AP's in HREAP mode disassociate from controller

[junior.beatty]

I have been having an issue with random AP3602I's in HREAP mode disassociating from the 5508 controller.  These AP's are in remote offices with 70Mb WAN back to the controller.  Randomly one or two AP's disassociate from the controller and I have to bounce the switchport to bring them back online.  The WLC is running 7.2.  Again this only occurs to one or two AP's not all of the AP's.  Has anybody experienced this problem and if so what did you do to resolve problem.

[Scott Fella]

Well what you need to look at is on the GUI... Monitor statistics ap join and see what was the reason the ap didn't join. Also, do you see any errors on the switch port? Is it with the same AP's?

Sent from Cisco Technical Support iPhone App

[Leo Laohoo]

Before you bounce the APs and when the APs would not connect to the controllers, what do you get when you run the command "sh power inline"?

Are the APs showing up as "IEEE" instead of the model number?

[George Stefanick]

Is there anything in the syslog or traps ?

Sent from Cisco Technical Support iPad App

[junior.beatty]

Thanks for the info, unfortunately I am still waiting on an AP to fail again.  Once that occurs I will follow everyones recommendations. 

[junior.beatty]

I had an AP finally disassociate from a controller.  First off I do not see the AP in the monitor statistics AP Join list, so I am unable to see why it did not join.  As for the switch port it is running clean.  When I run the "sh power inline" command, I get on the switchport the AP is attached to 15.4 watts and the model of the AP.  The only thing I see in logs either syslog or traps is the fact that the AP disassociated and the date and time when it disassociated.  This occurred on Sunday at 1:15 PM.  Nobody is at that facility at that time.  Unfortunately since this AP is at another facility an hour and a half away, I am unable to run a console connect to the AP to see what happens.  TAC suggested doing a console connect, but since it is difficult to determine which AP is going to have the problem and when it is going to have a problem this is not possible.  I will be reopening a TAC case to see if they can assist further.

[Scott Fella]

So is the AP joined to the WLC now or not? If so, click on the AP and look at the joined time and uptime. What is the value.

Sent from Cisco Technical Support iPhone App

[junior.beatty]

Thanks Scott, yeah I went ahead and bounced the switchport and it has reassociated to the WLC.  But I did just have a second AP disassociate and I found the following under Monitor Statistics AP Join

Timed out while waiting for ECHO response from AP

Unfortunately I bounced the switchport on this AP to get it back online.

[Scott Fella]

TAC should be able to see if its a bug. Sounds like it.

Sent from Cisco Technical Support iPhone App

[woodmd7]

Was there ever a bug identified for this or a solution? 

I have the same issue with multiple 3602i AP's w/ AC modules in flex-connect mode to a 5508 running 7.5.102 that have dropped association to the controller and bouncing the switchport doesn't bring them back up.  I am also getting the "

Timed out while waiting for ECHO response from AP" failure reason.

[Saravanan Lakshmanan]

Bump the AP retransmission count and interval to max - 8,5.

Make sure don't have load balancing, routing, firewall issue going on between AP and WLC.