Talk:MAC address/Archive 1

Comments from User:Ta bu shi da yu
Excellent work, but nothing on LAA, UAA or BIA and nothing on unicast, multicast, broadcast and functional addresses of frames! I've added all of these things, and some other articles - Ta bu shi da yu 12:18, 28 Jul 2004 (UTC)

Fibre Channel uses EUI64 ?!
I think FC uses 64-bit addresses. Can anyone check that and edit the page as nessesery ?

KLC Consulting Frequent SPAMMER In This Section
It has been observed that any "How To Change MAC" external links added on this topic which refers directly or indirectly to KLC Consulting's website (http://www.klcconsulting.net/smac) which hosts the SMAC software, that the IP address of the anonymous poster is Class A IP address on network 68.0.0.0, 65.0.0.0 & on static IP 192.223.226.6. This has been observed since many months (you may check the history section for proofs). If you collect all the IP addresses then you would find that its clearly indicating that the SMAC commercial software is been advertised in this section and KLC Consulting is actively involved in adding the external links on regular intervals. —The preceding unsigned comment was added by 221.135.179.193 (talk) 06:02, 9 April 2007 (UTC).

Locally Assigned Addresses
The IEEE has assigned some OUIs with the 0x020000 bit set. Can anyone create and/or use a LAA OUI in MAC addresses on their LAN? --Jakllsch 19:37, 23 March 2006 (UTC)
 * The article's information about LAAs was previously wrong. I've updated the paragraph, but the diagram is still incorrect (it still shows the LA bit as second least sig, when it should be second most sig bit)if someone wants to get around to this.  I trust information directly from IEEE to be the most accurate: http://standards.ieee.org/regauth/groupmac/tutorial.html.  This should clear up some confusion about why it looked like the OUI and LAA address spaces were overlapping.131.107.0.73 21:37, 5 February 2007 (UTC)

The IEEE tutorial article states: "The U/L bit indicates whether the MAC address has been universally or locally assigned." IMHO this does not state at all if 0=locally or 1=locally. 213.95.68.8 (talk) 14:28, 12 September 2008 (UTC)

CAUTION! You should read the IEEE tutorial more carefully since they use a Big-Endian Byte Order with a Little Endian Bit Order! Quoting the IEEE tutorial : "The Individual/Group address bit is the least significant bit." At least this should be clear to everybody. So in my opinion the picture in the article is correct but only uses a different bit order! 195.34.146.134 (talk) 13:37, 15 September 2008 (UTC)

Uniqueness
i want to ask one thing...when MAC address is unique for all networked machines...why do they need an IP address at all?
 * They are unique (well, almost) but the heirarchy under which they're organised is useless for routing. Your ethernet card's MAC address says "I'm a txp9000, made by XYZ Semiconductor".  A machine on the other side of the world that wanted to send you a packet wouldn't know how to route that packet.  By contrast, IP addresses are heirarchical by route (very very roughly).  So if the remote machine sends a packet to 66.1.94.133 its router can look up a table that says "send all 66.1.x.x packets to router at 66.1.0.1".  This way the remote router doesn't need to know the (impossibly long) list of exact IP addresses in your company or school - it only needs to know who is responsible for that "block" of IP addresses.  It can do this because IP addresses are heirarchical by location in the wiring plan, whereas MAC addresses are only heirarchical by manufacturer (and don't tell you where on the network a given node is). -- John Fader (talk | contribs) 18:13, 22 Apr 2005 (UTC)

Number of network devices built by a manufacturer
Because of the 3 octets OUI, a manufacturer can only assign 24 bits to a given network device. So, if a manufacturer builds more than 16 million network cards, it will have to stop producing network devices. 16 million is a big number, but not that many. Why have they reserved so many bits to identify a manufacturer? There are probably far less network device manufacturers than there are network devices built by big manufacturers. --Earendel 10:06, 15 June 2006 (UTC)


 * They can always get more. 121a0012 01:22, 16 June 2006 (UTC)


 * Yes, many companies have more than one OUI

OS X
Can someone figure out if you can actually change a MAC address in OS X and update that section? Right now it looks quite silly. It says something like 'theoretically this should work, but i don't know nor am I motivated enough to check'

Burned in addresses
Currently, the article states
 * MAC addresses permanently attached to a product by the manufacturer are known as "burned-in addresses" (BIA) or sometimes as "Universally Administered Addresses" (UAA). The BIA can be overridden with a "Locally Administered Address" (LAA).

Does this mean that if one uses one of the methods to change the MAC address explained later in the article, one is using a LAA? Also, the article on burned-in address seems to disagree with the above statement; it states that there are two types of BIAs, namely UAAs and LAAs. AxelBoldt 22:24, 27 May 2005 (UTC)


 * No, it means that if one changes the MAC address, one is supposed to use an LAA instead of a UAA. I doubt it ever occurred to the designers of MAC-48 at the time that one would deliberately use the BIA of one station on a different station; after all, &lsquo;they are just numbers and have no commercial significance&rsquo;.  121a0012 01:52, May 28, 2005 (UTC)

Myth about why 48 bits were chosen
For years I have been repeating something I heard somewhere: that the inventers of the Ethernet decided there should be enough MAC addresses to uniquely identify every 5-kilogram piece of the earth. Tonight I finally did the math for myself. (Thanks, Google, for programming the "mass of earth" into your calculator as a constant.) Google says the mass of the Earth / (2^48) = 2.12246221 × 10^10 kilograms. Not even close. The best source I found for the real story (using a quick Google search) is, "John Schoch, a former PARC researcher, ... notes that PARC ... gave [Ethernet] a ridiculously large 48-bit addressing scheme—far more than anyone imagined would be needed at the time." <>< tbc 06:21, 20 December 2005 (UTC)

