Hi

I have six Deco P9 devices connected in AP mode with PLC/Wireless blackhaul.

Network works well for several years.

I bought two more Deco P9 units and want to add it to the existing mesh network of six Deco P9, totalling it to 8 devices.

I've succefully added seventh Deco P9 to the network. The issues appeared instantly.

Random Deco P9 units began to reboot without any reason for this.

I've tried to isolate this issue and it seems that network stabilized when I stopped using "Deco" Android app on the phone.

When using Deco unit web interface it shows constant almost 100% CPU load and periodically can't update the info from the units with "operation failed" error code.
I've provided browser network console log screenshot below.

Removing 7th Deco P9 unit some how fixes the situation.

It seems that you have 15 seconds timeout in your code in web interface and app for client_list, device_list requests.

If timeout is reached mobile app or web interface aborts connection for the reqest and immediately sends a new request, hammering main Deco unit with excessive load. Bescuse it seems that aborted request is still running on Deco and app is sending a new one. Slave deco units loose connect to main unit because of high load and randomly reboots to restore connection.

I've measured the web request /admin/device?form=device_list with 6 units on my network it usually takes from 11 to 13 seconds and with 7 devices it is from 16 to 19 seconds.

Same for /admin/device?form=client_list  request.

I'll be glad if you can evaluate the performance issue in the firmware and apps. 

It seems that you may need to increase request timeout in the app/web interface according to the number of devices.

Also I think it is possible to cache information for /admin/device?form=device_list and /admin/device?form=client_list calls and update it with cron every minute in the firmware OS.