APs losing connection "The CAPWAP connection with HiveManager was lost."

[support150]

We have one Hivemanager located in our HQ, and several APs in different locations with different IP ranges.

Since Friday we can see that some of the APs in our offsite-locations, lose connection to the hivemanager and people are unable to connect to them.

Strange thing is that some APs in one offsite-location are still connected, while others are offline. So we have 2 running APs on one location, while 2 have no connection on the same location.

This is happening to us on 5 different offsite-locations, while all the APs in our HQ are still connected.

Any way to efficiently troubleshoot this?

Errors we are seeing for these lost APs are simply "The CAPWAP connection with HiveManager was lost."

Thanks

[samantha_lynn]

You can disable the debug commands by adding "no" in front of the command. For example:

no _debug capwap info

no _debug capwap basic

no _debug capwap stat

Also, a reboot of the device will disable the debugs, as an alternative.

[brian_wing]

@Sam Pirok​ once logs have been collected, how do you set the logging back to standard so we don't fill up the buffer with garbage?

[samantha_lynn · Accepted solution]

To get more details on CAPWAP issues the APs may be having, we'll want to turn on a few debugs to get more details in the buffered logs, wait for the issue to happen again, then pull tech data from the AP.

You'll want to SSH in to an AP that is disconnecting and run each of the following commands.

CAPWAP debugs:

_debug capwap info

_debug capwap basic

_debug capwap stat

Once you've enabled those, we want to wait for the AP to disconnect from the HiveManager again, then get tech data. I see you've tagged this post with Connect so the best way for you to get tech data would be via the CLI. This guide will walk you through how to do that- https://thehivecommunity.aerohive.com/s/article/Collecting-Tech-Data-via-CLI

If you could email me the results at communityhelp@aerohive.com, I can let you know what we're seeing and what we need to do to resolve this issue for you.