All of my 3702e generate input errors on switchports

[Christian S.]

Hello everybody,

I started replacing our 1262n with new 3702e and found a curious problem. All switchports on different switches where the new 3702 are connected have input errors. These error counters get incremented by 1 every 30 seconds, even if there are no clients connected to the AP on this port. These ports and the cabling are ok, I can change back to a 1262 and get no more input errors, but really all of the 3702 generate 1 error every 30 seconds.

So, what's wrong with them? What about others with these APs, are your interfaces free of errors or do you experience the same problem?

I have no special config on my ports, they are just simple access ports with the capwap-vlan applied, all APs are in local mode:

Current configuration : 141 bytes

!

interface FastEthernet2/2

description apxxx

switchport access vlan xxx

switchport mode access

spanning-tree portfast

end

If others experience the same issue then there's something wrong with the code but if I'm the only one I'd take the time to do further investigations.

Thank you and kind regards,

Christian

[Rasika Nayanajith]

You are connecting these 3700 into Fastethernet ports, Try to connect these to Gig ports & see if any error occurs

100Mbps is real bottleneck & no point connecting 3700 AP to those ports.

HTH

Rasika

**** Pls rate all useful responses ****

[Christian S.]

Thank you Rasika, but I also have GE ports (same config als above) and keep getting the same input errors there, so the problem is the AP/code, not the speed of the port. What is this AP transmitting every 30 seconds that triggers an input error on the switch?

GigabitEthernet4/48 is up, line protocol is up (connected)

  Hardware is Gigabit Ethernet Port, address is 0015.fa2b.xxx (bia 0015.fa2b.xxx)

  Description: apxxx

  MTU 1500 bytes, BW 1000000 Kbit, DLY 10 usec,

     reliability 255/255, txload 1/255, rxload 1/255

  Encapsulation ARPA, loopback not set

  Keepalive set (10 sec)

  Full-duplex, 1000Mb/s, link type is auto, media type is 10/100/1000-TX

  input flow-control is off, output flow-control is off

  ARP type: ARPA, ARP Timeout 04:00:00

  Last input 00:00:12, output never, output hang never

  Last clearing of "show interface" counters 3d05h

  Input queue: 0/2000/0/0 (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 22000 bits/sec, 16 packets/sec

     417072 packets input, 90387588 bytes, 0 no buffer

     Received 23724 broadcasts (14420 multicasts)

     0 runts, 0 giants, 0 throttles

     9220 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored

     0 input packets with dribble condition detected

     3955987 packets output, 827858253 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 output buffer failures, 0 output buffers swapped out

Regards,

Christian

[Rasika Nayanajith]

Hi Christian,

I have couple of 3700 in my network running for last 2-3 weeks & do not get any input errors on the swtichport they are connected.

DS01#sh cdp nei g2/0/12 de

-------------------------

Device ID: L3702-1

Entry address(es):

  IP address: 10.15.7.41

Platform: cisco AIR-CAP3702I-Z-K9,  Capabilities: Router Trans-Bridge

Interface: GigabitEthernet2/0/12,  Port ID (outgoing port): GigabitEthernet0.1

Holdtime : 170 sec

Version :

Cisco IOS Software, C3700 Software (AP3G2-K9W8-M), Version 15.2(4)JB3, RELEASE SOFTWARE (fc1)

DS01#sh int g2/0/12

GigabitEthernet2/0/12 is up, line protocol is up (connected)

  Hardware is Gigabit Ethernet, address is 30f7.0dec.9c8c (bia 30f7.0dec.9c8c)

  Description: WAP - PW L1

  MTU 1500 bytes, BW 1000000 Kbit/sec, DLY 10 usec,

     reliability 255/255, txload 1/255, rxload 1/255

  Encapsulation ARPA, loopback not set

  Keepalive set (10 sec)

  Full-duplex, 1000Mb/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 00:00:13, output 00:00:35, output hang never

  Last clearing of "show interface" counters never

  Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 3233

  Queueing strategy: fifo

  Output queue: 0/40 (size/max)

