cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
3116
Views
0
Helpful
6
Replies

Error disabled - bit error rate too high

Hi Guys,

 

I'm trying to connect a Quantum Scalar i40 Tape Library (with HP Tape Drive) to a MDS-9148.

Below are the MDS version information:

BAMDSPR01# sh version
Software
  BIOS:      version 1.0.21
  loader:    version N/A
  kickstart: version 6.2(7)
  system:    version 6.2(7)
  BIOS compile time:       01/27/14
  kickstart image file is: bootflash:///m9100-s3ek9-kickstart-mz.6.2.7.bin
  kickstart compile time:  3/12/2014 18:00:00 [03/30/2014 10:19:34]
  system image file is:    bootflash:///m9100-s3ek9-mz.6.2.7.bin
  system compile time:     3/12/2014 18:00:00 [03/30/2014 12:17:33]

  Reason: Reset due to upgrade
  System version: 5.2(1)
  Service:

 

Port Configuration:

BAMDSPR01# sh run int fc1/5

 

interface fc1/5

 

  switchport speed auto max 8000
  switchport description Scalar i40
  switchport trunk mode off
  port-license acquire
  no shutdown

 

Tape Drive Configuration:

 

Once we start the backup, interface shutdown with "Error disabled - bit error rate too high".

If we force speed to 2Gb/s on both side, then backup is fine.

Any idee where the issue could come from ?

 

Thanks & Best Regards,

Jean-Christophe.

6 Replies 6

dakester
Cisco Employee
Cisco Employee

Hi Jean-Christophe,


1.
The switch is just reporting the problem, so don't shoot the messenger.
You can verify this my moving the cable to another interface and SFP.

2.
Since it works ok at 2G, it may not be a cabling issue, but don't rule it out yet.  The cables may need to be replaced, including patch panel.

3.
Use multimode cables for multimode/shortwave optics.  Use singlemode cables for singlemode/longwave optics.
The output of the following command will let you know the SFP model and if it's short or long wave.

show interface fcx/y

4.
The distance supported at 8G is much shorter than the distance supported at 2G, so verify the distance between lasers.
http://www.cisco.com/c/en/us/products/collateral/storage-networking/mds-9000-series-multilayer-switches/product_data_sheet09186a00801bc698.html


5. I'm just curious What happens at 4G?

6. Open a case with the Tape vendor to
   check for firmware or driver bugs
   hardware issues
   verify fill word (bit pattern used between frames to keep line synchronization)
   Replace the transceiver.

6. 8G uses a different fill word (ARBFF) than 1/2/4G (IDLE). It's probably not a fill word pattern issue, but we can't rule it out.
If ARBFF does not work, try IDLE.
IF IDLE does not work, set the 8G fill-pattern back to ARBFF.

switch(config-if)# switchport fill-pattern ?
  ARBFF  Configure fill pattern ARBFF
  IDLE   Configure fill pattern IDLE
 
rtp-san-23-02-9148# show port internal info interface fc1/10 | i fill
    admin port fill-pattern (ARBFF) < default

 

 


My experience is this is an issue with the device at the other end of the cable.

If the vendor presses back, click on Open a TAC Case.

 

Regards,

David

This behaviour has been seen before; are you 100% sure, that the tape Interface is not in FC Loop mode, and therefore link initialization is failing. Setup manually FC Point to Point on the storage side.

https://supportforums.cisco.com/discussion/12326176/error-disabled-bit-error-rate-too-high-mds-switch

Hi Walter,

 

Thanks a lot for your answer.

From what I can see on the web interface, tape interface is in P2P mode.

There is no CLI interface avalaible...

I've opened a case with our reseller and update the thread once it is resolved.

Thanks.

Jean-Christophe.

Hi David,

 

Thanks a lot for your reply.

I'am pretty sure issue comes from the tape drive :-)

I have already tried to change SFPs and cables without luck, so I'm pretty sure hardware is fine.

There is not much than 3 meters between tape drive and fc switches.

@ 4G, the same issue happens.

I've opened a case with our reseller an will let you know when the case is resolved.

I'll try to change the fill world and test to see if it still fails (Right now ARBFF fill pattern is configured).

 

Once again thanks a lot for your input, it is much appreciated.

 

Hi JC

Is rate mode dedicated ?

Hi Walter,

 

Yep, rate mode is dedicated:

BAMDSPR01# show interface fc1/5
fc1/5 is up
    Port description is Scalar i40
    Hardware is Fibre Channel, SFP is short wave laser w/o OFC (SN)
    Port WWN is 20:05:54:7f:ee:05:00:10
    Admin port mode is auto, trunk mode is off
    snmp link state traps are enabled
    Port mode is F, FCID is 0x010300
    Port vsan is 1
    Speed is 2 Gbps
    Rate mode is dedicated
    Transmit B2B Credit is 8
    Receive B2B Credit is 32
    Receive data field Size is 2112
    Beacon is turned off
    admin fec state is down
    oper fec state is down
    5 minutes input rate 0 bits/sec,0 bytes/sec, 0 frames/sec
    5 minutes output rate 0 bits/sec,0 bytes/sec, 0 frames/sec
      1729841912 frames input,183341908472 bytes
        0 discards,0 errors
        0 invalid CRC/FCS,0 unknown class
        0 too long,0 too short
      109335773330 frames output,225286081576472 bytes
        15 discards,0 errors
      8 input OLS,7  LRR,0 NOS,5 loop inits
      26 output OLS,15 LRR, 19 NOS, 22 loop inits
      32 receive B2B credit remaining
      8 transmit B2B credit remaining
      8 low priority transmit B2B credit remaining
    Interface last changed at Mon Jan 12 15:17:48 2015

    Last clearing of "show interface" counters 38w 6d
BAMDSPR01#