|
Home > Archive > alt.certification.cisco > April 2003 > Switching and Duplicate MACs?
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Switching and Duplicate MACs?
|
|
| Knisa 2003-04-22, 10:24 pm |
| I just finished reviewing the switching section of my CCNA Study Guide, but
something occured to me that has me curious...
In theory, there shouldn't be any duplicate MAC addresses out there.
However, it is possible for them to come up. How would a switch handle
having two identical MAC addresses on different interfaces?
It goes without saying that you're going to have problems if this happens,
but I'm mostly curious about what the switch is going to do... ignore the
second one it sees? Overwrite the table entry every time it gets a frame
from the other MAC? Shutdown both interfaces? Shutdown the highest cost
interface, believing that it's in a loop? Something else?
TIA,
Steve
--
"Find out just what any people will quietly submit to and you have the exact
measure of the injustice and wrong which will be imposed on them." -
Frederick Douglass
| |
| Hansang Bae 2003-04-22, 10:24 pm |
| In article <JWmpa.2982$zt3.348592@nnrp1.ptd.net>,
vtoneohfive@hotmail.com says...
> I just finished reviewing the switching section of my CCNA Study Guide, but
> something occured to me that has me curious...
>
> In theory, there shouldn't be any duplicate MAC addresses out there.
> However, it is possible for them to come up. How would a switch handle
> having two identical MAC addresses on different interfaces?
>
> It goes without saying that you're going to have problems if this happens,
> but I'm mostly curious about what the switch is going to do... ignore the
> second one it sees? Overwrite the table entry every time it gets a frame
> from the other MAC? Shutdown both interfaces? Shutdown the highest cost
> interface, believing that it's in a loop? Something else?
It will keep overwriting the CAM table. And complain about quick MLS
cache changes if there's enough cycling going on (of the MAC addresses)
--
hsb
"Somehow I imagined this experience would be more rewarding" Calvin
*************** USE ROT13 TO SEE MY EMAIL ADDRESS ****************
******************************
******************************
********
Due to the volume of email that I receive, I may not not be able to
reply to emails sent to my account. Please post a followup instead.
******************************
******************************
********
| |
| Knisa 2003-04-23, 12:24 am |
| Ah, thanks =).
Steve
--
"Find out just what any people will quietly submit to and you have the exact
measure of the injustice and wrong which will be imposed on them." -
Frederick Douglass
"Hansang Bae" <uonr@alp.ee.pbz> wrote in message
news:MPG.190fd0099912ee7998995f@news-server.nyc.rr.com...
> In article <JWmpa.2982$zt3.348592@nnrp1.ptd.net>,
> vtoneohfive@hotmail.com says...
> > I just finished reviewing the switching section of my CCNA Study Guide,
but
> > something occured to me that has me curious...
> >
> > In theory, there shouldn't be any duplicate MAC addresses out there.
> > However, it is possible for them to come up. How would a switch handle
> > having two identical MAC addresses on different interfaces?
> >
> > It goes without saying that you're going to have problems if this
happens,
> > but I'm mostly curious about what the switch is going to do... ignore
the
> > second one it sees? Overwrite the table entry every time it gets a
frame
> > from the other MAC? Shutdown both interfaces? Shutdown the highest
cost
> > interface, believing that it's in a loop? Something else?
>
> It will keep overwriting the CAM table. And complain about quick MLS
> cache changes if there's enough cycling going on (of the MAC addresses)
>
> --
>
> hsb
>
> "Somehow I imagined this experience would be more rewarding" Calvin
> *************** USE ROT13 TO SEE MY EMAIL ADDRESS ****************
> ******************************
******************************
********
> Due to the volume of email that I receive, I may not not be able to
> reply to emails sent to my account. Please post a followup instead.
> ******************************
******************************
********
| |
|
| Legacy Hubs will have a mac table timer that may or may not flush
automatically if it doesnt recieve traffic from the address
Your 1900 + and up cisco switches will flush old entries for you
ATM Switches have will depend on vendor.
This is quite common on corporate networks that utilize a lot of laptops
that travel from conf room and such. On your cisco ethernet/fast/gig
switches you should never have a mac duplication due to traveling pc's
unless you have portfast enabled. FDDI and ATM are another story though.
