@Virgo 

When I was posting that I was in the process of downgrading them to 1.2.9, It's too early to tell at this point.

These units have run flawless for over a year and a half with the old 1.1.0 firmware as there was nothing else available, and it wasn't until they were upgraded to 1.2.10 that this started.

They are being POE powered from a switch. And I don't really have a way to power them any differently currently.

  @Virgo 

So I guess it didn't take long, One of the APs crashed out again on 1.2.9, rolling it back to 1.1.1, which is the version they were on for over a year and a half (I incorrectly called it 1.1.10 earlier)

Hank21 wrote

Hello @Napsterbater,

 

Thanks for reporting this issue to TP-Link business community!

 

May I know the firmware version of your controller? Do you also suffer the same issue on the EAP 1.2.9 firmware?

Just in case, please check first if there is a problem with unstable power supply, but OFC if the problem has not recurred after the downgrade, this can be ruled out as the cause.

 

Napsterbater wrote

my 2 660HD are randomly crashing and reporting disconnected, a simple reboot (POE switched off then on from switch) allows them to come online for 1-7 days roughly before crashing again.

 

 

When EAPs show "Disconnected", please check whether the EAP can still communicate with the controller, you may ping it if EAPs still get valid IP addresses.

Did you set VLAN/ACL on the controller?

Are there any abnormal statistics on EAPs?

  @Hank21 

The APs continued to crash on 1.2.9, I have reverted them to 1.1.1. I am at 2 days 1hr of uptime so a little early, but still monitoring. The APs are POE powered, not able to currently test another method, but if they stay stable on 1.1.1 that would rule it out. And of course these ran for 1.5+ years (w/reboot and power outages of course) on 1.1.1 with no issue and it didn't start until they went from 1.1.1 to 1.2.10.

Controller is on 5.13.23 Windows Server 2022

No response at ALL from the AP, even packet count from the switch port after a counter reset shows 0 packet received from the device over 15 mins, until rebooted.

The SSIDs have VLANs assigned/set, Mangement is Native VLAN "1", no EAP (or any) ACLs. I only have 2 EAP660HDs on this controller, no Swtiches Other APs or anything else.

  @Napsterbater Uptime is upto 3days 3hrs on 1.1.1, it is looking like something with 1.2.9 and 1.2.10 is causing the crashes.

  @Hank21 

Recived and replying now.

Thanks.

BTW, for anyone who finds this.

Support was useless.

No resolution, reverted the units to 1.1.1 and left them.

1.2.9 and 1.2.10 is useless.