Early AM lag and loss of network while still connected - odd kernel errors shown - XT8

[ThatGuyinNC]

Hi all,

Apologies if this has been discussed before, but I have just recently started having issues with my XT8 mesh during the very early morning hours which prevents loading of webpages, and requires a restart of the system. I am currently running the 9.0.0.4.386_57317 beta firmware as it's been the most stable (until this issue) since 42905. When looking at the log it is almost as if it has stopped populating around the 4am time and then when I have finally either reset the system or "optimized" I find the below or similar kernel errors:
Mar 3 11:23:34 kernel: eth0 (Int switch port: 1) (Logical Port: 1) (phyId: 1f) Link DOWN.
Mar 3 11:23:37 WAN Connection: Fail to connect with some issues.
Mar 3 11:23:43 kernel: Flushing net_device wds0.0.2.
Mar 7 05:01:29 kernel: nf_conntrack: expectation table full
Mar 7 05:01:29 kernel: nf_conntrack: expectation table full
Mar 7 05:01:29 kernel: nf_conntrack: expectation table full
Mar 7 06:13:21 kernel: ^[[0;33;41m[ERROR archer] archer_mcast_activate,577: ADD_PORT: WLAN SSID has already been added: egress_port 7, current 0x0002, new 0x0002^[[0m
Mar 7 06:13:21 kernel: ^[[0;33;41m[FHW] _fhw_activate_hw ERROR: tuple_in <0x00300da7> ; flow-modification failed
Mar 7 06:13:21 kernel: ^[[0m
Mar 7 06:13:36 kernel: ^[[0;33;41m[ERROR archer] archer_mcast_activate,577: ADD_PORT: WLAN SSID has already been added: egress_port 7, current 0x0002, new 0x0002^[[0m
Mar 7 06:13:36 kernel: ^[[0;33;41m[FHW] _fhw_activate_hw ERROR: tuple_in <0x00300ccc> ; flow-modification failed
Mar 7 06:13:36 kernel: ^[[0m

My thought is that I need ANOTHER factory reset and rebuild. I was hoping someone could share if they had had an issue such as this and if perhaps if it stems from my settings (very close to default), or if my system hardware is faulty. My backhaul rate is all over the place, I have a 1GBs connection and roughly pull about 21.27 Mbps PHY and 36.46 Kbps Data on my living room node. My set up is 1 router connected to PPoE to my Fiber line, 2 wireless backhauled XT8s and 1 ethernet backhaul XT8.

I have attached the log in the event it helps diagnose. I can also share anything else if someone thinks they can assist.

Thank you so much in advance, I'm at a loss on what else I can do.

 

[dethknite]

From what I can gather in the description... do you perhaps have Backhaul Mode enabled on your main router connected to the PPoE? It sounds like you are running mixed mode with your APs.. which in my understanding is okay as long as the main router isn't set to backhaul. Setting the main router to backhaul enables all wifi bands to be used for wifi, but all APs have to then be connected via ethernet. If you do not set it to backhaul though, one of the wifi bands will be used for the AiMesh communication, and the APs will use whatever you set as priority ETH/wifi and switch if either fails. I may be way off.. but wanted to mention in case it helps.

 

[ColinTaylor]

1. You have enabled access to your router's web interface from the internet. I suggest you disable that for security reasons unless you really need it.

2. I only see two problems in the log. One at Mar 3 11:23 and another at Mar 5 08:25. In both cases it looks like there was a problem with the connection to your modem/ONT. Check your ISP equipment and the cables connected to it. Contact your ISP to see if there any faults on the line.

 

[ThatGuyinNC]

From what I can gather in the description... do you perhaps have Backhaul Mode enabled on your main router connected to the PPoE? It sounds like you are running mixed mode with your APs.. which in my understanding is okay as long as the main router isn't set to backhaul. Setting the main router to backhaul enables all wifi bands to be used for wifi, but all APs have to then be connected via ethernet. If you do not set it to backhaul though, one of the wifi bands will be used for the AiMesh communication, and the APs will use whatever you set as priority ETH/wifi and switch if either fails. I may be way off.. but wanted to mention in case it helps.

Thanks for the reply. I have one node that is ethernet backhaul but the other two nodes and the router are all utilizing the 2nd 5Ghz band, (which has it's SSID hidden). As far as the setting, it's set for wireless backhaul on all except the ethernet connected one. I have experimented with that setting and learned the hard way what that can impact.

