I'm seeing some odd behavior that I'm trying to isolate on my home network. And if this post belongs in a more specific manufacturer section, please lmk.
- ASUS ZenWiFi AX XT8s
- one main router with two identical meshed APs
- dedicated wired Ethernet back-haul
- latest STABLE fw version 3.0.0.4.388_21617
Periodically (and often after a router fw upgrade), when I try to ping a numeric IP address from my WiFi iPhone targeting another local wireless client (in this case, a few Raspberry Pi Zero W's), I will get "Request timed out"s ... for a while, but, then it will eventually clear up and start pinging successfully (Note: numeric IP, not a host name resolution issue). It happens less from my hard-wired clients (i.e.: laptop, Pi server, etc.) but it still happens occasionally.
I am not using the "tri-band" setup on my router (called "SmartConnect" by ASUS, band-steering elsewhere, etc.). I am using each WiFi SID (2.4Ghz, 5-1Ghz, 5-2Ghz) separately, with their own auth creds... but, it should automatically route / switch between IP clients on those bands just fine... right? They are part of the DHCP IP range.
This seems to me like a stale ARP cache issue. But, I thought reboots typically resolved that? Am I wrong? (Note: I have rebooted everything, the issue seems to eventually return) Periodic manual pinging does seem to resolve it. If this is a corrupt ARP cache, where? On the client? On the main router or nearest AP mesh node? If rebooting doesn't permanently resolve it, should I maybe truly submit an ARP clear cache cmd on the clients, SSH into the routers and do the same?
