03-01-2017 01:06 PM - edited 03-08-2019 09:33 AM
Hello,
I have 4x Ruckus R500 APs with a 1200 Zone Director on version 9.13.3.0 build 22. The APs are connected to a Cisco WS-C2960X-48LPS-L POE switch on version 15.2(2)E6.
The issue is that the 4 ports that these APs are connected to randomly go down throughout the day as shown here:
Mar 1 13:03:43: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/3, changed state to down
Mar 1 13:03:47: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/3, changed state to up
Mar 1 13:03:49: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/3, changed state to up
Mar 1 13:04:14: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/3, changed state to down
Mar 1 13:04:15: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/3, changed state to down
Mar 1 13:04:18: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/3, changed state to up
Mar 1 13:04:20: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/3, changed state to up
The AP themselves do not show as rebooting but I have verified that users do lose connection when this happens. I cannot reproduce the issue and I cannot find any sort of pattern. Some days one of the ports will not drop at all but the other 3 will. Or some days it might be only one or 2 that drop randomly.
I have verified the cabling multiple times, I even pulled out brand new patch cables and used them and still get the same result. I tried moving one of the APs to an older cisco switch and left it there for a day and it did not drop at all. So that would leave me to believe it's something on that switch itself but I can't seem to track it down. There are no POE events on the switch when a port goes down. I also have other POE devices (mostly phones) that are plugged into the same switch and do not have an issue.
The only thing that stands out to me is that there are a lot STP topology changes on the AP VLAN that originate from the switch ports that the APs are connected to and they line up with when the switch ports drop but I'm not sure if that's a cause or effect of the ports dropping. I originally had spanning-tree bpdu guard and loop guard configured on the ports but I removed them from half of them but there was no change in behavior.
I did open a case with Ruckus support and they looked at the logs and said they see nothing on their side that would cause this. Unfortunately, I only carry hardware support on these switches so I can't open a TAC case.
Any help would be appreciated as I'm starting to run out of ideas of where to look.
Thanks!
03-01-2017 06:50 PM
Lets see if spanning tree might be related. On one of the ports try adding:
spanning-tree bpdufilter enable
Then see if that ports behaviour changes as compared to the others.
03-02-2017 06:58 AM
Hi Philip,
Thanks for the reply. I enabled bpdufilter on one of the AP ports. I will report back the results.
03-03-2017 06:43 AM
Hi Philip,
Unfortunately I am still seeing drops on the port with bpdufilter enabled. Doesn't look lik I'm receiving any BPDUs from the APs either: BPDU: sent 39269, received 0
03-05-2017 02:14 PM
That means the issue has nothing to do with spanning tree, so you can remove that line now.
Does the 2960 have plenty of PoE budget available?
03-02-2017 03:46 AM
Hello
Try hardcoding speed/duplex settings,enable stp portfast.
Turn off any control traffic not required (cdp,dtp,flow control etc..)
res
Paul
03-02-2017 06:52 AM
Hi Paul,
Thanks for the reply. I hardcoded the speed and duplex, enabled portfast, and turned off any control traffic on one of the APs. Unfortunately I can't manually reproduce the issue so I will report back if it drops.
03-03-2017 06:24 AM
Hi Paul,
Unfortunately I am still seeing the drops after hardcoding the speed/duplex and enabling portfast.
03-06-2017 01:35 PM
Just want to update this in case anyone else runs into it. It took me awhile to track it down but my issue was that I was hitting this bug:
https://bst.cloudapps.cisco.com/bugsearch/bug/CSCus35889/?referring_site=bugquickviewclick
The "energy-efficient ethernet" feature was causing the issue. Applying the command "no power efficient-ethernet" to the ports having the issue resolved it.
03-12-2017 05:41 PM
Well done for tracking that down. I would never have guessed that.
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: