Ports on MDS9148 showing as red crosses.

[Julian.Milano]

I have 2x MDS9148 switches which are used predominately for fibre connections to servers, SANs, RecoverPoint devices, etc. I have added three servers to ports 25, 26 & 27 but when I open Device Manager GUI, those ports are showing with red crosses on both switches. I've run a license check and this is the result:

MYFIBRESWITCH250# show port-license
 Available port activation licenses are 0
-----------------------------------------------
  Interface   Cookie    Port Activation License
-----------------------------------------------
  fc1/1       16777216        acquired
  fc1/2       16781312        acquired
  fc1/3       16785408        acquired
  fc1/4       16789504        acquired
  fc1/5       16793600        acquired
  fc1/6       16797696        acquired
  fc1/7       16801792        acquired
  fc1/8       16805888        acquired
  fc1/9       16809984        acquired
  fc1/10      16814080        acquired
  fc1/11      16818176        acquired
  fc1/12      16822272        acquired
  fc1/13      16826368        acquired
  fc1/14      16830464        acquired
  fc1/15      16834560        acquired
  fc1/16      16838656        acquired
  fc1/17      16842752        acquired
  fc1/18      16846848        acquired
  fc1/19      16850944        acquired
  fc1/20      16855040        acquired
  fc1/21      16859136        acquired
  fc1/22      16863232        acquired
  fc1/23      16867328        acquired
  fc1/24      16871424        acquired
  fc1/25      16875520        acquired
  fc1/26      16879616        acquired
  fc1/27      16883712        acquired
  fc1/28      16887808        acquired
  fc1/29      16891904        acquired
  fc1/30      16896000        acquired
  fc1/31      16900096        acquired
  fc1/32      16904192        acquired
  fc1/33      16908288        ineligible
  fc1/34      16912384        ineligible
  fc1/35      16916480        eligible
  fc1/36      16920576        eligible
  fc1/37      16924672        eligible
  fc1/38      16928768        eligible
  fc1/39      16932864        eligible
  fc1/40      16936960        eligible
  fc1/41      16941056        eligible
  fc1/42      16945152        eligible
  fc1/43      16949248        eligible
  fc1/44      16953344        eligible
  fc1/45      16957440        eligible
  fc1/46      16961536        eligible
  fc1/47      16965632        eligible
  fc1/48      16969728        eligible
MYFIBRESWITCH250#

In Device Manager, when I click on the Summary tab, it lists all the ports on the switch, however I can only see ports 1 to 24 in the list.

If I click on the port and click Configure, I see a Port VSAN configured, Mode: Admin = Auto, Speed: Admin = Auto, Status: Admin = Up, Oper is Down, FailureCause is LinkFailure but I had they guys on site plug & unplug the fibre cables directly from server to switch but nothing has changed.

I'm not sure why this is happening. Is there anything I can check to see why this is happening?

[Walter Dey]

Can you please post output of

show interface fc 1/25-26

Do you have proper FC SFP's ?

[Julian.Milano]

Thanks Walter, here's the output of the command:

MYFIBRESWITCH250# show interface fc  1/25-26
fc1/25 is down (Link failure or not-connected)
    Port description is MyServer01_P0
    Hardware is Fibre Channel, SFP is short wave laser w/o OFC (SN)
    Port WWN is 20:19:54:7f:ee:cb:1f:d8
    Admin port mode is auto, trunk mode is on
    snmp link state traps are enabled
    Port vsan is 10
    Receive data field Size is 2112
    Beacon is turned off
    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
      303352111 frames input, 142769179344 bytes
        0 discards, 0 errors
        0 CRC,  0 unknown class
        0 too long, 0 too short
      271174482 frames output, 18700640324 bytes
        0 discards, 0 errors
      5 input OLS, 5 LRR, 0 NOS, 0 loop inits
      5 output OLS, 0 LRR, 10 NOS, 0 loop inits
    Interface last changed at Wed Jun 22 12:50:14 2016


fc1/26 is down (Link failure or not-connected)
    Port description is MyServer02_P0
    Hardware is Fibre Channel, SFP is short wave laser w/o OFC (SN)
    Port WWN is 20:1a:54:7f:ee:cb:1f:d8
    Admin port mode is auto, trunk mode is on
    snmp link state traps are enabled
    Port vsan is 10
    Receive data field Size is 2112
    Beacon is turned off
    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
      280062528 frames input, 19104795816 bytes
        0 discards, 0 errors
        0 CRC,  0 unknown class
        0 too long, 0 too short
      567506743 frames output, 678276845236 bytes
        0 discards, 0 errors
      2 input OLS, 2 LRR, 0 NOS, 0 loop inits
      2 output OLS, 0 LRR, 4 NOS, 0 loop inits
    Interface last changed at Wed Jun 22 12:50:12 2016


MYFIBRESWITCH250#

[Walter Dey]

Do you see any additional information in the log file ? show logging log ?

- what server ? OS ? HBA ?

- could it be that the server has a FC Loop interface ?

- is the patch cable multimode ?

- send / receive crossed ?

- the port seems to have been used before ? send /receive Counters ?

[Julian.Milano]

I assume that for logging I can click on "Logs\SHow Onboard Logs" in Device Manager?

This is what I see:

------------------------------------------------------------------------------------------------------------
 show logging onboard starttime 7/20/16-0:0:0 endtime 8/9/16-0:0: 0----------------------------
    Supervisor Module:
----------------------------


MYFIBRESWITCH250#

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

Server details are:

Model: Dell R730 (New)
OS: None (Bare Metal)
HBA: FC.Slot.5: QLogic QLE2562 8Gb 2Port Fibre Channel Adapter - 21000024FF7590FA

FC Loop Interface: Not sure how to check this.

Multi-Mode cable: Since the setup is in another country, I can only assume that the cables are multi-mode.

Send/Receive Crossed: I checked with my Sys Admin who installed the cables and he says the "prongs" are dual-channel and connected as one connector so I don't think it would be possible to cross them. The cables are wired directly from server to MDS.

Port usage: You are correct, the ports were initially connected to our RecoverPoint devices but since we had limited licenses on the ports in the MDS, we removed the RPAs and instead used the ports for the new servers. So these ports were working before.

In Device Manager, If I choose Interface\Enabled, I get a list of enabled ports from 1 to 24 only?

In Device Manager, I noticed the ports were set to Trunk so I set them to NonTrunk, tried Disable, Enable, nothing changed.

Seems that since these ports were attached to an RPA device, this may be the reason they are not working with a server connected? Can I reset the ports to "defaults"?

[Julian.Milano]

Interesting find: I asked the DC guys to swap a cable going to port 14 (from our NetApp device) to port 25 (the one we cannot get to come online) and port 25 is now live while port 14 is now down (both cables were interchanged at the MDS side).

So, it looks like the server may need to have an OS loaded with the appropriate FC drivers before its ports become alive?

[Walter Dey · Accepted solution]

So, it looks like the server may need to have an OS loaded with the appropriate FC drivers before its ports become alive?

Yes indeed !