. It took a while to figure out what was wrong. – sysadmin1138 Jul 26 '10 at 20:33

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.