Controllerless Networks

We have been experiencing a strange problem with our WiFi network since the last firmware update.  We will suddenly lose the ability to access the virtual controller (we are using 16 IAP-93’s), and sometimes lose internet access at one or several building’s AP’s.  If I start unplugging buildings at the main switch, eventually I end up regaining access to the virtual controller.  I can then narrow down which building(s) the problem is coming from.  As soon as I plug in the building(s) causing the problem, the GUI will almost instantly report “AP does not respond” and no longer functions at all.  Which building(s) causes the problem has not been consistent, either.  If I then go to that building(s) and physically unplug the AP’s and plug them back in, everything comes back to life- for a while.  Within a few hours or days though, the same thing happens again.  We have had this system for over a year with zero problems until this last firmware update.  That update took the system down for about an hour, and when it finally came back up, using the virtual controller GUI became much slower than it had been before.  What can we do?

Hoiw many users you got connected to the instant network?

how do you see the reourses of the IAP when you regain access? are they on 100% of cpu?

A cluster of 93 which that can go up to 16 they just can go up to 256 concurrent connectoins...

I think this is performance issue... and it goes soo high than then you get unable to access the VC... when you unplug APS then well it has less work to do :) so you regain access

Thats my guess....

Cheers

Carlos 

I have been seeing a similar issue. I have a new Instant AP network up and running with 9 ap's on it. It has no more than a few clients on it that we use for ticket scanning. Ove the past week all ap's except for the VC just disappear??? Sometimes rebooting the VC fixes the issue and sometimes it doesn't. Today it happened again and I rebooted the VC. Now the VC is unresponsive.

I can look at the dhcp server and see active leases. Some of the ap's I can ping and get to via SSH. Some I cannot?

Firmware version: 6.2.1.0-3.4.0.3_40346

HELP!

We have a very similair problem here running 6.2.1.0-3.3.0.3 code. AP's run fine for months then randomly we find that AMP registers the AP cluster as down and we have odd issues connecting. HTTPS from AMP to the VC just hangs,SSH from a terminal server connects and we can log in but any command ( ie. sh run ) just hangs the session but SSH from the local switch works fine.The SSID's seem unnafected by this so seems directly connected to the VC address in some way.All infrastructure seems unchanged and all other kit seems to be working fine. We have tried closing all switch ports except one to try proving it but even the one AP will not permit management over the VC address so in this case probably not a memory and or capacity issue.We have over 8000 AP's mostly in nests of between 4 - 8 AP's so upgrading code is a major undertaking for us.

Regards

Mike