NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.

Forum Discussion

MrRussell's avatar
MrRussell
Follower
Oct 04, 2026

RS300 drops LAN traffic in a 3-second pattern for hours at a time, V1.2.7.2

RS300 on V1.2.7.2, which is the newest firmware on the support page. WAN is 2.5G to an AT&T BGW320 in IP Passthrough, so the RS300 holds the public IP. One LAN port goes to a managed switch, about 40 wired and wireless clients, all three radios on.

 

Several times a week the router drops and delays wired traffic for 45 minutes to 5 hours, then recovers on its own. During an episode:

 

- Pings from a wired host to the router's own LAN address lose 5 to 13 percent. Traffic the router forwards to the internet drops at the same instants.

- The switch and every other wired host lose 0 percent, so the LAN itself is fine.

- The loss comes in bursts every 12 or 15 seconds, locked to an exact 3.000 second clock. Each burst drops most packets for about half a second and delays the rest by about 20 ms.

- Router CPU sits at 2 to 15 percent, memory is flat, no process changes, nothing in the logs at the time.

- A reboot clears it for about a day. It has come back after every reboot.

 

I pulled Debug Log Captures during an episode, right after, and in a clean period. In the br0.pcap taken mid-episode, the echo requests that were lost never show up at all, and the ones that were delayed reach the bridge 15 to 26 ms late and are answered in under 2 ms. Forwarded traffic crosses the router in 0.14 ms and replies arriving on the WAN port show no delay. So the packets are being dropped or held on the LAN ingress side before they reach Linux, somewhere between the LAN ports and the SoC. No interface counter records the drops and the switch port facing the router shows zero discards or errors.

 

Ruled out so far: the ISP gateway and WAN link (WAN ingress measured clean mid-episode), Wi-Fi clients and roaming (same client behaviour on days with zero loss), channel interference (chanim stats identical clean vs episode), any device joining or leaving at onset (16 onsets checked), the switch and cabling (counters clean). I have not factory reset it yet.

 

This looks like the same class of problem in these threads: periodic loss on wired and Wi-Fi after weeks of running fine, the LAN side stopping to answer pings while routing continues, and in one case a warranty replacement doing the same thing:

https://community.netgear.com/discussions/nighthawk-wifi-7-be/help-please-rs300/2465442

https://community.netgear.com/discussions/nighthawk-wifi-7-be/rs300-lan-port-not-responding/2405310

https://community.netgear.com/discussions/nighthawk-wifi-7-be/rs300-wifi-lag-problems/2427333

 

Is this a known issue with the LAN switch block or the internal link to the SoC on the RS-series? If anyone from NETGEAR wants the three debug captures and the loss graph, I can send them by PM.

1 Reply

  • FURRYe38's avatar
    FURRYe38
    Guru - Experienced User

    Name and brand of the managed switch? 

    Does this happen on all LAN ports?

    Does this happen with a non managed switch? 

    CAT# LAN cable connected between the RS router and this managed switch? 

    Has a factory reset and setup from scratch been performed since last FW update? 

     

    Have you opened a support case with NG support about this?