12-04-2012 07:10 AM
Hello all, I have searched for, and found, a few issues related to mine. Although, I never located a solution or root of the issue I'm having. I work for a university and we have been deploying Aruba AP 105s across our campus. We've deployed roughly 2000 APs and are about 1/3 of the way done with our deployment.
The more access points that we put out the more we're noticing this problem. We have a monitoring program called spectrum. Every morning when we come in there will be about 5 new access points that are down. When we go to the physical AP we get a blinking red power light that seems to strobe in 10 second intervals.
We are running Juniper EX 3200, 4200 & 4200 POE+ switches.
We have looked at the power settings in the Juniper switches and the APs seem to be pulling around 5.4 - 6.5 volts. I have tried to find something in the switch logs that would indicate a power surge or anything that might cause the AP to become unresponsive.
The AP itself is only responsive when a power injector is used to supply the power. Even after a factory default the Ethernet port cannot be used to power the AP back on. I have also tried to find some sort of log on the AP itself but I'm not sure there is anything of the sort on the 105s?
Any thoughts or suggestions would be appreciated. Thank you!
12-04-2012 07:48 AM
The AP-105 pulls 15.4 watts when it's working. From my Cisco 3560s:
Fa0/46 auto on 15.4 Ieee PD 3 15.4
Fa0/48 auto on 15.4 Ieee PD 3 15.4
I haven't experienced issues with the 3560s; the 4200 switches should be able to handle the 105s because according to their literature they put out at least Class 3 15.4W...the EX3200 appears to do it but only on 8 ports.
You can certainly put a lot of force on a non-POE+ switch with 24 Aruba APs. That's the total output available on a Cisco 3560, so make sure you're not putting more APs on the switch than the available POE can handle.
12-04-2012 07:54 AM
We have the switch ports set to a capability of 15.4 volts. The APs just don't seem to pull that much from it. So basically (how I understand it) is that the ports have a roof cap of 15.4 volts and whatever is plugged into the PoE port just pulls whatever voltage it needs?
We also have a maximum of 6 APs on any given switch. It's a dorm setting so there are tons of wired ports which means quite a few switches so we've tried to balance the power load between the switches available in the closet.
Thanks for your help! :)
12-04-2012 10:00 AM
This has been escalated to Aruba Networks' support team. One of our support engineers should be reaching out to you soon.
--Give Kudos: found something helpful, important, or cool? Click Kudos Star in a post.
--Problem Solved? Click "Accepted Solution" in a post.
12-17-2012 07:57 AM
We have just begun deploying AP-105 access points via POE and have had two out of 10 fail with the same symptoms in the last few days. They stop responding, then the only the red power button flash every 10 seconds. No activity at all via the console port.
These are powered from a Cisco 2960s switches.
Any more information on the problem Aruba?
12-17-2012 10:35 AM
Was there any log on the 2960 switch that contains the following error: " Power Controller reports power Imax error detected"?
If there was, I would invite you to review the following post: https://supportforums.cisco.com/thread/157276 for more information.
Another action on the Aruba side could be to upgrade to 184.108.40.206 after reviewing the release notes page 9 for resolved issues (bug # 57624)
12-17-2012 01:20 PM
There was not any errors in regard to the one you were looking for " Power Controller reports power Imax error detected"? on the switch where the two AP's have failed.
In regards to the bug fix on the new code, as far as I can tell the AP-105's did not fail after a reset or reboot. They failed while working.
Any thing else we can look at?
12-17-2012 01:23 PM
Have you found anything out on why the AP's are failing? We are just starting to roll these out and would love to know what is going on before they get deployed.
01-02-2013 12:44 PM
We have upgraded to the latest software version and since that time (around 10 days) we have not experienced any other 105 failures.
So if you have not upgraded yet you may want to do that.