cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
34389
Views
10
Helpful
18
Replies

100/1000 Mbs interface, auto- negotiation

sarahr202
Level 5
Level 5

Hi every body.

I was reading   about how 1000 -T intererface will default to full duplex.

Let say we have two swithes,sw1 and sw2

sw1

100/1000-T  interface--------------------------------------------------------------------------------------100-T   Sw2

Let say  on sw2,  duplex and speed are manually configured as:

speed 100

duplex full

Now will sw1 default to half duplex?

Thanks a lot.

12 Accepted Solutions

Accepted Solutions

Mohamed Sobair
Level 7
Level 7

Hi Sarah,

SW1 will  negotiate with SW2  and will be set to  100/full duplex as well.

HTH

Mohamed

View solution in original post

Mohamed Sobair
Level 7
Level 7

You asked my How it negotiates if Sw 2 has negotiation disabled?

I will ask you how SW 1 will determine its port speed of 100 while its a Gig Interface ?

My answer is that , Sw1 will negotiate the speed according to SW 2 manual settings..

HTH

Mohamed

View solution in original post

It chooses 100Mbps because it is its lowest available speed (safe choice)

It chooses Half-Duplex because it is its safest duplex choice

I do have a documentation to back this up;

http://www.cisco.com/en/US/products/hw/switches/ps700/products_tech_note09186a00800a7af0.shtml

View solution in original post

Mohamed

When auto-negotiation is disabled on one switchport the switch at the other end of the link can use the Parallel detection function to determine the speed. Put simply the switch does not negotiate the speed with the other switch but it can still work it out.

However what it cannot work out without auto-negotiation is the duplex. Now as far as i know 10/100 ethernet interfaces default to half-duplex and gigabit interfaces default to full duplex. So you could interprete this to mean the port would set itself to 100 Full because of it's default setting is gigabit. But i think that would be wrong. Once the port has negotiated down to 100Mbps it then has to default to half-duplex because that is all it can do safely as Edison says.

Jon

View solution in original post

Mohamed Sobair
Level 7
Level 7

John,

Thanks for your reply and clarification.

you said:

((When auto-negotiation is disabled on one switchport the switch at the other end of the link can use the Parallel detection function to determine the speed. Put simply the switch does not negotiate the speed with the other switch but it can still work it ou)).

I was refering to this point , as long as the other swith port speed sets to 100, the other switch negotiate it Or in a correct way it uses the parallel detection function of the other port speed. Thats why it becomes 100 Mbps port.

However, when I said its duplex of full, I didnt mean negotiating a duplex, but rather it defaults to its Full duplex settings becuse its a GIG Interface.

I still dont understand why the Gig port becomes half duplex in this situation? Do you have any reference for that ?

Mohamed

View solution in original post

Mohammed

However, when I said its duplex of full, I didnt mean negotiating a duplex, but rather it defaults to its Full duplex settings becuse its a GIG Interface.

Yes, i think that's where we interpret it differently but to be honest i can't test it because i don't have any gigabit capable switches handy. You are absolutely correct in that a gig interface defaults to full duplex as i said in my last post. But my understanding is that if the gig interface has to fall back to 100Mbps it must also fall back to half-duplex because of the speed it is running at.

However as i said in my last post it really depends on interpretation of "gig interfaces default to full duplex". And your interpretation may well be correct and mine wrong.

Jon

View solution in original post

And your interpretation may well be correct and mine wrong.

Shaking head..

View solution in original post

ediortiz wrote:

Shaking head..

View solution in original post

Reza Sharifi
Hall of Fame
Hall of Fame

Sarah/All,

I was able to test this with 2 3750-E switches.  In this scenario I am using port 1/0/48 on both switches.  The switches are called switch B and switch C

When I set the duplex to full and speed to 100 on switch C switch B's interface shows 100 Full

Here is the config and show command for switch C

Switch-C#sh run in gi1/0/48
Building configuration...

Current configuration : 63 bytes
!
interface GigabitEthernet1/0/48
speed 100
duplex full
end