Confused About Terminology
Why is the term 'medium access control address' used by many sources (including the IEEE) but not used or referred to by this article? I'm confused as to the difference (if any) between this term and 'media access control address'. 216.166.159.185 06:14, 23 December 2005 (UTC) M. Glenn Lewis


 * That's a great question - When referring to the MAC address, there is no difference between media and medium access control address. Technically media is the plural form of medium but the IEEE standards appear to use media and medium for different parts of the standard.  I'll research this soon and update the article as necessary to clarify. Sfisher 05:08, 25 December 2005 (UTC)


 * Found it; per IEEE standard 802 "Overview and Architecture" from 2001, medium and media are interchangable in the MAC abbrevation and that standard uses medium. Sfisher 20:22, 2 January 2006 (UTC)

"Changing the MAC address" section
The whole section is not encyclopedic and should be removed. 121a0012 18:18, 7 May 2006 (UTC)


 * I agree; I think that maybe only the fact that it can be changed should be left in. —Preceding unsigned comment added by 68.6.58.5 (talk • contribs)


 * In the Wikipedia offical policy, it states that Wikipedia is not "an indiscriminate collection of information", specifically an instruction manual or how-to (title 8). This section could be construed as a how-to thereby violating section 8 of the guidelines. However, I find the information highly informative, relevant, and valuable whilst not largely violating the Wikipedia guidelines. Unless there is an external source that contains the same information presented here, and I have not found any, I propose it be kept. -Etienne 01:30, 12 July 2006 (UTC)
 * I've collected the MAC address changing information and posted it here. If you're happy with it, remove all the information and write a short section that MAC addresses can be changed and refer to this external link -Etienne 18:44, 19 July 2006 (UTC)
 * This is also an option Changing_MAC_addresses —The preceding unsigned comment was added by Etienne.navarro (talk • contribs) 19:42, 28 July 2006 (UTC).
 * What Wikisource Includes says, under What is excluded:
 * in the Original contributions section,
 * Works created by Wikisource users or otherwise not published in a verifiable, usually peer-reviewed forum do not belong at Wikisource. Wikisource is not a method for an author to get his or her works published and make them available to other people, nor is it a site to discover "new talent".
 * in the Evolving works section,
 * Wikisource's mission is to collect and preserve works in their original form. In light of this, works whose content is expected to constantly change over time (possibly for the purpose of keeping the work updated), to improve the content matter of what has already been published, or to make the text more comprehensive), are excluded from Wikisource's scope.
 * A few examples include
 * Open-ended texts where the author relies upon cooperative efforts by many contributors to finish and improve the work;
 * Compilations where there are many sources of a particular text, and/or the text is to be constantly updated as more relevant information is found and added;
 * Lists (see also below).
 * For more information about static texts, see Wikisource:Text integrity.
 * so I'm not sure that list belongs in Wikisource. Guy Harris 03:09, 29 July 2006 (UTC)


 * So does it belong in Wikipedia? --Etienne 02:32, 31 July 2006 (UTC)

''``Changing the MAC address of the new interface will usually solve the problem. However, this action is generally frowned upon and if the service provider is able to detect that a MAC address is spoofed and prohibits spoofed MAC addresses, this method will not work."''

Is it possible for a service provider to detect that a MAC address is being spoofed? I suspect not! The service provide may guess... For example, for months I use x megabytes of bandwidth per month. I then make a request using another MAC address (my router) which fails. Soon after I revert back to my old MAC address (spoofed by the router). My bandwitch usage subsequently increases to 2x megabytes a month, one could guess that I now have two machines behind the router, but this cannot be proven by the service provider.--Bah23 12:09, 26 January 2007 (UTC)

Required by the F.C.C.?
Every once in a while I hear some schpiel about network cards having MAC addresses is because the FCC requires it, usually claimed by people who know nothing about computers.

Googling around a few minutes all I see is similar hear-say.

I'm leaning toward the this-is-a-wives-tale opinion, but I thought I'd just get confirmation anyway. I see mention of the IEEE handing out MAC addresses on the Network card article, so maybe this is what people think of -- most laymen don't know what the IEEE is or that it even exists, are so maybe they got confused with the FCC considering it's a more "well-known" entity. --I am not good at running 08:48, 19 May 2006 (UTC)


 * Correct, the FCC has nothing to do with assigning MAC addresses. Network cards may have a separate FCC ID, but that is unrelated to the MAC address. Sfisher 05:41, 24 June 2006 (UTC)

ethernet address
"Ethernet address" redirects here. Please write a couple of words about it in the intro. `'mikka (t) 17:37, 17 July 2006 (UTC)

By the way, please clarify the entry in the "aether" disambiguation page: "in internet routing, the term ether is associated with hosts" What the heck is it? (Previously "hosts" was a redirect to "hosts file") `'mikka (t) 17:37, 17 July 2006 (UTC)

bits

 * The standard (IEEE 802) format for printing MAC-48 addresses in human-readable media is six groups of two hexadecimal digits, separated by hyphens (-) in transmission order, e.g. 01-23-45-67-89-ab. This form is also commonly used for EUI-64 .

