in an effort to try and isolate the issue i've disabled 2.4 & 5ghz on the EAP772s (still with no SSIDs) so it's running 6ghz only for mesh. Changing from 320mhz width down to the default 80 width has helped from this morning, but I've still seen 2 disconnects since doing this.

I've added pings into the diagnostics for the EAP's too.

From a wired connection into Cab-0E-4C, the CAB-2-5G_08-AD, and both EAP772's ping solid. This sugests the mesh is ok. Only LivingRoom-0E-49 looses connection.

Looks like i'm dealing with a broadcast storm. I've enabled storm control limits based on https://support.omadanetworks.com/ca/document/13281/ (which i thought i already had configured globally, but clearly was done on just on a few select ports).

time to get wireshark out!!

  @Vincent-TP 

Indeed the storm control was a different issue. partly HomeAssistant scanning an entire /21 range (it's been moved to a more suitable /25 IoT vlan) I've also changed the storm thresholds to 40mbit (40% of slowest 100mbit link). I've not see any storm issues so far since those changes.

I've only just seen your response today. Yesterday I replaced the EAP772's mesh with a network cable for further testing, which has been 100% perfect for 24hours - not a single ping loss or stability issue so i'm confident the network is good in terms of potential loops and no mis-behaving clients.

I'll put the EAP772 mesh back in place soon and run the iwconfig cmd. Does this matter when i do the iwconfig ie: does the port block issue need to be actively occuring, or can it be done anytime after i see the fault, even if it's not actively occuring?

also, can i just confirm you want this done on the root AP? based on what i've found upto now, the mesh appears to be stable (i can ping both EAP772 AP's), but the EAP772-36-80 seems to get it's wired interface dropped periodically.

  @Vincent-TP i just edited my last post that you probably didnt see: Can i just confirm you want this done on the root AP? based on what i've found upto now, the mesh appears to be stable (i can ping both EAP772 AP's reliably), but the EAP772-36-80 seems to get it's wired interface dropped periodically.

  @Vincent-TP 

Link has just failed.

BusyBox v1.20.2 (2025-03-21 11:13:28 CST) built-in shell (ash)
Enter 'help' for a list of built-in commands.

/bin $iwconfig
wifi1     no wireless extensions.

soc0      no wireless extensions.

mld-wifi0  IEEE 802.11  Mode:Master  
          RTS thr:off   Fragment thr:off
          Power Management:off
          
miireg    no wireless extensions.

soc1      no wireless extensions.

bond0     no wireless extensions.

txvap1    IEEE 802.11bea  ESSID:"TP_TX_VAP"  
          Mode:Master  Frequency:6.295 GHz  Access Point: AA:29:48:C1:61:F2   
          Bit Rate:5.7648 Gb/s   Tx-Power=21 dBm   
          RTS thr:off   Fragment thr:off
          Power Management:off
          Link Quality=94/94  Signal level=-48 dBm  Noise level=-92 dBm (BDF averaged NF value in dBm)
          Rx invalid nwid:13  Rx invalid crypt:0  Rx invalid frag:0
          Tx excessive retries:0  Invalid misc:0   Missed beacon:0

wifi0     no wireless extensions.

bkhap1    IEEE 802.11bea  ESSID:"mesh_a82948c161f0"  
          Mode:Master  Frequency:5.2 GHz  Access Point: A8:29:48:C1:61:F1   
          Bit Rate:1.4412 Gb/s   Tx-Power:21 dBm   
          RTS thr=1 B   Fragment thr:off
          Power Management:off
          Link Quality=0/94  Signal level=-92 dBm  Noise level=-92 dBm (BDF averaged NF value in dBm)
          Rx invalid nwid:2779  Rx invalid crypt:0  Rx invalid frag:0
          Tx excessive retries:0  Invalid misc:0   Missed beacon:0

wifi2     no wireless extensions.

eth0      no wireless extensions.

lo        no wireless extensions.

bkhap2    IEEE 802.11bea  ESSID:"mesh_a82948c161f0"  
          Mode:Master  Frequency:6.295 GHz  Access Point: AA:29:48:C1:61:FB   
          Bit Rate:5.7648 Gb/s   Tx-Power=21 dBm   
          RTS thr=1 B   Fragment thr:off
          Power Management:off
          Link Quality=0/94  Signal level=-92 dBm  Noise level=-92 dBm (BDF averaged NF value in dBm)
          Rx invalid nwid:5  Rx invalid crypt:0  Rx invalid frag:0
          Tx excessive retries:0  Invalid misc:0   Missed beacon:0

br0       no wireless extensions.

/bin $ 

Logs tab shows the following: 

so, i can see that it thinks vlan1 has a loop. however the ONLY thing thats changed is from: 1x 20m cable through the house between switches (same port & port profile) to: 2x cables into EAP772 & EAP772 and using the mesh to link the two switches together.

it appears the EAP772 in mesh are somehow causing a periodic loop, with the downstreams APs' wired port being disabled.

Something that has been niggling me is how i'm expecting this to work. I'm using the EAP772's in mesh (currently 6ghz only, no SSID broadcast) to act as a bridge between floors of the house. I couldnt find any reason this shouldn't work when i researched this deployment pre-purchased.

  @Vincent-TP 

Port8 is currently to the EAP772-36-80 AP wired por. Appologies, i didnt mention that change and the original map screenshot had it connected to port2. i basically used port8 to test a wired link for 24hours that produced no loop issues, and then moved the EAP772's to the same ports to eliminate any port-specific differences/issues.

Good call on disabling loopback control. I've been so fixated on the loop issue, disabling loopback didn't even cross my mind! I'll see how it behaves over the next 24hours and report back.

thanks!