SFP module strange behavior

Unanswered Question
Jul 25th, 2010

Hello!

I have cisco-finisar 1Gig module installed in SFP port in RSP720-3CXL-GE.

Here is some diag:

core2#sh idprom interface gigabitEthernet 6/1 detail
<cut>
Transceiver Serial EEPROM Contents:
Common block:
Identifier        :
Connector         :  [0x07]
Transceiver
  Speed            : [0x00]
  Media            : [0x00]
  Technology       : [0x00]
  Link Length      : [0x02]
  GE Comp Codes    : 1000BASE-LX [0x12]
  SONET Comp Codes : [0x00]
Encoding          : 8B10B [0x01]
BR, Nominal       : 21x100 MHz [0x15]
Length(9u)        : 100 x 100 m [0x64]
Length(50u)       : 0 x 10 m [0x00]
Length(62.5u)     : 0 x 10 m [0x00]
Length(Copper)    : 0 m [0x00]
Vendor Name       : CISCO-FINISAR
Vendor OUI        : 0x0 0x90 0x65
Vendor PN         : FTRJ1319P1BTL-C5
Vendor rev        : 10.0
CC_BASE           : 0x99
Extended ID Fields
Options           : Loss of Signal implemented TX_DISABLE is implemented and disables the serial output
BR, max           : Unspecified
BR, min           : Unspecified
Vendor SN         : FNS11360P14
Date code         : 061212
CC_EXT            : 0x8D

The strange thing about it is that when observing transiever status I see the flap of that kind:

core2#sh interfaces transceiver | i Gi6/1
Gi6/1       0.0       0.00       0.0     -40.0     -40.0
core2#sh interfaces transceiver | i Gi6/1
Gi6/1       0.7       0.01 --    0.3 --   N/A ++  -40.0

I could guess that this is the power problem, perhaps sfp module installed badly or the transmitter is out of service.

But..the magic is that the interface is operational, trafic flows and no threshold messages are dropped into log:

core2#sh int gigabitEthernet 6/1
GigabitEthernet6/1 is up, line protocol is up (connected)
  Hardware is c7600 1Gb 802.3, address is 001c.584c.556c (bia 001c.584c.556c)
  Description: peer
  MTU 9000 bytes, BW 1000000 Kbit, DLY 10 usec,
     reliability 255/255, txload 46/255, rxload 70/255
  Encapsulation ARPA, loopback not set
  Keepalive set (10 sec)
  Full-duplex, 1000Mb/s, media type is LH
  input flow-control is off, output flow-control is off
  Clock mode is auto
  ARP type: ARPA, ARP Timeout 04:00:00
  Last input never, output never, output hang never
  Last clearing of "show interface" counters 17w6d
  Input queue: 0/2000/1255/0 (size/max/drops/flushes); Total output drops: 118
  Queueing strategy: fifo
  Output queue: 0/40 (size/max)
  5 minute input rate 276684000 bits/sec, 47459 packets/sec
  5 minute output rate 183261000 bits/sec, 50184 packets/sec
     946156131401 packets input, 670142539615934 bytes, 0 no buffer
     Received 181011 broadcasts (179518 multicasts)
     0 runts, 0 giants, 0 throttles
     25 input errors, 22 CRC, 3 frame, 1255 overrun, 0 ignored
     0 watchdog, 0 multicast, 0 pause input
     0 input packets with dribble condition detected
     1020707481696 packets output, 452456257677017 bytes, 0 underruns
     0 output errors, 0 collisions, 5 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


So...is it some bug with all finisar modules, or it is a module frome a corrupted production set, or it is an IOS bug of "sh interface transiever"  command ?

Any ideas?

Thanks in advance.

I have this problem too.
0 votes
  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
martin_knorre Thu, 08/26/2010 - 01:26

Hi,

I had a similar problem with a 2960 Switch, I tried it with different SFP modules with the result of, the switch is broken. Can you try more SFP modules? Do you use GLC-T or SX-MM for the connection? I hope that only the Gbic os broken.

Regards Martin

Alexander Demin Thu, 08/26/2010 - 05:08

From  my post it is evident that I use either SM or MM capable trnasiever (1000BASE-LX/LH)

My trnasiever is anyway operational though indication is incorrect.

Perhaps, you (and I) should look for other than Finisar transiever vendor, compartible with cisco

Look for where the transiever was manufactured. Better it is Japan, not Malayasia or China.

ARMANDO ALVARADO Mon, 08/30/2010 - 13:17

These are the SFP  modules we use on all our switches GLC-SX-MM the ones for LH is for long haul for long distances. If you need a part number I can probably get that for you. Hopes this helps.

Actions

This Discussion