Deco X50-5G - Has become unusable after reboot (CPU: 100% / DHCP: Not working) - Possible solution
Hello.
I'm not quite sure what's happening with my X50-5G.
I've been using the 1.2.0 firmware for a while now, without a problem... Daily schedule reboot at 4am. No connection drops, nor any other problem.
But since a few days back every time the scheduled reboot happens everything goes down the drain...
The update of the firmware to the version 1.2.1 didn't help.
Symptoms:
No device can connect to the X50-5G, after the reboot. Its CPU Usages stays at 100%, but the values seem somewhat all over the place.
For instance the "Online Time". Yesterday at 8:20am it was reading: "9m". At 8:40am it was "17d 23h 15m".
On both occasions the CPU was at 100% and all devices, even those connected by Ethernet cable were unable to connect; as if the DHCP was not working and no device had been given its IP.
I do suspect the firmware (1.6.3 Build 241112112 Rel. 78921) of the X50 (3 pack), that was updated more recently, may be the cause that's affecting the entire system. At least the recent date of its implementation seems somewhat to check when the problems started. In the app I can see that the firmware was available on 27/12/2024.
After some tests; turning off all the X50 APs from the X50-5G, the problem seems to disappear. Even when the scheduled reboot happened the next day there were no problems.
I Imagine there's something going on during the 'handshake' between the X50 access points and the X50-5G, after the scheduled reboot, that crashes and overuses the CPU until not even the DHCP is working.
Once again, the fact that the System logs are constantly being cleared out doesn't help anyone. The moment you are able to connect to the router the logs are already gone. I do imagine the software team are worried about the memory used by the logs (the LTE/NSA/SA services really do create a lot of debug info), but a solution such as having the option to remotely save the logs (such as a remote network server - NAS) as many other manufactures use would be ideal.
Hey, David.
So it's okay to downgrade from the 1.2.1 Build 20241217 Rel. 63056 to the 1.2.0_build20241230_Beta?
The mesh system was all wired by Ethernet cable. I never had much luck using it in wireless mode.
Unfortunately, in my case, the problem still persists; even though with a different kind of 'flavor'.
The CPU usage seems to have stabilized. But there are still problems with the devices and connections. It seems that it doesn't differentiate between WiFi or Ethernet.
For instance, I usually have around 38-45 devices connected via the main router & mesh APs. But after the scheduled reboot, everyday at 4am, some devices aren't able to connect.
For example, my tablet is able to connect and browse the Internet, but my phone, of the same brand, can't even connect to the WiFi. One Desktop PC is able to connect, but my Synology NAS isn't - both connected via Ethernet Cable. Then there are the devices that appear on the router interface as connected, but are unreachable via their IP, ping, ssh: ESP32s, Raspberries, Intel NUC, etc.
Normal behavior:
After scheduled reboot:
In the end, I got tired of it and once again turned off all mesh APs (X50).
Immediately, even without rebooting the main router (X50-5G), the effect was evident:
I did share my logs with the technical support engineer member that reached out, by email. So let's see how that goes.
For now I have left my mesh APs disconnected, and after 2 scheduled reboots I have had no problems.