  5 minute input rate 1000 bits/sec, 1 packets/sec

  5 minute output rate 60000 bits/sec, 45 packets/sec

     266933193 packets input, 134292697106 bytes, 0 no buffer

     Received 2488086 broadcasts (1998578 multicasts)

     0 runts, 43888 giants, 0 throttles

     0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored

     0 watchdog, 1998578 multicast, 0 pause input

     0 input packets with dribble condition detected

     1515778157 packets output, 594818778307 bytes, 0 underruns

     0 output errors, 0 collisions, 6 interface resets

     0 unknown protocol drops

     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

Please run TDR test of your AP connected switchports & see if any layer 1-2 issue with the cable used. Here are the two commands you need to run to test this

DS01#test cable-diagnostics tdr interface g2/0/12

TDR test started on interface Gi2/0/12

A TDR test can take a few seconds to run on an interface

Use 'show cable-diagnostics tdr' to read the TDR results.

DS01#show  cable-diagnostics tdr interface g2/0/12

TDR test last run on: January 25 15:01:55

Interface Speed Local pair Pair length        Remote pair Pair status

--------- ----- ---------- ------------------ ----------- --------------------

Gi2/0/12  1000M Pair A     19   +/- 10 meters Pair A      Normal             

                Pair B     19   +/- 10 meters Pair B      Normal             

                Pair C     19   +/- 10 meters Pair C      Normal             

                Pair D     19   +/- 10 meters Pair D      Normal

Refer below post done by Leo explaining in & out of this TDR process.

https://supportforums.cisco.com/docs/DOC-18983

HTH

Rasika

**** Pls rate all useful respones ****

[Christian S.]

Thank you Rasika, that's interesting. Since you don't have any errors it shouldn't be a general problem with AP or code, so I'll do further investigations.

In my opinion the cabling isn't the problem because if I connect one of the old APs to the same cable everything is fine again. Cable-diagnostics doesn't give any results (Unknown/Terminated), maybe because of the power-injector (AIR-PWRINJ4) between switch and AP? I can test it with our Fluke on monday and post the results, but to be honest if there were a L1/2 problem then there would be many input errors, not only one every 30s regardless if the AP is idling or under heavy load. Wouldn't you agree?

Which switchtype are you using Rasika? My APs are connected to 4506 running 12.2(53)SG8 with default mtu size. I'll try to reproduce it in the lab with a different switch on monday and see what's happening... Some sort of protocol is transmitting one packet from AP to LAN every 30s (which wasn't the case with the 1262's) but I have no idea what this could be. Seems as if the 4506 doesn't like it.

Thank you so far and kind regards,

Christian

[Scott Fella]

It too don't have any errors:

IL-AUR-3750X#show cdp nei g1/0/43 det

-------------------------

Device ID: CDW-3702i-01

Entry address(es):

  IP address: 192.168.201.104

Platform: cisco AIR-CAP3702I-A-K9,  Capabilities: Router Trans-Bridge

Interface: GigabitEthernet1/0/43,  Port ID (outgoing port): GigabitEthernet0.1

Holdtime : 152 sec

Version :

Cisco IOS Software, C3700 Software (AP3G2-K9W8-M), Version 15.2(4)JB3, RELEASE SOFTWARE (fc1)

Technical Support: http://www.cisco.com/techsupport

Copyright (c) 1986-2013 by Cisco Systems, Inc.

Compiled Wed 18-Dec-13 22:30 by prod_rel_team

advertisement version: 2

Duplex: full

Power drawn: 16.800 Watts

Power request id: 17184, Power management id: 2

Power request levels are:16800 15400 13000 0 0

Management address(es):

IL-AUR-3750X#show inter g1/0/43     

GigabitEthernet1/0/43 is up, line protocol is up (connected)

  Hardware is Gigabit Ethernet, address is 44d3.ca9f.58ab (bia 44d3.ca9f.58ab)

  MTU 1500 bytes, BW 1000000 Kbit/sec, DLY 10 usec,

     reliability 255/255, txload 1/255, rxload 1/255

  Encapsulation ARPA, loopback not set

  Keepalive set (10 sec)

  Full-duplex, 1000Mb/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 00:00:10, 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 3000 bits/sec, 2 packets/sec

  5 minute output rate 5000 bits/sec, 3 packets/sec

     5555102 packets input, 1132309118 bytes, 0 no buffer

     Received 69561 broadcasts (56433 multicasts)

     0 runts, 0 giants, 0 throttles

     0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored

     0 watchdog, 56433 multicast, 0 pause input

     0 input packets with dribble condition detected

     15503657 packets output, 10194425530 bytes, 0 underruns

     0 output errors, 0 collisions, 1 interface resets

     0 unknown protocol drops

     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

Thanks,

Scott

Help out other by using the rating system and marking answered questions as "Answered"

[Christian S.]

