AP's disassociate

[Scott O'Brien]

Hi All,

we have a controller runiing 7.2.110.0, we have some AP's ( mainly 3502's) that keep disassociating from the controller, the AP's dont reboot or loss power, they just disassociate. there uptime does not change just the Controller Associated Time.

here is a message i got from the traps:

AP 'biochem-345-1-new', MAC: a0:cf:5b:3e:36:30 disassociated previously due to AP Reset. Uptime: 5 days, 13 h 13 m 24 s . Last reset reason: power loss.

[Amjad Abdullah]

How do you power the APs?

Sent from Cisco Technical Support iPad App

[Scott Fella]

That logs shows that it was reset due to power. If your supplying power via PoE switch, make sure you are not oversubscribing the power on that switch. Also click on the wireless tab and look at the APs listed on the bottom. Those are the latest APs to join the wlc. If the uptime is long and the join time is recent, then it's a network connectivity issue. Check your cable run, patch panel and maybe move the ap to a different port to test. Also take a look at the switch log and see if there is something wrong at the port or switch level. If APs can be isolated to a switch, well, there is your answer.

Sent from Cisco Technical Support iPhone App

[Leo Laohoo]

If your WAPs are powered up using a Cisco switch, can you please specify EXACTLY the model of your switch?

[Scott O'Brien]

the AP's are powered on via a power brick, the aps uptime is long but it is just the join time is recent.

[Scott Fella]

So that means that it must be a connectivity issue.  So from the AP ethernet port to the WLC is the issue.  It can be due to the power brick getting faulty, but the best way is to swap out the power brick and see if the issue fixes itself or not. 

[Scott O'Brien]

its happening to a few few AP's so i dought its the power brick, also the AP's are on all diffrent switches in all diffrent locations...

[Scott Fella]

Is it happening to the same AP's? If so, swap the power brick and see if the problem follows to another AP that gets powered up with that brick.

Sent from Cisco Technical Support iPad App

[Leo Laohoo]

If it's happening to a few WAPs then it's the cable that connects from the injector to the WAP.

Again, what is the EXACT model of the Cisco switches? 

[Scott O'Brien]

no consistancy on the model of switch we use 2 models mainly 2960-(24 or 48) - TS-L or WS-C3750X-48P.

its happening on both models, seems to be happening to a few to many to be the cable from the injector to the AP.

will keep tracking to see if there is any consistancy with the AP's.

[Leo Laohoo]

its happening on both models, seems to be happening to a few to many to be the cable from the injector to the AP.

There's one way to settle this.

Do you have WAP(s) that are connected to the 3750X?  Are you sure your 2960 is NOT 2960G?

If you have WAPs connected to the PoE 3750X then run the following commands:

1.  Command:  test cable tdr interface ;

2.  Wait for approximately 5 to 7 seconds;

3.  Command:  sh cable tdr interface ; and

4.  Post the output.

[Scott O'Brien]

Hi Leolaohoo,

sorry now 3750x's i tryed the cmd on a WS-C3750-48TS Poe but it only ran on gig interfaces and all the AP's run on FA.

I have looked at what AP's are doing it and it seems like they all are doing it ( we have just unter 500 AP's spread over 2 5508's)

it dosnt look like they are disconnecting the clients but they defantly drop off from the controller and come back ( they only drop off for a few seconds by the looks)