BE65 logs showing pppd: len <0 and wandetect3 errors every 10 seconds.

BE65 logs showing pppd: len <0 and wandetect3 errors every 10 seconds.

BE65 logs showing pppd: len <0 and wandetect3 errors every 10 seconds.
BE65 logs showing pppd: len <0 and wandetect3 errors every 10 seconds.
Yesterday
Model: Deco BE65  
Hardware Version: V2
Firmware Version: Deco BE65(EU)_V2_1.3.2 Build 26040912

Hi all — hoping someone (or TP-Link support) can help with this.

 

Setup

 

My network is fully wired backhaul:

Internet connection: Full fibre 900Mbps down, 110Mbps up from Zen [UK ISP] whose network operates a PPPoE auhentication. The ONT handles the VLAN ID which is 101, so not configured in the Deco.

 

- ONT → Living Room **BE65 (main)**

- Living Room BE65 →

  - Kitchen BE65 (wired)

  - 2.5Gb switch (TL‑SG105S‑M2) →

    - Office BE65 (wired)

    - BE25 Outdoor (via Cat6A → Omada PoE 260S)

 

All Decos are wired — no wireless backhaul anywhere.

 

Symptoms and logs

 

Devices (mainly Google Nest speakers) keep reconnecting.

 

Logs show repeating errors like:

 

```

pppd: len < 0

netifd: internet: len < 0

Failed to find serving BSS

client not found in ARP table

wandetect3: fopen failed

```

 

### Pattern:

- Device connects normally  

- ~6 seconds later → errors appear  

- Then:

  - PPP errors repeat every ~10 seconds  

  - clients drop off ARP table  

  - devices reconnect  

 

This goes on constantly.

 

Checks so far:

 

Network / firmware

- All Decos on latest firmware (1.3.2)

- Full factory reset + rebuild

- Manual firmware reinstall in all devices.

 

Created a new network in the Deco app and tested as below

- Started with 1 BE65 only

- Added nodes one by one over hours

- Issue still present even with minimal setup

 

Wi-Fi / device testing

- Disabled mesh per device in the Deco app.

- Locked devices to specific nodes to try and stop them constantly flapping. 

- Forced 2.4 GHz / 5 GHz

- Moved things to the IoT network which is set to 2.4 GHz only.

 

Config changes

- DNS: ISP + 8.8.8.8

- IPv6 on/off

- QoS On/off

- MTU: 1480 / 1492 (PPPoE)

- Beamforming on/off

- Fast roaming on/off

 

Nothing changes, even with suspected trouble devices removed, the WAN never drops, packet loss is virtually zero. 13000 pings to Google and less than 4 dropped. The logs carry on showing the same every 10 seconds or so. 

---

 

## ❓ Questions

 

1. Is `pppd: len < 0` every ~10 seconds a known issue on BE65 (1.3.2)?

2. Could this be causing the BSS / ARP errors and device reconnects?

3. Is there a firmware fix or workaround available?

 

 

This is easiest to spot with Google Nest devices — they reconnect very aggressively when something isn’t right.

 

Any help or confirmation would be really appreciated.

 

  0      
0
#1
Options
2 Reply
Re:BE65 logs showing pppd: len <0 and wandetect3 errors every 10 seconds.
8 hours ago

  @carguy143 

Hi,
The mentioned log may not relate to the drop issue. It's suggested to confirm the following information for further analysis.

 

  1. How many client devices are connected to the Deco network? Do they all experience the drop phenomenon?
  2. Take the Google Nest speakers as an example, when the issue occurs, does it disconnect from the Deco's Wi-Fi and will reconnect automatically? Or is it always connected to the Deco's Wi-Fi but does not have internet?
  3. How often does the drop phenomenon occur? Do all client devices drop the connection simultaneously?

 

Best Regards

  0  
0
#2
Options
Re:BE65 logs showing pppd: len <0 and wandetect3 errors every 10 seconds.
3 hours ago

  @Solla-topee 

 

Thanks for your reply. 

 

1. Typically 25 to 30 devices over a mix of wired and wireless connections. Anything fixed in place is typically hard wired apart from the Nest Speakers, and the doorbell. 

 

2. The Google Nest speakers will sometimes state "they can't connect" "oops, there's a problem with the connection", be slow to respond to requests to cast music to them from services like YouTube music, or sometimes, one speaker in the pair will play fine, but the other will fade in and out almost like it's struggling to stay in sync with the other in the pair.

 

3. Most everything else seems to be working fine. I have been running a broadband quality monitor which pings my connection from the outside world and the only packet loss or drops that shows is when I have been changing a setting or rebooting the kit to try and stabilise it. 

 

In addition to the previous items i have done more testing

 

I disconnected all the Google speakers last night by powering them off, left them off for about 18 hours and then reconnected them to the IoT network. obviously stopped them roaming and band switching like they do (they were switching between 2.4 and 5Ghz) but, all the other logs I mention above still persist. 

 

I have tried with QoS both off and on, and even set the speeds at lower than my actual throughput by about 10% which has reduced any minor bufferbloat but as I live alone, my network isn't too stressed anyway with 900 Mbps down and 110 Mbps Upload bandwidth on the FTTP connection.  

 

Factory reset the ONT. 

 

Swapped the cable between ONT and main Deco. 

 

Left the Decos to sort themselves out with a wireless backhaul (none of the LAN cables connected) to rule out issues with any of my cabling. 

 

I manually factory reset each BE65, and swapped them around so a different one acts as the main one, directly into the ONT and also factory reset the BE25-Outdoor to see if that was causing the extra logs.

 

Today, before reconnecting the nest speakers, I factory reset each one and set it up like new, but this time on the IoT network, which is limited to 2.4 GHz only.

 

I have also today used DHCP reservation in the Deco to all devices on my network, too. 

 

Sadly none of it made any difference and the logs still show all those other errors. 

  0  
0
#3
Options