RT-AX82U - Link DOWN/Up

[goldern]

I have a Fiber Connection with an Optical Network Interface acting as the main router, before the Asus RT-AX82U, the problem is, it has no DHCP IP reservation and I don't have access to the PPPoE access credentials to setup the ONU as Bridge and use Asus RT-AX82U.

After being successful at Port Forwarding after buying an static IP from the internet provider, and also using Pi-Hole as the main DNS server, the router randomly (up to 5 times a day) disconnect everyone in the network and spits the following message in console:

Aug  8 22:50:17 kernel: eth4 (Int switch port: 4) (Logical Port: 4) (phyId: c) Link DOWN.
Aug  8 22:50:19 kernel: eth4 (Int switch port: 4) (Logical Port: 4) (phyId: c) Link Up at 1000 mbps full duplex
Aug  8 22:50:49 wlceventd: wlceventd_proc_event(511): eth6: Disassoc 50:BC:96:53:FE:10, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0
Aug  8 22:50:49 wlceventd: wlceventd_proc_event(511): eth6: Disassoc 50:BC:96:53:FE:10, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0
Aug  8 22:50:52 wlceventd: wlceventd_proc_event(530): eth6: Auth 50:BC:96:53:FE:10, status: Successful (0), rssi:0
Aug  8 22:50:52 wlceventd: wlceventd_proc_event(559): eth6: Assoc 50:BC:96:53:FE:10, status: Successful (0), rssi:-58

Which made me think that there was a problem at "eth4", which is the first port in the router, then I proceeded to disconnect the cable which was in the first port eth4 (my Intelbras camera), but even without any cable on "eth4" it still appears the same error and disconnects everyone for 1-2 minutes up to 5 times a day, even the clients connected to another router that receives the same cable as my Asus one.

That error didn't appear before and only appeared after the PPPoE was moved to the Fiber interface instead of my router (also after changing from dynamic ip to static ip), why would the internet fall for 1-2 minutes up to five times a day after supposedly a phantom cable that isn't connected being disconnected?

Also, here's another event that the same thing happened (only difference was that at this time the eth4 cable was connected):

Aug  8 19:48:42 kernel: eth4 (Int switch port: 4) (Logical Port: 4) (phyId: c) Link DOWN.
Aug  8 19:48:45 kernel: eth4 (Int switch port: 4) (Logical Port: 4) (phyId: c) Link Up at 1000 mbps full duplex
Aug  8 19:48:50 WAN Connection: WAN(0) link up.
Aug  8 19:48:50 rc_service: wanduck 24112:notify_rc restart_wan_if 0
Aug  8 19:49:24 wan: finish adding multi routes
Aug  8 22:49:24 ntp: start NTP update
Aug  8 19:49:24 kernel: Archer TCP Pure ACK Enabled
Aug  8 19:49:24 dhcp client: bound 192.168.130.2/255.255.255.0 via 192.168.130.1 for 691200 seconds.
Aug  8 19:49:26 WAN Connection: WAN was restored.

I'm scratching my head in confusion about what might be happening, any help is welcome!

 

[ColinTaylor]

Upload the complete syslog to pastebin.com and provide a link to it. Without see all the information in context it's difficult to guess what the problem is.

eth4 isn't necessarily the same as the port labelled LAN4 on the case, if fact it's very unlikely it is. The only way to know for sure which logical port matches a physical port is to plug/unplug each socket and monitor the syslog messages.

 

[goldern]

Upload the complete syslog to pastebin.com and provide a link to it. Without see all the information in context it's difficult to guess what the problem is.

eth4 isn't necessarily the same as the port labelled LAN4 on the case, if fact it's very unlikely it is. The only way to know for sure which logical port matches a physical port is to plug/unplug each socket and monitor the syslog messages.

The snippet I've posted is the entire syslog up until the error/disconnect happens, also, recently a crash happened, as it's 26000 characters I'll post it in the txt file.

Also, I've unplugged/plugged into eth4 and it shows that LAN1 is eth4, and there was no cable connected to LAN1/eth4, probably what caused the crash as it's outputting eth%d instead of eth4 as there's no cable there, also had a similar issue with the router as it wasn't exactly saving changed options even after confirm changes and reboot

I think it's a crash because it has the line "May 5 02:05:05 kernel: proc_dostring_crashlogsave: crash log filename is /jffs/crashlog.log" and the time suddenly jumped from 9th Aug to 5th May