I think the red part should read This form, with eight groups of two hexadecimal digits instead of six, is commonly used for EUI-64, but I am hesitant to change it because I'm not familar with EUI-64 beyond what I have read here. --203.6.205.131 01:41, 31 July 2006 (UTC)

EUI-64, EUI-48 and MAC-48
"Note: The IEEE now considers the label MAC-48 to be an obsolete term which was previously used to refer to a specific type of EUI-48 identifier used to address hardware interfaces within existing 802-based networking applications and should not be used in the future. Instead, the term EUI-48 should be used for this purpose."

"IPv6 — one of the most prominent standards that uses EUI-64 — applies these rules inconsistently. Due to an error in the appendix to the specification of IPv6 addressing, it is standard practice to extend MAC-48 addresses (such as IEEE 802 MAC address) to EUI-64 using 'FF-FE' rather than 'FF-FF.'"

This is not an error then (IPv6's behaviour in regard to EUI-64)... because http://standards.ieee.org/regauth/oui/tutorials/EUI64.html says:

Restricted encapsulated values To support encapsulation of EUI-48 and MAC-48 values within small subsets of the EUI-64 values, the first four digits of the manufacturer's extension identifier shall not be FFFF16 or FFFE16. Thus, the 64-bit values of the following form are never-assigned EUI-64 values: ccccccFFFEeeeeee16  (an EUI-48 extension) ccccccFFFFeeeeee16  (a MAC-48 extension)

So... if MAC-48 is a special case of EUI-48 and is obsolete, then it can be (consistently) interpreted as EUI-48 and thus converted to EUI-64 with FFFE inserted. This seems to be entirely correct.

Pavlix 14:04, 11 November 2006 (UTC)

sudo su
I'm fairly sure you don't want to do "sudo su"- you just want to do "sudo". 128.61.38.107

unique identifier?
In computer networking a Media Access Control address (MAC address) is a unique identifier attached to most network adapters (NICs).

Its only an unique identifier IF no one spoofs it.--Bah23 12:09, 26 January 2007 (UTC)
 * I agree. I therefore added a little "quasi" to this sentence. --Abdull 16:58, 15 May 2007 (UTC)


 * I disagree with quasi. It makes it sound as if there could be a legitimate duplicate mac address.  It is infeasible to have a legitimate duplicate.  While I agree you could have a duplicate thanks to mac spoofing, a little clarification on this point is necessary.  —Preceding unsigned comment added by 76.100.92.38 (talk) 05:21, 20 February 2008 (UTC)

Link cleanup
I trimmed back the external links by removing all of the tools and how-to links on the basis of Wikipedia is not an instruction manual. I removed the Wireshark links because they were redundant with the remaining two.

I was prompted to look at the links by the back-and-forth placement of nthelp.com and klcconsulting.net links. The owner of klcconsulting.net (one of the editors) rightfully pointed out to me that the nthelp.com site just points to the klcconsulting.net site. The nthelp.com site states the material was copied, which is a copyright violation and even linking to such a site is not allowed by Wikipedia. My preference is to simply leave all of those links out of the article; they add little to understanding the topic. Please discuss new links here before adding them.

Finally, editors are reminded that the conflict of interest guideline indicates adding links to one's own site is to be avoided; suggest changes on the talk page instead and if neutral editors agree, let them place the link. ✤ JonHarder talk 17:55, 2 March 2007 (UTC)
 * Thank you, Jon. ~a (user • talk • contribs) 21:47, 4 March 2007 (UTC)
 * I approve of the continuing cleaning of how-to material from the article. ✤ JonHarder talk 02:16, 6 March 2007 (UTC)
 * I also approve of it. Guy Harris 03:11, 6 March 2007 (UTC)

Uppercase and lowercase
i've noticed some mac addresses use uppercase letter and some lowercase - does it matter if you type them in the correct case or not? Maybe this can be mentioned in the article - thanks —The preceding unsigned comment was added by Shakehandsman (talk • contribs).

Unclear sentence in introduction
"MAC addresses, unlike IP addresses and IPX addresses, are not divided into "host" and "network" portions, so a host cannot determine, from the MAC address of another host, whether that host is on the same layer 2 network segment as the sending host or a network segment bridged to that network segment and, if it's not, cannot determine the MAC address of a router that is on the same network segment as the sending host or a segment bridged to that network segment and that can help route the packet to the destination host."

I get lost reading this about halfway through, can it be clarified? 83.104.231.152 19:59, 17 June 2007 (UTC)

00:00:00:00:00
What does this address mean? I have read online that it usually denotes an error (usually because the MAC can't be read), but that it is also a valid address. 203.33.3.12 00:18, 27 June 2007 (UTC)

Unicast/multicast bit in picture?
There is a bit in the picture that can hold the value "unicast" or "multicast". Does anyone know what the difference is in this situation? Why set or clear that bit? Tommy 00:18, 29 July 2007 (UTC)


 * To quote IEEE Std 802-2002, "IEEE Standard for Local and Metropolitan Area Networks: Overview and Architecture":


 * The Individual/Group (I/G) address bit (LSB of octet 0) is used to identify the destination address as an individual address or a group address. If the I/G address bit is 0, it indicates that the address field contains an individual address. If this bit is 1, the address field contains a group address that identifies one or more (or all) stations connected to the LAN. The all-stations broadcast address is a special, predefined group address of all 1’s.


 * I.e., if the bit isn't set, the address is the address of a network adapter; if it is set, it's a multicast address or a broadcast address. Guy Harris 00:40, 29 July 2007 (UTC)

Thanks Guy. I updated the article according to your explanation. Tommy 12:38, 29 July 2007 (UTC)

wikisource link broken
it seems the Wikisource text on Changing MAC addresses has vanished.

deletion log there says it was transwikied to wikibooks, but i can't seem to find it there.

anyone got a proper link for it?

