No link between two SFP-10G-LRM

[FordRacer]

I have two 3750X switches that need to connect using SFP-10G-LRM GBIC over 62.5/125 OM1 MMF.  It's a short distance, the buildings are not more than 75 feet apart.  Each switch has a C3KX-NM-10G network module installed.  There is a mode conditioning 62.5/125 patch cable at each switch. show inventory on each switch shows the network module, and the GBIC's in the correct ports.  There are no 1Gbe SFP's in either switch. The fiber pair is good, I moved the existing switches to a different pair the and am using the original pair for this connection.  I have changed polarity, shut/no shut on the interfaces.  Nothing I have tried will bring up the link.  One of the switches does connect to the core using a SFP-10G-LR, it's in the second 10G SFP interface in the network module. That link is working fine. I have verified protocols on the interfaces.  When I first connected the link the amber light blinked once, then went dark. As a test i transferred the SFP's (1 Gbe) from the existing switches to the new switches and the link came up, so I know the network modules are OK.  Are there any special configurations needed for this link, force duplex or speed, shut down the Gbe interfaces?  I don't have any more of the SFP-10G-LRM GBIC's so I can't just swap them out.

[Marvin Rhoads]

They're actually SFP+ transceivers.

You may have a marginal link for purposes of reliable 10 Gbps operation.

Are you using the Cisco MCP part number CAB-MCP-LC=?

What do you see from the diagnostic command "show interface te1/1/x transceiver detail" at both ends?

[FordRacer]

We are using a Tripp Lite, mode conditioning patch cable, ST to LC. I can only connect to the one switch, I have connect directly to the other switch.  I should be able to get to it tomorrow morning.

Here is the result of the command you suggested on the easy to access switch.

ITU Channel not available (Wavelength not available),

Transceiver is internally calibrated.

mA: milliamperes, dBm: decibels (milliwatts), NA or N/A: not applicable.

++ : high alarm, +  : high warning, -  : low warning, -- : low alarm.

A2D readouts (if they differ), are reported in parentheses.

The threshold values are calibrated.

                              High Alarm  High Warn  Low Warn   Low Alarm

          Temperature         Threshold   Threshold  Threshold  Threshold

Port       (Celsius)          (Celsius)   (Celsius)  (Celsius)  (Celsius)

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

Te1/1/2     37.3                75.0        73.0        -8.0      -10.0

                              High Alarm  High Warn  Low Warn   Low Alarm

           Voltage            Threshold   Threshold  Threshold  Threshold

Port       (Volts)            (Volts)     (Volts)    (Volts)    (Volts)

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

Te1/1/2    3.24                  3.63        3.46        3.13       2.97

           Optical            High Alarm  High Warn  Low Warn   Low Alarm

           Transmit Power     Threshold   Threshold  Threshold  Threshold

Port       (dBm)              (dBm)       (dBm)      (dBm)      (dBm)

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

Te1/1/2     -1.0                 5.0         4.0        -5.0       -6.0

           Optical            High Alarm  High Warn  Low Warn   Low Alarm

           Receive Power      Threshold   Threshold  Threshold  Threshold

Port       (dBm)              (dBm)       (dBm)      (dBm)      (dBm)

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

Te1/1/2    -40.0                 0.0        -1.0       -17.0      -18.0

[Marvin Rhoads]

The cause for failure to get link is shown by the last line in your output:

Te1/1/2    -40.0                

-40.0 dBm is the noise level of the transceiver - meaning there's no light received from that strand where it plugs into the transceiver.

Anything less than -18 dBm (the "low alarn threshold") won't establish link reliably.

I'd go through and clean and re-seat the patch cables at both ends.

If you don't have a fiber cleaning kit (a Cletop for your patch cable tips and a ferrule cleaner for the SFP+ and patch panel), I'd recommend investing in a set - under $200 for the pair on Amazon.com.

[FordRacer]

It's also possible that one of the transceiver's is bad, one may not be sending or receiving correct?

[Marvin Rhoads]

Possible -yes, but much less likely. 95% (or more) of the time I see issues like this it is in the fiber cabling and not the transceiver itself.

The section just above the last one in your output indicates the one you logged into was sending at -1.0 dBm. When you get the output from the other side, it should have a similar reading.

