cancel
Showing results for 
Search instead for 
Did you mean: 

Swapping fibres in VSP SMLT to EXOS LAG = SMLT port down

Swapping fibres in VSP SMLT to EXOS LAG = SMLT port down

dabbler
New Contributor

I have 2 VSPs, VSP1 and VSP2 that have SMLT fibre connections to EXOS switch stacks that use sharing.

I built a new EXOS switch stack to replace an existing one, created a new SMLT and all that worked fine. I then disconnected both fibres from the existing EXOS stack and connected both fibres from the new EXOS stack as I need to use the existing VSP SMLT ports.

The fibre on VSP2 works and provides connectivity but the fibre port on VSP1 is now down. Changing fibre cables, SFPs, rebooting the EXOS and VSP did not help.

Anyone know why moving both fibres on an SMLT connection to a new switch stack would cause this?

5 REPLIES 5

XTRMUser
Contributor

You mentioned fiber with SFP's. In my experience, you disconnect the fiber and then pull out the SFP. I wonder if you got some dirt, and need to clean the SFP and fiber ends. I don't know if you use EDM, but if you do, check for errors on the SFP ports (EXOS and VSP ends). I don't know the CLI commands for checking that.

I pulled out the fibers and SFPs and I also swapped the SFPs with spares and changed the fibres cables to ones that I knew worked OK, but nothing helped. I then swapped the 2 fibres back to the original switch stack and the same thing happened... one port on VSP1 shutdown so I thought it may be related to SMLT. Does in need a 'reset' ?

I also checked for errors using "show int gig error" but there are 0 errors on the SMLT ports.

RobertD1
Contributor II

Type show int gig state command and check reason for port down state. If you have SLPP configured and that detected a loop during the move then this could be a reason. SLPP will disable the port and needs to be re-enabled to make it work again. If you saved the VSP config and rebooted it would keep the port in an admin down state. 

On EXOS use the show sharing command to check the sharing status. Make sure the fibre ports used belong to the LAG.

Here is the output of the command, unfortunately no reason given on ports 1/3 (the new switch stack that the fibres were moved to) and 1/13 (the original stack I tried moving the fibres back to)

dabbler_0-1719566534271.png

I checked the EXOS sharing too and the ports are correct. I used the exact same config on the new stack as the original stack.

Should I maybe have shut the SMLT ports down before moving the fibres?

 

GTM-P2G8KFN