142.165.95.83 22:55, 20 September 2007 (UTC)

Nokia and naming
Nokia is officially calling the MAC address a "WLAN number". It uses the number as a unique identifier on its Nokia 770 (and presumably its other internet tablets). Whenever I need to engage in some kind of official communication about my N770, Nokia demands my WLAN number (see http://www.nokiausa.com/A4410046 for more info). Is that usage sufficient to update the first line (also known as...)? samwaltz (talk) 19:11, 24 December 2007 (UTC)
 * Do you have a link to a reputable/reliable source? ~a (user • talk • contribs) 21:41, 24 December 2007 (UTC)
 * Aside from the address at nokiausa.com, given above? I'll do a check, and post a few more in the morning. I would, however, think that nokiausa.com would be a sufficiently notable source. samwaltz (talk) 08:34, 31 December 2007 (UTC)

I/G U/LMistake
Corrected as per IEEE: IEEE 193.239.0.5 (talk) 21:20, 19 October 2008 (UTC)

privacy
can someone say something about the privacy implications of MAC addresses? Can they be seen b other net users? Does that matter? Ca they be linked to individuals etc? It's unclear from this article what they are for. —Preceding unsigned comment added by 67.40.250.209 (talk) 16:35, 25 August 2008 (UTC)


 * They can only be seen by people on the same physical network. In other words, when stuff hits a router, people on the one side can not see the MACs of those on the other side.  Of course this is baring any intrusion that may be taking place.  It does not really matter if other people see them though.  As for being linked to individuals I think you mean can anyone say who uses a particular MAC address...  Short answer, no.  However, if a company who sells you a device logs the MAC along with their sale record to you they could potentially know (but who is to say you haven't sold it to someone else yourself).  As for what they are used for: MAC addresses is what computers use to talk to each other.  Many people think its simply just IP, but in reality an IP address is basically just an alias for your MAC address.  Say you have two computers on the same network with the IPs of 192.168.2.1 and 192.168.2.2.  In order for them to communicate, they listen to the data line to see packets from MACs they do not know.  When they see one, they secretly record the IP associated with that packet and log it in their ARP (Address Resolution Protocol) table.  If they have not yet figured out what MAC belongs to an IP they send out an ARP request.  Basically this request just says: "Who is 192.168.2.2?".  192.168.2.2 would then reply with "I'm 192.168.2.2 and my MAC is bla bla".  Then that device can now communicate with the other device using the MAC address.


 * To get just a little more technical, MAC addresses work on layer 2 of the OSI model. Layer two is the data link layer.  This is the layer that computers actually make connections with each other and they do so with the MAC address.  Layer 3 is the network layer in which IP resides. To sum it up, MAC addresses are used to make the actual direct connection and IP is used to determine the network on which a computer (or MAC address) resides.  Fortunatly, computers on different networks do not need to know the MAC address of computers on other networks, even when they are communicating with each other.  This is because they are communicating over a network (Layer 3) instead of directly (layer 2).


 * Hope that helps. Jwjkp (talk) 13:27, 23 December 2008 (UTC)

Globally Unique (OUI Enforced) Bit
Can anyone explain why neither XP nor Vista seems to set the Globally Unique (OUI Enforced) Bit (bit 2 of the first byte) when they generate an address for, say, a VPN software interface, e.g.

PPP adapter name withheld:

Connection-specific DNS Suffix. :       Description. . . . . . . . . . . : WAN (PPP/SLIP) Interface Physical Address. . . . . . . . . : 00-53-45-00-00-00       Dhcp Enabled. . . . . . . . . . . : No       IP Address. . . . . . . . . . . . : withheld Subnet Mask. . . . . . . . . . . : 255.255.255.255       Default Gateway. . . . . . . . . : withheld DNS Servers. . . . . . . . . . . : withheld Primary WINS Server. . . . . . . : withheld Secondary WINS Server. . . . . . : withheld

--Dinosaurclover (talk) 21:07, 23 November 2008 (UTC)

Fixed finding MAC under Windows XP
The instructions for finding the MAC address under Windows XP recommended using arp -a, which shows the MAC of other machines on the Local Area Network. The command ipconfig/all shows the MAC address for all network devices. David Harbaugh


 * The "arp -a" works fine if the network is working properly. 
 * Sadly, the "arp -a" command can fail if:

the remote host is disconnected, the remote host is off (except for Wake-on-LAN), the remote host is on a different subnet,Proxy ARP is in use,(you get the Proxy's MAC address instead)the ARP/IP address pair has timed out and been purged from the local ARP table.two hosts are trying to use the same IP address, so only the latest IP/ARP address pair is stored in the local ARP table,a static ARP/IP address pair has been entered in the local ARP table, so new MAC addresses for the static IP address are just tossed away,the remote system is running a protocol like DECnet which uses a locally administered address, so the "burned in" address isn't seenthe local host is not running the TCP/IP protocol :-) 
 * To attempt to get a MAC address on a remote subnet, say across a router, the NBTSTAT -A IP address and NBTSTAT -a remote_SMB_hostname commands will retrieve the remote MAC address for many systems, but Samba systems respond back with MAC Address = 00-00-00-00-00-00 instead.Lent (talk) 16:45, 25 January 2010 (UTC)

Internal format of MAC
There is some formats of MAC addresses, I've read about them in "Computer Networks" by Andrew Tanenbaum.

