We had an unusual issue with Trango 900 AP's not ARPing to the correct gateway. This only happened on sites that aggregate back to our Cisco 7200 LACs. Our LAC would host several 192.168.x.x management networks but only the Trango couldn't answer their gateway unless pointed to the primary gateway address.
For example: #Cisco LAC ip address 192.168.25.1 255.255.255.0 secondary ip address 192.168.31.1 255.255.255.0 secondary ip address 192.168.27.1 255.255.255.0 secondary ip address 192.168.33.1 255.255.255.0 secondary ip address 192.168.24.1 255.255.255.0 primary In order for a Trango AP (192.168.31.100) to answer is to configure the gateway to be 192.168.24.1 instead of 192.168.31.1. This is unusual behavior from the Trango devices. They appear to not follow the RFC or have something broken in their IP stack. On Fri, Oct 30, 2015 at 1:22 PM, Scott Vander Dussen <[email protected]> wrote: > We replaced 3x Dragonwaves in LACP with a single AF 24HD, also updated > switches on both sides with newer HP Procurves. > > > > Before: > > Me – (Procurve 1810) – (3X Dragonwave HC in LACP) - (3X Dragonwave in LACP) > – (Procurve 2848) – (APs & BHs) > > > > After: > > Me – (Procurve 1920-24G) – (Ubiquiti AirFiber 24 HD) - (Ubiquiti AirFiber 24 > HD) – (Procurve 1920-48G) – (APs & BHs) > > > > Everything works as expected with the exception of subscribers behind three > different Trango M900APs. > > The M900APs are manageable, but the connected clients are not. > > The M900AP can always ping the clients but the Procurve 1920-48G switch and > other devices further back cannot either. > > The M900AP clients also are not able to establish PPPoE sessions. > > > > Its gets weird- > > If we reboot the M900AP all the clients become manageable and establish > PPPoE connections but this only lasts unpredictably anywhere from 10 minutes > to 3-ish hours. > > > > Weirder- > > Only this tower that received the Procurve 1920-48G has problems with the > M900APs. Other towers that are backhauled off this switch and also received > new Procurve 1920-24Gs switches and have M900APs do NOT have this same > problem. > > > > I’m running out of ideas to try :/ Nothing else changed in the network. > Basically it’s something with the AF or the Procurves, I think? > > > > Incidentally, we put these 3 APs on the old Procurve 2848 and routed them > back over the Dragonwave and it worked just fine for a few days. > > > > Ideas? > > `S