[FordRacer]

Below is the info from the transceiver on the other side of the link.

As far as patch cables, we are using brand new Tripp Lite N422-01M.

ITU Channel not available (Wavelength not available),
Transceiver is internally calibrated.
mA: milliamperes, dBm: decibels (milliwatts), NA or N/A: not applicable.
++ : high alarm, +  : high warning, -  : low warning, -- : low alarm.
A2D readouts (if they differ), are reported in parentheses.
The threshold values are calibrated.

                              High Alarm  High Warn  Low Warn   Low Alarm
          Temperature         Threshold   Threshold  Threshold  Threshold
Port       (Celsius)          (Celsius)   (Celsius)  (Celsius)  (Celsius)
--------- ------------------  ----------  ---------  ---------  ---------
Te1/1/2     38.1                75.0        73.0        -8.0      -10.0

                              High Alarm  High Warn  Low Warn   Low Alarm
           Voltage            Threshold   Threshold  Threshold  Threshold
Port       (Volts)            (Volts)     (Volts)    (Volts)    (Volts)
---------  ---------------    ----------  ---------  ---------  ---------
Te1/1/2    3.46        +         3.63        3.46        3.13       2.97

           Optical            High Alarm  High Warn  Low Warn   Low Alarm
           Transmit Power     Threshold   Threshold  Threshold  Threshold
Port       (dBm)              (dBm)       (dBm)      (dBm)      (dBm)
---------  -----------------  ----------  ---------  ---------  ---------
Te1/1/2     -1.0                 5.0         4.0        -5.0       -6.0

           Optical            High Alarm  High Warn  Low Warn   Low Alarm
           Receive Power      Threshold   Threshold  Threshold  Threshold
Port       (dBm)              (dBm)       (dBm)      (dBm)      (dBm)
-------    -----------------  ----------  ---------  ---------  ---------
Te1/1/2    -40.0                 0.0        -1.0       -17.0      -18.0

[Marvin Rhoads]

So both ends are showing no light level on the Rx side.

99% of the time I've seen this it has been due to reversal of polarity (Tx going to Tx and Rx to Rx instead of the desired Tx-Rx + Rx-Tx).

I know you said you tried changing polarity already but could you do it once more? Remember - only change it at one end.

[FordRacer]

Yes we have changed polarity on one side only.  The boss wanted to see for himself so we messed around with it more yesterday afternoon, including a polarity change on one side.  Still no dice.

Last week we also tried looping the circuit, removing one half of the cable (TX) from the SFP and going directly into the RX side of the fiber heading out, still didn't fire up on the SFP on the receive side.

The curious thing is how, when I first connected the link I had the amber light blink once, then it went away, like I had signal for a brief moment.

I have support on these switches, it may be time for a support case.

[Marvin Rhoads]

A TAC case may help. They may just RMA you a new SFP or two.

If your cabling vendor has guaranteed the fiber plant, it may be useful to request they send out a tech to verify the end to end loss on the pairs you're using.

[S.ashok S]

Hi Marvin,

I have been facing the same issue as Fordracer.

I have changed different MC patch cords and also modules but the issue has not resolved. The same OFC fiber is working with 1G SFP.

Could you please help me to resolve the issue.

Thanks and regards,

Ashok

[Tom Randstrom]

It would be helpful to understand what fiber type OM1/2/3/4 is being used on the fiber link, what is the length of the fiber link and what the output shows for ""show interface te1/1/x transceiver detail" for both ends of the link.  

Thanx!

[S.ashok S]

Hi Tom,

Thank you for your reply. 

We are using OM1 Fiber backbone and Cisco 10GBase-LRM modules, as per Cisco document we should use CMP on both ends to support 10G on OM1 backbone.

We have cleaned OFC backbone at both ends, connected CMP cables, now it is working for 3 buildings our of 10 buildings.

We are trying to isolate the issue for other buildings too.

Thanks and regards,
Ashok

[Fezziwig]

Digging this one up from the grave, hopefully you get an e-mail notification...

What does it mean if the optical transmit power is -40 dBm?

[Marvin Rhoads]

An optical tx power reading of -40 dBm means that there is no light being transmitted. That's the noise floor of the sensor that reads the light level.