"Knisa" <vtoneohfive@hotmail.com> wrote in message
news Iopa.2986$zt3.350272@nnrp1.ptd.net...
> Ah, thanks =).
>
> Steve
> --
> "Find out just what any people will quietly submit to and you have the
exact
> measure of the injustice and wrong which will be imposed on them." -
> Frederick Douglass
>
>
> "Hansang Bae" <uonr@alp.ee.pbz> wrote in message
> news:MPG.190fd0099912ee7998995f@news-server.nyc.rr.com...
> > In article <JWmpa.2982$zt3.348592@nnrp1.ptd.net>,
> > vtoneohfive@hotmail.com says...
> > > I just finished reviewing the switching section of my CCNA Study
Guide,
> but
> > > something occured to me that has me curious...
> > >
> > > In theory, there shouldn't be any duplicate MAC addresses out there.
> > > However, it is possible for them to come up. How would a switch
handle
> > > having two identical MAC addresses on different interfaces?
> > >
> > > It goes without saying that you're going to have problems if this
> happens,
> > > but I'm mostly curious about what the switch is going to do... ignore
> the
> > > second one it sees? Overwrite the table entry every time it gets a
> frame
> > > from the other MAC? Shutdown both interfaces? Shutdown the highest
> cost
> > > interface, believing that it's in a loop? Something else?
> >
> > It will keep overwriting the CAM table. And complain about quick MLS
> > cache changes if there's enough cycling going on (of the MAC addresses)
> >
> > --
> >
> > hsb
> >
> > "Somehow I imagined this experience would be more rewarding" Calvin
> > *************** USE ROT13 TO SEE MY EMAIL ADDRESS ****************
> > ******************************
******************************
********
> > Due to the volume of email that I receive, I may not not be able to
> > reply to emails sent to my account. Please post a followup instead.
> > ******************************
******************************
********
>
>
| |
| Stuart Robinson 2003-04-23, 9:24 am |
| > Legacy Hubs will have a mac table timer that may or may not flush
> automatically if it doesnt recieve traffic from the addres
Why would a hub have a MAC table ?
Stuart.
| |
| Knisa 2003-04-23, 11:24 am |
| I was referring to two devices with the same MAC address on at the same
time, not just one computer getting moved around. For example, many NICs
have software settable MACs. Someone looking to cause trouble could set two
identical MACs on different interfaces, and have them trying to communicate
with other devices. I was wondering how the switch would deal with this.
I'm sorry if I wasn't clear enough with my first inquiry. Thanks,
Steve
--
"Find out just what any people will quietly submit to and you have the exact
measure of the injustice and wrong which will be imposed on them." -
Frederick Douglass
"Bob" <robert.watson@covad.net> wrote in message
news:b860ct$2m6$1@sun-news.laserlink.net...
> Legacy Hubs will have a mac table timer that may or may not flush
> automatically if it doesnt recieve traffic from the address
>
> Your 1900 + and up cisco switches will flush old entries for you
>
> ATM Switches have will depend on vendor.
>
> This is quite common on corporate networks that utilize a lot of laptops
> that travel from conf room and such. On your cisco ethernet/fast/gig
> switches you should never have a mac duplication due to traveling pc's
> unless you have portfast enabled. FDDI and ATM are another story though.
| |
| Bernie 2003-04-23, 7:24 pm |
| On Wed, 23 Apr 2003 02:32:09 GMT, "Knisa" <vtoneohfive@hotmail.com>
wrote:
>I just finished reviewing the switching section of my CCNA Study Guide, but
>something occured to me that has me curious...
>
>In theory, there shouldn't be any duplicate MAC addresses out there.
>However, it is possible for them to come up. How would a switch handle
>having two identical MAC addresses on different interfaces?
>
>It goes without saying that you're going to have problems if this happens,
>but I'm mostly curious about what the switch is going to do... ignore the
>second one it sees? Overwrite the table entry every time it gets a frame
>from the other MAC? Shutdown both interfaces? Shutdown the highest cost
>interface, believing that it's in a loop? Something else?
There is another possibility. Some switches maintain a separate MAC
table for each VLAN. On such switches, duplicate MACs on different
VLAN ports are not an issue at all. The switch keeps track of the MAC
as it pertains to each particular VLAN.