 

[ThatGuyinNC]

1. You have enabled access to your router's web interface from the internet. I suggest you disable that for security reasons unless you really need it.

2. I only see two problems in the log. One at Mar 3 11:23 and another at Mar 5 08:25. In both cases it looks like there was a problem with the connection to your modem/ONT. Check your ISP equipment and the cables connected to it. Contact your ISP to see if there any faults on the line.

Thanks for the reply. I have the main XT8 connected directly to ONT port in my house. I have bypassed the ISPs modem on their suggestion as I am not getting my full 1Gbs connection through the XT8 system. That's a whole other issue I have recently given up on. (The ISP thought this would help, currently I pull about 600mbs at the router, I have taken it as an issue with Asus routers and 1Gbs connections that others have noted as well.) I had them out last week to review the line again and they are pulling from the connection outside around 900-950mbs. I used to pull this back in May of 2021 through the Asus XT8 but it stopped for whatever reason.

I have also tried swapping the XT8s and moving one of the others into the router position and still seem to get this weird latency around 4-5am everyday. I just purchased a Cat8 to connect the ONT to the router in hopes it was just the cable causing the issue. So you think that the ISP is having frequent issues around this time of day?

Will removing the remote connection prevent access via the app? I was considering removing this ability but the WebGUI says I will lose the ability to control via the app. Sometimes I do need to do a remote quick reboot for my wife while out and it's the quickest and easiest way.

 

[ColinTaylor]

I don't know what "weird latency" you're referring to as you didn't mention latency before. The only obvious thing in the log at about 4am is you rebooting the router at 04:07 on Mar 4.

If your ISP says they're getting full speed outside then I'd put their modem (or do you mean router?) back in service and test the speed again from a PC connected directly to that device.

If you need remote access via the app then you'll need to leave that port open.

 

[ThatGuyinNC]

I don't know what "weird latency" you're referring to as you didn't mention latency before. The only obvious thing in the log at about 4am is you rebooting the router at 04:07 on Mar 4.

If your ISP says they're getting full speed outside then I'd put their modem (or do you mean router?) back in service and test the speed again from a PC connected directly to that device.

If you need remote access via the app then you'll need to leave that port open.

sorry for the latency I was referring to the early am issue where it seems to drop network but remains connected. Usually a reboot fixes it, but it's just recently started requiring that restart in the am. I did notice some better speeds initially with the XT8 compared to the ISP router, but I haven't tested with it in a while, and not since I went from average 850mbs to 500mbs. I'll have to contact the company to get that set back up again. So those other kernel issues you see in the log aren't of concern? I just recently just started seeing these and I haven't gotten into any new firmware since the previously mentioned beta.

 

[ColinTaylor]

The nf_conntrack messages could possibly indicate a problem. But it might also be a consequence of a problem with your ISP connection. There's not a lot you can do on the router about that other than seeing if a subsequent firmware update improves things. At the moment my money is on an ISP issue.

 

[ThatGuyinNC]

The nf_conntrack messages could possibly indicate a problem. But it might also be a consequence of a problem with your ISP connection. There's not a lot you can do on the router about that other than seeing if a subsequent firmware update improves things. At the moment my money is on an ISP issue.

If it's ISP related then I gotta call them to figure out the issue.
I just had the network go down today and when looking at the system log at that time I found the below. The system log stopped populating after this too. I had to do a reboot of the system. Now it's fine, but it's very odd.

Mar 9 16:30:55 kernel: ^[[0;33;41m[ERROR archer] archer_mcast_activate,577: ADD_PORT: WLAN SSID has already been added: egress_port 6, current 0x0001, new 0x0001^[[0m
Mar 9 16:30:55 kernel: ^[[0;33;41m[FHW] _fhw_activate_hw ERROR: tuple_in <0x00300da7> ; flow-modification failed
Mar 9 16:30:55 kernel: ^[[0m
Mar 9 16:31:02 kernel: ^[[0;33;41m[ERROR archer] archer_mcast_activate,577: ADD_PORT: WLAN SSID has already been added: egress_port 6, current 0x0001, new 0x0001^[[0m
Mar 9 16:31:02 kernel: ^[[0;33;41m[FHW] _fhw_activate_hw ERROR: tuple_in <0x00300ccc> ; flow-modification failed
Mar 9 16:31:02 kernel: ^[[0m