Issues with APs disconnecting throughout the day

[Benjamin_DiehlF]

Hello, (I do apologize for my limited knowledge, just started learning about the system last week)

The school I work for is apart of the ExtremeCloud IQ Pilot system. During the day (several times a day) the dashboard will say that APs have disconnected (the normally green cloud icon will go red). It will take a minute or two for it to refresh again and say that it is connected. Checking the notifactions it says that “X” AP has lost connection to CAPWAP. I was wondering if this was a normal occurrence or if something was wrong with our Network somewhere along the line.

Physically I have checked a few APs wires and the switches and everything checks out there.

We have a range of APs models but I doubt that is the cause of the issue

[dparsons]

On a side note, does your location use IPV6? If not, you may wish and I highly recommend to filter it on the radio side to drop the broadcast packets from the clients. Also recommended is to block all traffic between the clients as this can also create extra load on the AP that can cause it to go unresponsive. Unless there is a true need for peer to peer traffic why support clients trying to negotiate IPV6 addresses with each other and other meaningless traffic.

[Benjamin_DiehlF]

Thank you, I will set this up and reply with an email as soon as I get the Data. 

[SamPirok · Accepted solution]

Hi Benjamin, thanks for the detailed explanation! My best guess is that there is just a bit of latency in the network and that the APs are not responding to the HiveManager as quickly as we’d like. There is a call and response system between the APs and ExtremeCloud IQ called echo packets. This system is in place to monitor whether the APs are responsive enough to receive updates from ExtremeCloud IQ. If the AP does not respond to an echo packet within a certain time frame, ExtremeCloud IQ will show that AP as disconnected until it does start responding again. 

We can confirm this is the issue by running CAPWAP debugs on the AP, waiting for the issue to happen again, and then pulling tech data. This guide reviews how to enable CAPWAP debugs:https://extremeportal.force.com/ExtrArticleDetail?n=000046614&q=capwap%20debug

This guide reviews how to get tech data: https://extremeportal.force.com/ExtrArticleDetail?n=000049350&q=Tech%20data%20CLI

If you want to send that tech data file to me at community@extremenetworks.com, I can review it and let you know what I find.