Switch-C#sh int gi1/0/48
GigabitEthernet1/0/48 is up, line protocol is up (connected)
  Hardware is Gigabit Ethernet, address is 0019.e7c0.dfb0 (bia 0019.e7c0.dfb0)
  MTU 1500 bytes, BW 100000 Kbit, DLY 100 usec,
     reliability 255/255, txload 1/255, rxload 1/255
  Encapsulation ARPA, loopback not set
  Keepalive set (10 sec)
  Full-duplex, 100Mb/s, media type is 10/100/1000BaseTX
  input flow-control is off, output flow-control is unsupported
  ARP type: ARPA, ARP Timeout 04:00:00
  Last input 00:00:08, output 00:00:00, output hang never
  Last clearing of "show interface" counters never
  Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
  Queueing strategy: fifo
  Output queue: 0/40 (size/max)
  5 minute input rate 0 bits/sec, 0 packets/sec
  5 minute output rate 0 bits/sec, 0 packets/sec
     624815 packets input, 62610400 bytes, 0 no buffer
     Received 284087 broadcasts (284087 multicasts)
     0 runts, 0 giants, 0 throttles
     0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
     0 watchdog, 284087 multicast, 0 pause input
     0 input packets with dribble condition detected
     2324754 packets output, 171372318 bytes, 0 underruns
     0 output errors, 0 collisions, 0 interface resets
     0 babbles, 0 late collision, 0 deferred
     0 lost carrier, 0 no carrier, 0 PAUSE output
     0 output buffer failures, 0 output buffers swapped out
Switch-C#

Switch B config and show command

Switch-B#sh run int gi1/0/48
Building configuration...

Current configuration : 63 bytes
!
interface GigabitEthernet1/0/48
spanning-tree portfast
end

Switch-B#

Switch-B#sh interfaces gi1/0/48
GigabitEthernet1/0/48 is up, line protocol is up (connected)
  Hardware is Gigabit Ethernet, address is 0023.eb14.2e30 (bia 0023.eb14.2e30)
  MTU 9000 bytes, BW 100000 Kbit, DLY 100 usec,
     reliability 255/255, txload 1/255, rxload 1/255
  Encapsulation ARPA, loopback not set
  Keepalive set (10 sec)
Full-duplex, 100Mb/s, media type is 10/100/1000BaseTX
  input flow-control is off, output flow-control is unsupported
  ARP type: ARPA, ARP Timeout 04:00:00
  Last input 00:00:01, output 00:00:02, output hang never
  Last clearing of "show interface" counters never
  Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
  Queueing strategy: fifo
  Output queue: 0/40 (size/max)
  5 minute input rate 0 bits/sec, 0 packets/sec
  5 minute output rate 0 bits/sec, 0 packets/sec
     2324827 packets input, 171376826 bytes, 0 no buffer
     Received 1984388 broadcasts (1984388 multicasts)
     0 runts, 0 giants, 0 throttles
     0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
     0 watchdog, 1984388 multicast, 0 pause input
     0 input packets with dribble condition detected
     624837 packets output, 62612942 bytes, 0 underruns
     0 output errors, 0 collisions, 0 interface resets
     0 babbles, 0 late collision, 0 deferred
     0 lost carrier, 0 no carrier, 0 PAUSE output
     0 output buffer failures, 0 output buffers swapped out
Switch-B#

HTH

Reza

View solution in original post

Reza

Thanks very much for testing that out +5.

Mohammed

Looks like it does indeed default to 100Mbps full duplex and not half duplex so i stand corrected +5

Jon

View solution in original post

Hello Sarah,

I may be wrong but my current understanding of auto-negotation is that when it is disabled we actually disable only the modulation of the pulse train with the line code capabilities 16 bit word.

However, even if FLP would be turned off completely (as it looks like in http://en.wikipedia.org/wiki/Autonegotiation ), the ethernet frames sent by neighbor would provide its real speed (10 vs 100, as I noted 1000 Mbps can be detected by the fact that there would be activity in all 4 pairs not only on two pairs).

Each ethernet frame has a 64 bits preamble that is used to synchronize to the frame (there is no framing structure with a permanent clock like in a serial interfaces)

so neighbor speed can be sensed in one way or another.

Here we have a theorical answer and an empirical answer from Reza's tests.

We could say that humility comes from experience: many times theory and practice can give different answers, and only doing a specific test the empiric answer can be found.

The test tells us about effective implementation on network devices

Be also aware that CDP is Cisco proprietary and in any case also its standard based conteurpart LLDP it is at an upper layer in the OSI stack: a CDP frame is already an ethernet frame with a specific encapsulation (using 802.2 and SNAP with a well known multicast MAC destination)

Hope to help

Giuseppe

View solution in original post

Hello Sarah,

if your objective is to perform interoperability tests it is correct to consider Reza's results the sign of an implementation that differs from accepted standards.

>> s  why  manufactures such as Cisco  want to implement  auto-negotiation process, clearly defined in 802.3 ab/z  differently and cause unexpected results

This can be the result of a software programming error just to make an example or of a broad interpretation of standards.

With a valid service contract you could open a service request to TAC.

Current best practice for 10/100/1000 is to use autonegotiation at both sides.