Is it possible for a PC to lose its MAC address configuration permanently? What I mean is, my PC crashed during a thunderstorm, after it came back on I could no longer access my network at home. Which is using DHCP. When I brought it to my office and connected it to my LAN I was able to connect. I also connected it directly into a device that is using DHCP and the PC recognized the connection. I then brought the PC home and reconnected it, in the hope that by some slim chance it would work, it did not. So that is why I am asking if it is possible for a PC to permanently lose a MAC address.
 * From your description, the PC sounds fine. More likely, the router box providing DHCP leases for IP addresses has a problem. A number ISP provided home routers seem to suffer from a "no more IP addresses for you" problem.  Usually power cycling the router will clear the problem.  Of course, another problem could be a physical cabling problem.  Also As your router box is likely connected directly to cables which run outside, this box could be suffering as a result of lightning from the thunderstorm.Lent (talk) 19:32, 25 January 2010 (UTC)

Do all WiFi capable devices have a MAC address?
I'm doing some research into a potential software product and need to understand if all WiFi enabled products have a unique MAC address. By this I mean games consoles, phones, PDAs etc etc. 93.97.144.211 (talk) 13:25, 18 February 2010 (UTC)
 * Any device which transmits 802.11 frames must use a MAC address. In theory, a device could ship without a pre-programmed ("burned in"), Universally Administered Address (UAA), requiring the operator to enter a Locally Administered Address (LAA), but it would still need a MAC address.  Practically everything which can transmit ships with a UAA pre-programmed.  Passive devices which only listen/monitor do not need a MAC address.  •  If you're researching a commercial application, you should hire a professional engineer, rather than trusting what random people on Wikipedia say.  — DragonHawk (talk|hist) 14:23, 18 February 2010 (UTC)

More sub-headings ?
Could we perhaps add some sub-headings to make it easier to locate subsections that deal with 'special' addresses and address types, such as say broadcast, multicast group? And maybe move EUI64 down into its own subsection with a subheading? I suggest that this would make the article more accessible. Thoughts? Suggestions for subheadings?

Could I perhaps get some help putting together a table of notable special/sacred numerical values, ranges too?CecilWard (talk) 16:39, 14 July 2010 (UTC)

On what goes in the Externals list
I notice a bit of revert controversy from time to time regarding whose MAC address lookup sites to include in the externals list. My feeling is, the fewer the better, otherwise everyone and his dog will say "you have all those, why is it fair not to include my MAC address information page also?"

On the specific subject of Michael Patton's "Ethernet Codes Master Page", user Kvng commented when reverting a removal of the link, "don't delete links just because they're dead. see WP:ROT". But I feel it is worth mentioning that WP:ROT has some very specific wording on this subject: "Except for URLs in the External links section that have not been used to support any article content, do not delete a URL solely because the URL does not work any longer." With the emphasis on solely and the explicit exception covering Externals, it almost seems that WP:ROT has been written specifically to allow for cleaning dead links out of the Externals section without further ado. On the other hand, I was able to find the new URL easily, and the content seems to be relatively useful and non-commercial, and the URL seems to be the proper canonical source for the material.

But in general, if the trend continues, the Externals section will decay into all kinds of duplication and link spam. I foresee a need for some rules on what to include. I'm not an expert editor, so I don't know what's proper. Maybe someone should post links to guidelines here? JMCorey (talk) 21:14, 22 June 2011 (UTC)


 * You are right. An IP poster had an external link to a ad-ladened site reverted by DragonHawk on June 13 and may be deleting everything in retaliation. I reverted the IP poster as did Kvng (the ISP in the three instances is CHONGQING PROVINCE NETWORK).


 * I spent some time reviewing the external links subsequently to make sure they were reasonable. I made a close judgement call that although the Michael Patton's "Ethernet Codes Master Page" link was available on cavebear's archive page, 1. the page itself was a mirror and the original is gone, 2, almost all the links were broken on the landing page, so I felt I wasn't doing anybody a favour by leaving it in, just a bunch of confusion. I've changed my mind and agree with your and Kvng's consensus to leave it in.


 * Each of the other external links seemed to have something different and unique to contribute when I reviewed them, especially the last one with access to Google's geolocation database. Darrell_Greenwood (talk) 23:15, 22 June 2011 (UTC)

Finding a MAC address
I needed to find various MAC addresses of a GNU/Linux and a Mac OS X machine. This involved taking one external link from this article, finding that it was irrelevant, taking another, finding that the explanation was outdated and/or obscure, and finally:


 * sudo ifconfig -a

I noticed in passing that with a Windows machine that's not grotesquely unsuited for today's internet, the command would be


 * ipconfig /all|more

I realize that WP is not a compendium of "howto" manuals, but suggest that this information would be useful and worth inclusion in the article. Comments? -- Hoary (talk) 07:16, 23 January 2010 (UTC)


 * Agreed. Added what is hopefully a stable non-commercial link under .gov. Darrell_Greenwood (talk) 03:33, 25 January 2010 (UTC)
 * Here's a search that yields ".edu" sites for these types of instructions.Lent (talk) 16:57, 25 January 2010 (UTC)


 * To answer the original question:
 * arp -i eth0 -n
 * will show the mapping from IP addresses to MAC addresses on your local network. — Preceding unsigned comment added by 74.96.165.138 (talk) 14:36, 25 August 2011 (UTC)