Thank you Scott, so it seems to be a problem on the 4506 which doesn't like some sort of communication from the new APs. Maybe I could get a tap between switch an power injector to see what's happening...

Regards,

Christian

[Scott Fella]

Keep us posted as this is pretty interesting:)  You have isolated it to just the 4506 correct?

Thanks,

Scott

Help out other by using the rating system and marking answered questions as "Answered"

[Christian S.]

Of course I'd keep you updated and yes, it's isolated to 4506 because now I remembered that I have one closet with a 3560 (all others are 4506) and just dialed in right now to check it - no input errors on that switch (could have thought of this earlier :-/ ). Interesting...

Regards,

Christian

[Saravanan Lakshmanan]

wlc code?

span the ap port and check the mtu size.

as test try different port speed and duplex config on 4500.

[Christian S.]

Sorry for the late response, I'm quite busy at the moment.

Since we're talking about 3k7 AP's it can only be AireOS 7.6.

If I'm not completely wrong span wouldn't make any sense because error-packets won't get mirrored.

Speed or duplex makes no difference.

I captured a 3k7 and 1260 with disabled wireless interfaces communicating with WLC using a network-tap between power-injector and switchport and found the following behavior:

Every 30 seconds the AP sends a single DTLSv1 capwap-control and one LLDP multicast, both far below mtu-limit. One of these two packets is responsible for the input-error count on the Cat4500.

The interesting thing is that both AP types send the same two packet types but only if a 3k7 is attached the switch counts an error.

Of course I could open a TAC but since everything is working fine and it only seems to be a cosmetic thing on the Cat4500 I won't do further investigations to this curious "problem".

Thanks to everybody so far,

Christian

[Rasika Nayanajith]

Hi Chris,

I am using 3750X switch for these AP connection.

Looks like it is to do with 4506 as per your observations. These 4506 modules may have different hardware/software limitation which we are not aware of . So get TAC involved if you have their support as well.

HTH

Rasika

[mlinder01]

to Scott Fella; You provide in this Post a "3750X#show cdp nei g1/0/43 det" which shows a 16.8W power draw for the Cisco AP 3702i. This is the Power Draw required at PSE (Switch) side correct?

Thanks

Manuel

[Scott Fella]

That is correct... The 3702's require more power if you want 4x4

Aironet 3700 without an add-on module
? 802.3at PoE+
? Enhanced PoE
? Cisco AP3700 power injectors (AIR-PWRINJ4=)
? Cisco AP3700 local power supply (AIR-PWR-B=)
Note: If 802.3af PoE is the source of power, the access point will dynamically shift from 4x4 to 3x3 and come up under PoE.
Aironet 3700 with an add-on module
? 802.3at PoE+
? Enhanced PoE
? Cisco AP3700 power injectors (AIR-PWRINJ4=)
? Cisco AP3700 local power supply (AIR-PWR-B=)
Note: If 802.3af PoE is the source of power, the access point with module will dynamically shift from 4x4 to 2x2 and come up under PoE.


Sent from Cisco Technical Support iPhone App

[Leo Laohoo]

Cable-diagnostics doesn't give any results (Unknown/Terminated),

That's because you are using either a 4500 or a 6500 and, to be honest, TDR feature on these two systems are a terrible joke.

If this is a 6500, it will depend on your IOS version, there are a few "line errors" bugs which are purely cosmetic.  Do you have a 1 RU switch, like a 2960S or 3750G/E/X which we can re-run the TDR on?  It doesn't matter if the switch isn't PoE.  We just want to see a more accurate TDR result.