  We see some times that a link to a server can negotiate at 100 Mbps or even at 10 Mbps. The way we deal with these cases is : we try to shut/unshut the port to make the two endpoints to negotiate again, if this doesn't solve we try to change a patch cable (a damaged cable can be the root cause).

Hope to help

Giuseppe

View solution in original post

18 Replies 18

Edison Ortiz
Hall of Fame
Hall of Fame

SW1 will set itself to a safe setting 100/Half Duplex since it did not receive a NWAY from SW2.

Regards

Edison

Mohamed Sobair
Level 7
Level 7

Hi Sarah,

SW1 will  negotiate with SW2  and will be set to  100/full duplex as well.

HTH

Mohamed

How SW1 would negotiate with SW2 if SW2 has negotiation disabled with static settings?

Mohamed Sobair
Level 7
Level 7

You asked my How it negotiates if Sw 2 has negotiation disabled?

I will ask you how SW 1 will determine its port speed of 100 while its a Gig Interface ?

My answer is that , Sw1 will negotiate the speed according to SW 2 manual settings..

HTH

Mohamed

It chooses 100Mbps because it is its lowest available speed (safe choice)

It chooses Half-Duplex because it is its safest duplex choice

I do have a documentation to back this up;

http://www.cisco.com/en/US/products/hw/switches/ps700/products_tech_note09186a00800a7af0.shtml

Mohamed

When auto-negotiation is disabled on one switchport the switch at the other end of the link can use the Parallel detection function to determine the speed. Put simply the switch does not negotiate the speed with the other switch but it can still work it out.

However what it cannot work out without auto-negotiation is the duplex. Now as far as i know 10/100 ethernet interfaces default to half-duplex and gigabit interfaces default to full duplex. So you could interprete this to mean the port would set itself to 100 Full because of it's default setting is gigabit. But i think that would be wrong. Once the port has negotiated down to 100Mbps it then has to default to half-duplex because that is all it can do safely as Edison says.

Jon

Mohamed Sobair
Level 7
Level 7

John,

Thanks for your reply and clarification.

you said:

((When auto-negotiation is disabled on one switchport the switch at the other end of the link can use the Parallel detection function to determine the speed. Put simply the switch does not negotiate the speed with the other switch but it can still work it ou)).

I was refering to this point , as long as the other swith port speed sets to 100, the other switch negotiate it Or in a correct way it uses the parallel detection function of the other port speed. Thats why it becomes 100 Mbps port.

However, when I said its duplex of full, I didnt mean negotiating a duplex, but rather it defaults to its Full duplex settings becuse its a GIG Interface.

I still dont understand why the Gig port becomes half duplex in this situation? Do you have any reference for that ?

Mohamed

Mohammed

However, when I said its duplex of full, I didnt mean negotiating a duplex, but rather it defaults to its Full duplex settings becuse its a GIG Interface.

Yes, i think that's where we interpret it differently but to be honest i can't test it because i don't have any gigabit capable switches handy. You are absolutely correct in that a gig interface defaults to full duplex as i said in my last post. But my understanding is that if the gig interface has to fall back to 100Mbps it must also fall back to half-duplex because of the speed it is running at.

However as i said in my last post it really depends on interpretation of "gig interfaces default to full duplex". And your interpretation may well be correct and mine wrong.

Jon

And your interpretation may well be correct and mine wrong.

Shaking head..

ediortiz wrote:

Shaking head..

Reza Sharifi
Hall of Fame
Hall of Fame

Sarah/All,

I was able to test this with 2 3750-E switches.  In this scenario I am using port 1/0/48 on both switches.  The switches are called switch B and switch C

When I set the duplex to full and speed to 100 on switch C switch B's interface shows 100 Full

Here is the config and show command for switch C

Switch-C#sh run in gi1/0/48
Building configuration...

Current configuration : 63 bytes
!
interface GigabitEthernet1/0/48
speed 100
duplex full
end

Switch-C#sh int gi1/0/48
GigabitEthernet1/0/48 is up, line protocol is up (connected)
  Hardware is Gigabit Ethernet, address is 0019.e7c0.dfb0 (bia 0019.e7c0.dfb0)
  MTU 1500 bytes, BW 100000 Kbit, DLY 100 usec,
     reliability 255/255, txload 1/255, rxload 1/255
  Encapsulation ARPA, loopback not set
  Keepalive set (10 sec)
  Full-duplex, 100Mb/s, media type is 10/100/1000BaseTX
  input flow-control is off, output flow-control is unsupported
  ARP type: ARPA, ARP Timeout 04:00:00
  Last input 00:00:08, output 00:00:00, output hang never
  Last clearing of "show interface" counters never
  Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
  Queueing strategy: fifo
  Output queue: 0/40 (size/max)
  5 minute input rate 0 bits/sec, 0 packets/sec
  5 minute output rate 0 bits/sec, 0 packets/sec
     624815 packets input, 62610400 bytes, 0 no buffer
     Received 284087 broadcasts (284087 multicasts)
     0 runts, 0 giants, 0 throttles
     0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
     0 watchdog, 284087 multicast, 0 pause input
     0 input packets with dribble condition detected
     2324754 packets output, 171372318 bytes, 0 underruns
     0 output errors, 0 collisions, 0 interface resets
     0 babbles, 0 late collision, 0 deferred
     0 lost carrier, 0 no carrier, 0 PAUSE output
     0 output buffer failures, 0 output buffers swapped out
Switch-C#