IPv6 error
I have moved the following comment from the article. -—Kvng 16:33, 17 December 2012 (UTC)


 * IPv6 — one of the most prominent standards that uses a Modified EUI-64 — treats MAC-48 as EUI-48 instead (as it is chosen from the same address pool) and toggles the U/L bit (as this makes it easier to type locally assigned IPv6 addresses based on the Modified EUI-64). This results in extending MAC addresses (such as IEEE 802 MAC address) to Modified EUI-64 using only FF-FE (and never FF-FF</tt>) and with the U/L bit inverted.RFC 5342


 * Note: The last statements regarding IPv6 seem to be incorrect. The RFC referred to only contains a note that IETF treats MAC-48 as EUI-48, because that "doesn't cause any problems in practice". It does not specify that everyone else should be doing the same. --

Labeling
Is there any standard for labeling on equipment? I've seen barcodes used but I haven't found any indication that there's a standard for this. For example, here's a Code 128 label:
 * MAC address barcode 00-17-4F-08-5D-69.jpg
 * --SpareSimian (talk) 01:00, 21 December 2012 (UTC)


 * I'm not aware of any standard. There's certainly no requirement to print the MAC address on equipment. -—Kvng 17:24, 24 December 2012 (UTC)

Is There a Way to Make This More Understandable to Non-Computer Science Students?
"A Media Access Control address (MAC address) is a unique identifier assigned to network interfaces for communications on the physical network segment. MAC addresses are used for numerous network technologies and most IEEE 802 network technologies including Ethernet. Logically, MAC addresses are used in the Media Access Control protocol sub-layer of the OSI reference model." Glad I asked,thanks a lot !76.166.245.241 (talk) 18:11, 24 November 2010 (UTC)


 * Try this: http://simple.wikipedia.org/wiki/MAC_address --2001:980:A4CB:1:7D09:88D5:B721:DAEB (talk) 19:21, 7 April 2013 (UTC)

OUI36
The article contains no information about the new OUI36! The IEEE has already begun to make reservations. — Preceding unsigned comment added by 84.176.137.225 (talk) 19:53, 28 September 2013 (UTC)

Uniqueness
Today I've learned that in this world of virtualization, MAC addresses are no longer globally unique. Best practice for virtualized environments requires only that MAC addresses be unique per subnet. Read the Setting up MAC addresses section of this VM documentation for a discussion of the new world order. They've not even bothered to recommend locally administered addresses. Eeek! ~KvnG 20:34, 15 July 2013 (UTC)


 * That is nothing new. MAC spoofing has been a common practice for decades. Spoofed MAC addresses in Windows did not require the "locally administered address" bit prior to Windows 7 and other operating systems continue to allow spoofing OUI MAC addresses to this date. Most home network products also allow complete freedom in spoofing their MAC. So, even the average home user has been perfectly capable of spoofing those "globally unique" OUI addresses for a long time. -91.157.57.36 (talk) 01:47, 2 April 2014 (UTC)

Error - Locally Administered Address and Multicast flags place on wrong bits
8 bits in Most mean octet used for Locally Administered Address (LLA) and Multicast must be placed to 1st bit and 0 bit accordingly.

This is has simple logic - 1st bit what will be in wire it will be... surprised? bit numberd 40 (0 bit from 5th octet) - we get Multicast packet, and reducing overhead in switch hardware so. LLA simple placed on next bit.

Author possibly takes obsoleted info from link on old IEEE pdf file.

For proof: http://www.certsoft.com/mac.htm (this main - they referring to IEEE standard) http://www.noah.org/wiki/MAC_address http://lizardsystems.com/wiki/change_mac_address/faq/change_mac_address_in_windows_7

Anybody can check this with proven utility - wireshark, by editing MAC on own PC

If it will be need, I can to check this in Linux or FreeBSD source code. — Preceding unsigned comment added by Kosta.Shtabalyuk (talk • contribs) 00:37, 5 November 2013 (UTC)


 * I assume this comment refers to the figure in the MAC address section. The figure is correct. Over Ethernet, bytes are transmitted least significant bit first. The first bit transmitted indicates whether the address is multicast. ~KvnG 05:44, 7 April 2014 (UTC)

Canonical presentation format
The article claims
 * The standard (IEEE 802) format for printing MAC-48 addresses in human-friendly form is six groups of two hexadecimal digits, separated by hyphens (-) or colons

but does not substantiate that claim with a reference. I have been able to find a couple of sources that promote the use of hyphens, but none that promote colons: which both state
 * https://standards.ieee.org/develop/regauth/tut/eui48.pdf
 * https://standards.ieee.org/develop/regauth/tut/eui64.pdf
 * An EUI-48[64] is properly displayed as shown with hyphens between numbers in canonical address representation...

There is also: which states:
 * https://tools.ietf.org/html/rfc7043
 * The Address field MUST be represented as six two-digit hexadecimal numbers separated by hyphens.

