ok. so we just upgraded a good portion of our AP’s from 7.3.6 to 8.2.2 now occasionally and very randomly we will loose IP connectivity to them . They show “down” on our monitoring system. you cant ping them or web into them using the IP address. Now if you telnet into the backhaul or another sector and ping the “down” AP it replys and shortly thereafter pingability and web access is restored. During theyre down period they continue to pass traffic as we do not recieve any complaints or reports or service loss from customers. As far as IP scheming they are local 10.x.x.x addresses. Have any of you guys seen this and/or found a solution?
Whats your region code on the AP? Are you near an Airport? What freq. are you using?
Try changing the region code to other and test to see if that resolves your problem.
they come default as other and we set them to United States. We tried them at other and they just acted strange. Were no where near an airport lol. And why would an airport cause loss of IP connectivity?
Set it back to other and see if that resolves your problem. If it does this means your dfs is kicking in because its sensing some sort of radar, i believe you regain connectivity after 15 min.
What freq. are you running?
its mostly happening to 900mhz AP’s
He says user traffic is unaffected- so the wireless link stays up (not a DFS problem).
Is the management IP address on the AP a public address accessible from the Internet? There was a known issue with the HTTP server in the radios becoming overwhelmed with random requests from the Internet and locking up. The fix was to give the radio a private, unroutable, IP address. Not sure if that was ever fixed in a software update.
yea all of our units (AP’s and SM’s) have local unroutable addresses of the 10 class Ex: 10.1.2.3
yes and the radar theory does sound a wee bit odd
When attempting to access the APs when they are unreachable, are you doing this from a host on the same network segment? I.E. are there any other devices in the way that could be causing the problem instead of the AP itself, such as the AP’s next-hop router?
we try to access from the office which is attached to our network. they dont respond to web or pings. Yet if we telnet into the backhaul to the AP we can ping said “down” AP from the backhaul and get a response.
Sounds like the problem is a device in between rather than the AP itself. Map out the entire path that the packet has to take to get from your office to the AP and back, particularly any routers. Then go get PingPlotter and trace to each device in the path until you see where the failure is occuring.