Switch B config and show command

Switch-B#sh run int gi1/0/48
Building configuration...

Current configuration : 63 bytes
!
interface GigabitEthernet1/0/48
spanning-tree portfast
end

Switch-B#

Switch-B#sh interfaces gi1/0/48
GigabitEthernet1/0/48 is up, line protocol is up (connected)
  Hardware is Gigabit Ethernet, address is 0023.eb14.2e30 (bia 0023.eb14.2e30)
  MTU 9000 bytes, BW 100000 Kbit, DLY 100 usec,
     reliability 255/255, txload 1/255, rxload 1/255
  Encapsulation ARPA, loopback not set
  Keepalive set (10 sec)
Full-duplex, 100Mb/s, media type is 10/100/1000BaseTX
  input flow-control is off, output flow-control is unsupported
  ARP type: ARPA, ARP Timeout 04:00:00
  Last input 00:00:01, output 00:00:02, output hang never
  Last clearing of "show interface" counters never
  Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
  Queueing strategy: fifo
  Output queue: 0/40 (size/max)
  5 minute input rate 0 bits/sec, 0 packets/sec
  5 minute output rate 0 bits/sec, 0 packets/sec
     2324827 packets input, 171376826 bytes, 0 no buffer
     Received 1984388 broadcasts (1984388 multicasts)
     0 runts, 0 giants, 0 throttles
     0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
     0 watchdog, 1984388 multicast, 0 pause input
     0 input packets with dribble condition detected
     624837 packets output, 62612942 bytes, 0 underruns
     0 output errors, 0 collisions, 0 interface resets
     0 babbles, 0 late collision, 0 deferred
     0 lost carrier, 0 no carrier, 0 PAUSE output
     0 output buffer failures, 0 output buffers swapped out
Switch-B#

HTH

Reza

Reza

Thanks very much for testing that out +5.

Mohammed

Looks like it does indeed default to 100Mbps full duplex and not half duplex so i stand corrected +5

Jon

Hello Jon, Reza,

neighbor speed detection is possible, detection of neighbor duplex is not possible without auto negotiation.

First of all, 1000 Mbps uses 4 pairs (all wires in RJ-45) 10/100 Mbps uses only two pairs (wires 1,2,3,6).

so if the switch sees activity on all pairs it detects neighbor is working at 1000 Mbps.

if neighbor is set to 100 Mbps or 10 Mbps  fast link pulses  allows to detect the speed.

Autonegotiation is the modulation of the pulses to carry few bits of information to tell capabilities (16 bits link code word)

http://en.wikipedia.org/wiki/Autonegotiation

Ethernet is baseband and there is no framing  structure between two ethernet frames there is silence. the fast link pulses are not ethernet frames but a form of baseband digital modulation that is interleaved over time with ethernet frames (a sort of TDM we could say).

I would say that after detecting neighbor speed is 100 Mbps a system not able to see neighbor duplex capabilities it should revert to half duplex, but Reza's tests shows something different.

Also recommendations have changed over time: years ago hardcoding  speed and duplex was recommended. Today autonegotiation is preferred.

see

http://etherealmind.com/ethernet-autonegotiation-works-why-how-standard-should-be-set/

Final note:

compliments to Sarah that putting (what looks like) simple questions make us thinking of basic concepts

Hope to help

Giuseppe

Thanks  Giuseppe for your kind remarks.     

I have one more question

Once auto-negotiation is turned off by configuring duplex and speed under the interface,  there should not be any FLP  ( fast link pulses) as auto-negotiation is off.  So in absence of FLP, what other activity would enable adjacent switch  to determine the speed? is it cdp? how about if cdp is turned off ?

Thanks and have a nice weekend.

I would like to thank  all of you who responded to my post. I  am really learning a lot  technical stuff from  all of your experiences and knowledge.

Jon  , i am really impressed with your humility.  I will try to emulate that.

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community:

Innovations in Cisco Full Stack Observability - A new webinar from Cisco