Although the use of colons is common, I believe it is overreaching to claim that it is supported by a standard. It would be better to state that colons are commonly used as an alternative representation. — Preceding unsigned comment added by 2403:3400:6001:125:381E:C47D:8682:73F7 (talk) 05:40, 7 September 2015 (UTC)

I discovered that an earlier revision of this article made a distinction between standard and alternative notation. The elevation of the colon notation to "standard" was made (incorrectly in my opinion) in this edit.
 * That edit was by Kbrose who will be able to throw some light on the matter. The RFC 7043 mentioned above is for how an EUI-48 address should be stored in a DNS server and is not really relevant to this text which refers to a "human-friendly form". Please remember to sign comments with four tildes. Johnuniq (talk) 10:39, 7 September 2015 (UTC)

I got hold of the IEEE 802-2014 standard in which section 8.1 defines the hexadecimal representation which uses hyphens. A bit-reversed representation that uses colons is also defined and accompanied by a note that states "the bit-reversed representation is of historical interest only and is no longer applicable to any active IEEE 802 standard." The standard gives the following example of a particular MAC address represented both ways:
 * Hexadecimal representation: AC-DE-48-12-7B-80
 * Bit-reversed representation: 35:7B:12:48:DE:01

Clearly the bit-reversed representation is not the same as the hexadecimal representation with colons replacing hyphens and is therefore different from the colon representation that is seen more commonly. I believe this adds additional weight to the argument for having the article acknowledge only the hyphen representation as standard with other representations identified as non-standard alternatives. 2403:3400:6001:125:381E:C47D:8682:73F7 (talk) 16:12, 8 September 2015 (UTC)

Is the MAC Address transmitted via the Browser?
I read the article and could not find an answer to the question above. It has to do with Internet Browsing Security and Anonymity. I've read some posts that indicate at least some people believe this information is given out when one browse's the internet, so that a person's computer's identity can be recorded as having visited a particular site, etc...

Yet, I've never read anything from any reputable Computer Security organization that discusses the need to monitor and control the MAC Address (vs. cookies, flash cookies, IP Address, etc...)

