Some background: we have a system with four Ruckus R750 APs that are meshed together. These are connected to a Cisco switch stack and the rest of our network. The DHCP is handled by the switch. The directly wired (Ethernet) network from the switch and the WiFi are the same subnet. This is okay, right?
The issue I noticed we are having is that computers that are wirelessly connected to the network are not reachable via ping or other protocols, despite being on the same subnet (10.10.46.0 /24) and even same AP, and even if Windows firewalls are temporarily taken down for testing to allow ICMP traffic. Whereas a wireless client can successfully ping a hard-wired client, interestingly. It looks like the ARP broadcasts go out but are not responded to, from what I've seen via Wireshark. Yet I'm combing through the settings on our Ruckus web interface and I can't see where this blocking/dropping is happening.
However this seems to be somewhat inconsistent. Because sometimes when devices fall asleep and come back, or are rebooted, they become reachable, and other times the opposite happens, and they lose that ability. Pinging from and to hard-wired clients seems to be much more reliable. But just when I think I find a pattern of behavior, something happens that undoes that. We also had more problems with unstable dropping wifi in the past but now that at least seems better.
I'm trying to figure out if this is a Ruckus issue or a Cisco issue (something screwing up the ARP table), and where I should look. We need this functionality and it's becoming an issue.