I’d also make sure you have enough power getting up the cable to drive it.
From: Af [mailto:af-boun...@afmug.com] On Behalf Of Ken Hohhof Sent: Thursday, July 7, 2016 9:34 AM To: af@afmug.com Subject: Re: [AFMUG] 450 AP intermittently nonresponsive,doesn't happen if no SM sessions OK, I just upgraded the AP to 14.1.2 and so far, so good. SMs next. From: Sean Heskett <mailto:af...@zirkel.us> Sent: Wednesday, July 06, 2016 5:08 PM To: af@afmug.com Subject: Re: [AFMUG] 450 AP intermittently nonresponsive, doesn't happen if no SM sessions i would upgrade the AP and all SMs to 14.1.2 (it's a very stable release) if it exhibits these issues once SMs start to register are you sure that you are not somehow creating a network loop through an SM, or some client is launching a DDoS and the subject of a DDoS attack? On Wed, Jul 6, 2016 at 3:51 PM, Ken Hohhof <af...@kwisp.com> wrote: Does this sound like any kind of a known issue? I have a 450 AP that has been in service for about a year that has started intermittently becoming nonresponsive from the Ethernet side (can't access GUI, doesn't pass traffic). I thought Ethernet problem, but neither the AP nor the switch shows any errors or events. It is more like the CPU is going to sleep or something, but the Ethernet chip is still active. I don't see anything in the Event Log though. We missed applying the Lite to Full license key on this AP and I thought maybe that was the problem since it was at 10 SMs, but I applied the license key and still see the problem. The weird thing is, if I set the Freq to None, or set the color code to something we don't use, the AP doesn't seem to exhibit the problem. Only once SMs start registering. I'm thinking it's maybe a power starvation issue, that the AP uses more power when actively talking to SMs, and maybe it's not getting enough voltage. But I also see the firmware is still on 13.2.1.3, and I see Cambium finally released 14.1.2 official. I am hesitant to upgrade the firmware though, unless there is some reason to believe what I am seeing is a firmware bug. Otherwise I am just adding to the confusion.