Also, duplicate MACs are not just possible with malicious users. VRRP
can in some cases reuse the same MAC address on separate VLANs, but
that depends on how it is configured.
--Bernie
| |
| Hansang Bae 2003-04-23, 7:24 pm |
| In article <b860ct$2m6$1@sun-news.laserlink.net>,
robert.watson@covad.net says...
> Legacy Hubs will have a mac table timer that may or may not flush
> automatically if it doesnt recieve traffic from the address
There seems to be some disconnect here. Hubs don't keep any state
information about MAC addresses. Some intelligent hubs could partition
ports that were chattering away, but that was about it.
> This is quite common on corporate networks that utilize a lot of laptops
> that travel from conf room and such. On your cisco ethernet/fast/gig
> switches you should never have a mac duplication due to traveling pc's
> unless you have portfast enabled. FDDI and ATM are another story though.
Not sure how to take this one. Portfast have nothing to do with mac
detection. Portfast is there to speed up the spanning tree port
forwarding.
--
hsb
"Somehow I imagined this experience would be more rewarding" Calvin
*************** USE ROT13 TO SEE MY EMAIL ADDRESS ****************
******************************
******************************
********
Due to the volume of email that I receive, I may not not be able to
reply to emails sent to my account. Please post a followup instead.
******************************
******************************
********
| |
| Hansang Bae 2003-04-23, 8:24 pm |
| > On Wed, 23 Apr 2003 02:32:09 GMT, "Knisa" <vtoneohfive@hotmail.com>
> >In theory, there shouldn't be any duplicate MAC addresses out there.
> >However, it is possible for them to come up. How would a switch handle
> >having two identical MAC addresses on different interfaces?
I forgot to add that MAC address uniqueness is applicable to catanets
(LAN segments) only.
--
hsb
"Somehow I imagined this experience would be more rewarding" Calvin
*************** USE ROT13 TO SEE MY EMAIL ADDRESS ****************
******************************
******************************
********
Due to the volume of email that I receive, I may not not be able to
reply to emails sent to my account. Please post a followup instead.
******************************
******************************
********
| |
|
| "Bernie" <Bernie@weekend.com> wrote in message
news:EFA313F9372F4D95.3A1DFD2D44B98962.7094AC32E64BDDEA@lp.airnews.net...
> On Wed, 23 Apr 2003 02:32:09 GMT, "Knisa" <vtoneohfive@hotmail.com>
> wrote:
>
> >I just finished reviewing the switching section of my CCNA Study Guide,
but
> >something occured to me that has me curious...
> >
> >In theory, there shouldn't be any duplicate MAC addresses out there.
> >However, it is possible for them to come up. How would a switch handle
> >having two identical MAC addresses on different interfaces?
> >
> >It goes without saying that you're going to have problems if this
happens,
> >but I'm mostly curious about what the switch is going to do... ignore the
> >second one it sees? Overwrite the table entry every time it gets a frame
> >from the other MAC? Shutdown both interfaces? Shutdown the highest cost
> >interface, believing that it's in a loop? Something else?
>
> There is another possibility. Some switches maintain a separate MAC
> table for each VLAN. On such switches, duplicate MACs on different
> VLAN ports are not an issue at all. The switch keeps track of the MAC
> as it pertains to each particular VLAN.
This is to fix some common issues - if 2 VLANs are connected to a device
which uses the same MAC on each port, then separate tables per VLAN are
required.
This happens with DECnet, OSI and SNA devices with multiple LAN ports fairly
frequently - those protocols often use the same MAC address on different
interfaces.
The same single MAC can occur in 2 different VLANs very easily - imagine a
VLAN network routing IP, but bridging another protocol such as SNA or
NetBEUI between VLANs. The world might be going towards all IP, but most of
it still uses other protocols as well.
Finally you can get "dup" MACs during network faults - spanning tree loops
often show up as the same MAC address arriving on different ports.
>
> Also, duplicate MACs are not just possible with malicious users. VRRP
> can in some cases reuse the same MAC address on separate VLANs, but
> that depends on how it is configured.
>
> --Bernie
--
Regards
Stephen Hope - remove xx from email to reply
|
|
|
|
|