Wireless Access

 

We are currently using 10 Gig XFP's on all of our M3 controller these are configured as static port-channels but we keep seeing a lot of input error on those interfaces , and the interesting thing is that we don't see many error on the uplinks ( Cisco WS-C6506-E )

 

 

Port Statistics
---------------
Port        PacketsIn   PacketsOut     BytesIn          BytesOut         InputErrorBytes     OutputErrorBytes CRCErrors

XG 0/10  2844422     484009        338559811     97735309          908 0 0
XG 0/11 1729616      309411        279242996     104709665        655 0 0

 

#show interface gigabitethernet 0/10 transceiver

Manufacturer : SOURCEPHOTONICS
Part Number : XPMR01CDFE , rev: 1afX
S/N : B9S2019005
Date Code : 110830 Lot:
Supported Modes : 10GBASE-LW10GBASE-LRI-64.1, LC Connector

 

#show interface gigabitethernet 0/11 transceiver

Manufacturer : SOURCEPHOTONICS
Part Number : XPXESRCDFA , rev: 1aBh
S/N : BD92005476
Date Code : 111210 Lot:
Supported Modes : 10GBASE-SW10GBASE-SR, LC Connector

 

I was wondering anybody is using 10 Gig's XFP's  and if they are seeing these errors .

 

Thanks

 

Vic

Did you ever operate this WITHOUT the port channel?

 

What is your port channel configuration on both sides?

 

Check your cable lenght - XFP standard is minimum 2 metres.

Facing Same issue

 

No Error on SWITCH side ----

 

Physical interface: xe-5/0/5, Enabled, Physical link is Up
  Interface index: 288, SNMP ifIndex: 569, Generation: 291
  Description: Aruba 6000-1
  Link-level type: Ethernet, MTU: 1514, Speed: 10Gbps, Duplex: Full-Duplex, BPDU Error: None, MAC-REWRITE Error: None,
  Loopback: Disabled, Source filtering: Disabled, Flow control: Enabled
  Device flags   : Present Running
  Interface flags: SNMP-Traps Internal: 0x4000
  Link flags     : None
  CoS queues     : 8 supported, 8 maximum usable queues
  Hold-times     : Up 0 ms, Down 0 ms
  Current address: b0:c6:9a:cd:c0:f8, Hardware address: b0:c6:9a:cd:c0:f8
  Last flapped   : 2012-11-17 21:05:47 GST (2d 16:41 ago)
  Statistics last cleared: Never
  Traffic statistics:
   Input  bytes  :          74264868477              9051728 bps
   Output bytes  :         112849233079              9213752 bps
   Input  packets:            111158543                 1360 pps
   Output packets:            168546024                 1505 pps
   IPv6 transit statistics:
    Input  bytes  :                   0
    Output bytes  :                   0
    Input  packets:                   0
    Output packets:                   0
  Input errors:
    Errors: 0, Drops: 0, Framing errors: 0, Runts: 0, Policed discards: 0, L3 incompletes: 0, L2 channel errors: 0,
    L2 mismatch timeouts: 0, FIFO errors: 0, Resource errors: 0
  Output errors:
    Carrier transitions: 1, Errors: 0, Drops: 0, Collisions: 0, Aged packets: 0, FIFO errors: 0, HS link CRC errors: 0,
    MTU errors: 0, Resource errors: 0

 

but on Aruba Controller

 

XG 0/11 is up, line protocol is up
Hardware is 10 Gigabit Ethernet, address is 00:0B:86:0E:67:8C (bia 00:0B:86:0E:67:8C)
Description: J-CORE-5/0/5
Encapsulation ARPA, loopback not set
speed (10 Gbps)
MTU 1500 bytes, BW is 10000 Mbit
Last clearing of "show interface" counters 0 day 0 hr 1 min 19 sec
link status last changed 2 day 16 hr 33 min 44 sec
    168154359 packets input, 112636015917 bytes
    Received 38950741 broadcasts, 0 runts, 0 giants, 0 throttles
    9916 input error bytes, 0 CRC, 0 frame
    30143556 multicast, 129203618 unicast
    110887173 packets output, 74091451936 bytes
    0 output errors bytes, 0 deferred
    0 collisions, 0 late collisions, 0 throttles
This port is TRUSTED

 

 

I have the same issue. Thought I was on a winner with the cable length but longer cables hasn't made a difference.

I am seeing input errors on ALL XFP modules on ALL controllers in my network and going back as far as my graphs go which is several months. So this has been occuring on multiple software versions.

 

No errors are seen on the connected switch/router. 

 

In addition two of my controllers recently started logging link down messages every 5 hours and 15 minutes. As a result VRRP is swapping and APs are reregistering. I would have thought this unrelated to the input errors except the outages sometime conincide exactly with input errors on all controllers.

 

The amount of errors is higher during business hours but not related to the amount of traffic on the controller since they occur equally on the redundant members. 

 

 

and I wonder if this is related:

 

#show datapath hardware statistics

--------------------------------------------
Aruba Mobility Processor software counters
--------------------------------------------

----------------+------------+------------+
Opcode | RX | TX |
----------------+------------+------------+
RAW/FREE | 19516418 | 28252266 |
BRIDGE | 25750229 | 0 |
ROUTE | 23555 | 0 |
SESSION | 307039 | 0 |
FORWARD | 12756615 | 0 |
APP | 311 | 0 |
LEARN | 331549 | 0 |
HWDROP | 6259 | 0 |
----------------+------------+------------+

Total drops: 6259
----------------+------------+
Reason | COUNT |
----------------+------------+
PINOT_DROP_UNKNOWN | 6259 |

 

i would suggest to raise a TAC case about this. i doubt anyone expect aruba themselves will figure this out for you.

I have rechecked my dates and these errors only started occuring after the upgrade to 6.1.3.2 which we are still running on.

Can anyone else with the errors confirm what version they are running?

I have the same code, but I have a tac case open and apparently there's a solution for this ill keep you posted

 

The TAC engineers have stated that I am using the non-approved Aruba third party XFP module ( SOURCEPHOTONICS ) and according to the document that I have attached the approved Aruba third party XFP module is Fiberxon.

 

Can you please confirm which one are you using by executing the following command :

 

show  interface gigabitethernet 0/11 transceiver

 

Thanks

 

Vic

I got a different answer. I am using the supplied optics (same manufacturer as the ones I would have used anyway!). I have told there is a fix for this in 6.1.3.4 but I have asked which bug this relates to in the release notes.

Hi Ben,

 

So you have the approved third party XFP module ?

 

Thanks

 

 

No, I have the ones they supplied with the controller, which are the exact same as the third party ones I would buy :)

Looks like upgrading to 6.1.3.5 has resolved the problem.