topic Re: AP3935 not working on Cloud IQ Controller in ExtremeCloud IQ Controller

AP3935 not working on Cloud IQ Controller

DPI — Wed, 12 Jul 2023 14:34:34 GMT

Hello,

we are moving our AP3935i from :
Wireless Controller version 10.51.24.0003
to
Cloud IQ Controller version 10.05.02.0019

AP are properly seen by XIQ-c.
Once configured in XIQ-c, AP is going to "in-service" mode, up and running with radio enabled.
But then, unexpected reboot in loop. (with various delay, from few seconds to 5 minutes)

Tested on several AP, and same behaviour.

we already have more than 500 antennas AP310 &410 running fine on XIQ-C). And i have around 500 more AP3935 to migrate, impossible to get rid of this model right now.

Any ideas ? thank you.

Already tried :
reset factory, deletion from old controllers (even if no evidence of remaining connection to old controllers)

From logs, i can see :
Configuration has been resent the maximum number times, but no acknowledge has been received. Releasing AP...

Any ideas ?
Thank you.

Re: AP3935 not working on Cloud IQ Controller

StephanH — Wed, 12 Jul 2023 20:02:31 GMT

Hello DPI,

two approaches,

try to set the MTU size to a small value. Possibly not everything from the Config is transferred successfully.
switch off the encryption of the connection between AP and controller. I assume that this was not yet on on the old controller. In the XIQ controller it is on by default.

Regards

Stephan

Re: AP3935 not working on Cloud IQ Controller

Jan_Reister — Thu, 13 Jul 2023 16:17:59 GMT

I've been there. Looks like the AP can't maintain a stable tunnel with the controller and it enters a reboot cycle.

check the XIQ-C logs, tcpdump, ap trace, CLI
make sure an initian tunnel is setup with the controller and the AP gets the config from XIQ
try lower the MTU to 1400 on both the device profile and via CLI on the AP itself;
push up the poll timeout in the device profile
set the controller ip address statically on the AP via CLI
check for packet loss and other indicators of a network issue.

In one case it was a network issue due to VRRP instability causing the gateway to flap too quickly, this caused "data port congestion" warning in the controller dashboard.

Another site had mismatched fiber/SFPs that caused connection degradation and packet loss.

Look lice older controllers like C5210 EWC and HWC and their AP firmware version were more tolerant.

Cheers,

Jan