6506 up/down ports....

[jimmysands73_2]

I work for a school district and a ticket came in late last night that perplexes me

A few gi ints and a few fa ints went into the up/down state (both on diff modules).  Total number of ints in this state are 10, they were up/up Thursday.  Sometime on Friday they went into up/down. 

I shut/no shut an int, didnt  help.

I have restarted one module, didnt help.

The log doesnt show any issues.

All the downstream switches are either 3548 XL or 2950s. 

GigabitEthernet2/1         unassigned      YES unset  up                    up

GigabitEthernet2/2         unassigned      YES unset  up                    down

GigabitEthernet2/3         unassigned      YES unset  up                    down

GigabitEthernet2/4         unassigned      YES unset  up                    up

GigabitEthernet2/5         unassigned      YES unset  up                    up

FastEthernet4/9            unassigned      YES unset  up                    up
FastEthernet4/10           unassigned      YES unset  up                    up
FastEthernet4/11           unassigned      YES unset  up                    up
FastEthernet4/12           unassigned      YES unset  up                    down
FastEthernet4/13           unassigned      YES unset  up                    down
FastEthernet4/14           unassigned      YES unset  up                    down

Here is a gi config:

0532-RTR-01#sh run int gi2/2
Building configuration...

Current configuration : 155 bytes
!
interface GigabitEthernet2/2
description SW-IDF701-01
mls qos trust dscp
switchport
switchport trunk encapsulation dot1q
switchport mode trunk
end

I will visit the site Monday to see whats going on, but any ideas?

[smogra · Accepted solution]

Hi Jimmy,

Up/Down indicates that while  the phsyical line is good - meaning you're seeing that there is a device  connected at the other side - but it is not negotiating the link  successfully.

This  could be caused by a unidirectional link of some type, a  hardware/software failure on the switch, or the other side could be  err-disabled or shutdown on the other side. This could also be a problem  if different protocols are used for trunking(ISL vs dot1q).

Hope that helps.

Cheers

Sweta

P.S. Please rate the answer if it was helpful.

[jimmysands73_2]

The odd part is that it happened to about 10 access layer switches (all of course now are now not reachable).  Its not blade specifc.  I have been trying to fix this remotely, but after shutting/no shutting the port, and also reset the module, think I will have to arrive onsite Monday to see what the heck is going on.

Jul  8 16:43:29.191: %C6KPWR-SP-4-DISABLED: power to module in slot 4 set off (Reset)
Jul  8 16:44:10.043: %DIAG-SP-6-RUN_MINIMUM: Module 4: Running Minimum Online Diagnostics...
Jul  8 16:44:18.703: %DIAG-SP-6-DIAG_OK: Module 4: Passed Online Diagnostics
Jul  8 16:44:19.043: %OIR-SP-6-INSCARD: Card inserted in slot 4, interfaces are now online

After the above, the ports in question are still up/down.  Some are fiber links, rest are ethernet.  The module above was the fa module.

I didnt reboot the whole sw, as I dont see how that would fix anything.  Though best to reboot it now or Sun vs. Monday morning I guess.  I am going to wait, I want to console into the access layer sw and see what it has to say.  I am not seeing any issues on the 6506.  Thanks for your input!

[Leo Laohoo]

Can't be UDLD because the ports would've gone into err-disable plus the 3500XL doesn't support UDLD.