01-13-2014 08:17 AM
It seems that all of our controllers are flakey running any version of 6.x but our 3200 seems to be the worst.
We were previously running 22.214.171.124 and our 3200 controller ran great. All the devices stayed connected. We upgraded to version 126.96.36.199 because of new devices that required version 6.x. We also upgraded the memory per Aruba support. Ever since then the controller has been unstable. We monitor all our devices using SolarWinds and we see the controller go down randomly. During this time, we may or may not be able to get to the controller. Most of the time when the controller is responding again all the devices connect back up. Every once in a while the devices will not come back up and show as dirty. If I make a change on the controller, such as enable and then disable spanning tree, everything comes back up.
We have had support look at the controller and our network configuration. They made some minor changes but it only seemed to help for a couple of days.
The controller says that it has not gone down and support says they could not find anything in the logs saying that it lost connection..
Any ideas of where to look?
01-13-2014 08:31 AM
Right when you lose access to the controller is there anyway to know how much memory it is using up?
How many APs are connecting to it?
ACDX #420 | ACMP
[If you found my post helpful, please give kudos!]
01-13-2014 08:42 AM
We have 11 devices connected to the controller. We are not sending syslogs to SolarWinds. I haven't been able to tell how much memory is being used. The last time the device lost connection was over the weekend whn there was little, if any use on the RAPS.
We have the following conencted:
We have seen this same type of activity on our 620 controllers but not as bad.
01-13-2014 03:29 PM
Is the controller rebooting, or losing connecitvity to the network? You can run the command "show switchinfo" to check the controller uptime.
Also, are there redundant links from the controller to the uplink switch and in that case is spanning-tree enabled?
01-13-2014 04:34 PM
It is loosing connectivity to the network.
Uptime is over 6 days
Link status just changed 20 hrs ago.
There are no redundant links. Spanning Tree is disabled
The port on our Cisco switch is a basic trunked port. It is not showing any errors.
01-13-2014 08:15 PM
When you experience this can you run show spanning-tree summary on your uplink side and see what's the status of your controller VLAN to see if it is getting blocked by stp on your uplink.
Is that VLAN just for your controller management ?
Lead Mobility Engineer @ Integration Partners
AMFX | ACMX | ACDX | ACCX | CWAP | CWDP | CWNA
01-14-2014 09:04 AM
I can try and catch it.
This vlan is for the controller and all the aruba devices. All the clients that connect to the devices also use this vlan.
01-14-2014 01:14 PM
We were at 188.8.131.52 and Aruba support had us downgrade to 184.108.40.206.
No, we are not running OSPF
Did going to 220.127.116.11 fix your issue?