Is the MAC Address a threat to one's anonymity and/or security? My sense is that it is not, and that it is only used and made available within the local network, and once the computer is outside that network, the external IP of the router/modem is the only identifying information that can been seen.

````Jonny Quick — Preceding unsigned comment added by Jonny Quick (talk • contribs) 14:51, 13 August 2011 (UTC)
 * Questions like this should be at Reference desk/Computing, but the answer is no. Privacy is threatened by the cookies and IP address you mentioned, and other ugly tricks which the refdesk can probably enumerate, but the MAC address of your modem/router (or possibly computer) are seen only by the ISP, and not anyone else. Apparently it is possible on a very insecure computer for something running in Internet Explorer to read a computer's MAC address, but on the network, your MAC address is only available at your ISP. Johnuniq (talk) 23:58, 13 August 2011 (UTC)


 * The above answer from Johnuniq (that the MAC address is not transmitted across the net) may be true for IPv4, but in the case of IPv6, it is less clear. There is some discussion of privacy in the IPv6 article.  -- HLachman (talk) 19:50, 6 March 2016 (UTC)

How MAC addressing relates to routers
From a networking point of view, how is the MAC address technology able to provide internet services? Jibbzion (talk) 11:16, 28 January 2017 (UTC)

A network node with multiple NICs must have a unique MAC address for each.
The article says: A network node with multiple NICs must have a unique MAC address for each. As far as I know, 802.3 ethernet allows for either assigning MAC addresses to ports or hosts. The only ones that I know that use the latter are Sun systems. Early Sun systems put the MAC PROM on the CPU board, used by either on-board or add-on ethernet ports. Has this ability been removed in later versions of 802.3? Gah4 (talk) 08:37, 2 June 2019 (UTC)


 * I don't think that is actually defined anywhere in 802. The standards talk about entities which might (or might not) indicate individual network terminations, ie. NICs. The Sun scheme worked because at that time multiple NICs in a system were extremely uncommon, even more so connecting them to the same network. If two NICs in a system share the same MAC you run into trouble in many scenarios (e.g. attaching each NIC to a separate VLAN on the same switch doesn't work with some switches, neither does link bonding/teaming). So I guess A network node with multiple NICs must have a unique MAC address for each is more a practical rule than one of standard. --Zac67 (talk) 10:07, 2 June 2019 (UTC)


 * Using a Sun as a router wasn't that uncommon. Connecting two to the same network would be unusual at the time, though. That was way before VLAN and even switches.  The two port bridge was expensive and rare.  I wonder about dedicated routers, which were mostly Cisco at that time. With non-removable ethernet (usually AUI) ports, the NIC idea doesn't really apply. Though the bigger Cisco routers might have removable NIC-like devices. Gah4 (talk) 19:25, 2 June 2019 (UTC)


 * Of course a citation would be welcomed but, until that arrives, the text should stay as it is. A unique MAC for each NIC is current standard practice and avoids some problems that identifies. ~Kvng (talk) 13:41, 5 June 2019 (UTC)

POP QUIZ: Why is page view history a near-perfect sinusoidal graph?
Check this out: https://tools.wmflabs.org/pageviews/?project=en.wikipedia.org&platform=all-access&agent=user&range=latest-90&pages=MAC_address

The page view history of this page is a near-perfect sinusoidal graph. Why?

Hint: if I can figure it out (and I did) so can you! --В²C ☎ 06:03, 14 December 2017 (UTC)
 * Because somebody is vandaling the page in that demeanor? ZBalling (talk) 19:39, 24 August 2019 (UTC)
 * Because it's not and because no one cares about MAC addresses on the weekend. ~Kvng (talk) 14:55, 27 August 2019 (UTC)

Bit 1 is the least significant bit?!
As far as I know, absolutely no software engineers call the least significant bit bit 1. The value of a bit is 2 raised to the power of the bit number, therefore the least significant bit is bit 0. Calling the least significant bit "bit 1" just creates confusion and reduces the perceived competence of the author. Also, "Universally administered and locally administered addresses are distinguished by setting the second least significant bit of the most significant byte of the address", contradicts the diagram. The diagram says that the least significant bit determines whether it is multicast. —Preceding unsigned comment added by 66.235.19.120 (talk) 17:24, 11 April 2010 (UTC)
 * MAC is mostly for hardware, so it doesn't matter so much what software engineers call it. Ethernet puts bytes on the wire LSB first. Most of the time it doesn't matter to software what order the hardware puts bits on the wire, as long as they come back the same way. Most people don't know what order bits are written on disk, for example, because it mostly doesn't matter. In the case of Ethernet, though, the first bit on the wire is special to hardware, and the second has special significance. However, a MAC address is not a numerical value, but a bit pattern. On the other hand, if individual bytes are loaded into registers, we can give them significance. Hardware engineers can number bits anyway that they want, and that includes numbering them starting with one. Since they aren't actual numerical values, there isn't much of a reason not to number them from one. Note, for example, that we never add, subtract, or multiply them. Gah4 (talk) 02:21, 14 September 2020 (UTC)
 * MAC is mostly for hardware, so it doesn't matter so much what software engineers call it. Ethernet puts bytes on the wire LSB first. Most of the time it doesn't matter to software what order the hardware puts bits on the wire, as long as they come back the same way. Most people don't know what order bits are written on disk, for example, because it mostly doesn't matter. In the case of Ethernet, though, the first bit on the wire is special to hardware, and the second has special significance. However, a MAC address is not a numerical value, but a bit pattern. On the other hand, if individual bytes are loaded into registers, we can give them significance. Hardware engineers can number bits anyway that they want, and that includes numbering them starting with one. Since they aren't actual numerical values, there isn't much of a reason not to number them from one. Note, for example, that we never add, subtract, or multiply them. Gah4 (talk) 02:21, 14 September 2020 (UTC)

Nonsensical sentence
The article says "The DECnet software assigns the last three bytes of the MAC address to be AA-00-04-00-XX-YY where XX-YY reflects the DECnet network address xx.yy of the host.", i.e., the last three bytes are set to 6 bytes, which obviously is impossible. I a not sure what is meant here, but maybe it should read "The DECnet software assigns MAC addresses of the form AA-00-04-00-XX-YY..."? 37.49.68.13 (talk) 10:49, 28 October 2021 (UTC)
 * You've read the previous sentence, haven't you? AA-00-04 is the DEC OUI and the last three bytes add 00-XX-YY for a MAC address of the form AA-00-04-00-XX-YY. --Zac67 (talk) 11:45, 28 October 2021 (UTC)
 * The previous sentence says AA-00-04 is DEC's OUI. There is such an entry listed but this is a strange one because the local bit is set. (The corresponding OUI without local set is A8-00-04 and that's not listed.) I think this is the hybrid situation that is referred to - setting the local bit but perhaps not used in the manner intended by the IEEE. ~Kvng (talk) 14:30, 2 November 2021 (UTC)
 * I think the best description is that it is a local assigned address. It goes with DECnet, independent of who makes the hardware or software. It is most commonly seen on Cisco routers, and I suppose also DEC hardware. I believe in all cases, it overrides the address in ROM. If you move the hardware to a different network, the number will change. I am not sure what IEEE says about it, though. Gah4 (talk) 15:27, 2 November 2021 (UTC)
 * More specifically, DECnet Phase IV uses 16 bit addresses, with 6 bits for area, and 10 bits for node within the area. They ran out of addresses much earlier than IPv4! They were used well into the 1990's though. I don't know about Phase V. Gah4 (talk) 15:38, 2 November 2021 (UTC)
 * I think the best description is that it is a local assigned address. It goes with DECnet, independent of who makes the hardware or software. It is most commonly seen on Cisco routers, and I suppose also DEC hardware. I believe in all cases, it overrides the address in ROM. If you move the hardware to a different network, the number will change. I am not sure what IEEE says about it, though. Gah4 (talk) 15:27, 2 November 2021 (UTC)
 * More specifically, DECnet Phase IV uses 16 bit addresses, with 6 bits for area, and 10 bits for node within the area. They ran out of addresses much earlier than IPv4! They were used well into the 1990's though. I don't know about Phase V. Gah4 (talk) 15:38, 2 November 2021 (UTC)

An Unreadable Sentence
From section: “Address Details”, (third paragraph, second sentence):

“The IAB uses an OUI from MA-L (MAC address block large) registry was previously named OUI registry, the term OUI is still in use, but not for calling a registry belonging to the IEEE Registration Authority, concatenated with 12 additional IEEE-provided bits (for a total of 36 bits), leaving only 12 bits for the IAB owner to assign to their (up to 4096) individual devices.”   (65 words)

With all due respect to whoever wrote this, it is largely unintelligible to the general reader for whom this encyclopaedia is primarily intended. The sentence is too long and is not very grammatical. There must be something amiss if it has to contain three sets of brackets. (Might a glossary of acronyms somewhere at the foot of the article help?) I would have tried editing the sentence but, having little in-depth knowledge of the subject, felt that there would be a real risk of my introducing unintended error into the text.

This is just one of many highly technical articles in the encyclopaedia where editors, completely at home in their own subject, can sometimes lose sight of the audience for whom they are writing; after all, they know what they are talking about and somehow expect everyone else to have the same level of knowledge.

I recognise the great difficulties involved in presenting complex content to a readership with widely differing degrees of familiarity with the subject. I simply ask editors here to: Remember the Readers! Freeman501 (talk) 14:49, 12 October 2022 (UTC)