Samsung TV Smart Hub error 202 over Ethernet (LAN port), but works fine over Wi-Fi on same router - CAP700

Networking Solutions Issue
1

Hello,

  • Device: ALTOS-CAP700, firmware CAP7_R_00.07.23.6

  • Connection: Telekom.de, 5G (NSA), CG-NAT (WAN IP 192.0.0.2/32)

  • LAN bridge: eth0, eth1, usb0 (STP off, IGMP off)

  • WiFi SSID: bridged to the same “lan” network/zone as Ethernet

Issue:

a Samsung Smart TV connected via Ethernet cable to a LAN port on the CAP700 gets Smart Hub error 202 (cannot connect to Smart Hub services), while the exact same TV connected via WiFi to the same router/bridge works perfectly every time.

Troubleshooting already performed:

- verified cable integrity with a cable tester — OK

- verified TV gets correct IP/gateway/DNS (192.168.1.x, gateway/DNS 192.168.1.1)

- ruled out CG-NAT as the cause: tested TV on two different mobile hotspots (Telekom.de and o2.de via iPhone) — Smart Hub worked perfectly on both, despite both also being behind CG-NAT

- confirmed WiFi and Ethernet are bridged into the same “lan” network/zone (verified in Network > Wireless > SSID config, “Network” field = lan)

- connected a laptop to the exact same Ethernet port/cable: experienced partial connectivity issues (~30% of websites, including banking sites, failed to load), while ping (ICMP) and speed tests (nPerf) were fine - classic PMTU black hole symptom

- fixed the laptop issue completely by setting “Override MTU” = 1472 on the LAN interface (matching the WAN MTU, which was already set to 1472 due to a previously diagnosed MTU mismatch with Telekom.de’s 5G network)

- after this MTU fix, the laptop has 100% working connectivity on the same port/cable. However, the Samsung TV still intermittently fails with error 202 on the same Ethernet connection.

Additional observation:

- on the TV, Smart Hub test sometimes passes initially, then fails after a failed attempt to open a streaming app (Pilot WP), and stays failed until switching back to WiFi.

Questions:

1. Is there any known difference in how the WiFi radio interface vs. the physical Ethernet ports handle MTU/fragmentation within the same LAN bridge on the CAP700?

2. Is there a way to manually control Ethernet port speed/duplex (auto-negotiation) on this model? I could not find this option anywhere in the WebUI (checked LAN interface > More > Physical Settings).

3. Are there any known compatibility issues between CAP700 and Samsung Smart TVs over wired Ethernet specifically?

Best Regards,

XynthCode

3

Hello @XynthCode,

Our apologies for the delayed response.

As additional troubleshooting steps, please try the following:

  • Navigate to Network → Firewall → Settings. Under the Routing/NAT Offloading section, disable Software Flow Offloading and test whether the issue persists.
  • Navigate to Network → Firewall → Zones, edit the LAN → WAN zone, and under General Settings, enable MSS Clamping.
  • Navigate to Network → Ports → Port Settings and verify the Speed and Duplex settings of the Ethernet port. If these options are not visible in the webUI - skiip this step for now.

Regarding your questions:

  • The Linux/OpenWrt bridge (br-lan) treats Ethernet and WiFi traffic identically at Layer 2, so the observed difference is unlikely to originate from the bridge itself.
  • Yes, it is possible, however if it’s not visible under Network → Ports → Port Settings an alternative method would be through the CLI.
  • At this time, there aren’t any known compatibility issues between the CAP700 and Samsung Smart TVs.

Please let us know whether any of these suggestions helped resolve the issue or if the behavior remains unchanged.

We appreciate your patience and look forward to your response.

Best regards,
P.

4

Hi,

I’m currently having a smiliar, almost same problem.
all of my devices connected via Ethernet to the router trough a passive switch or connected directly, cannot reach the internet. My PC shows me that he is connected, but I cant load any websites. Same on all the other devices on ethernet. WiFi works no problem.
The interesting part is, that e.g. yesterday the problem went away for a short period of time, after connecting my ethernet cable to my WAN-port set to function as LAN port in the GUI. It worked a couple of minutes on my laptop connected directly to the router. After unplugging it and plugging my PC directly into the router, it only worked for a couple of seconds after reverting back to being unable to load anything from the internet. The day before though, switching the ports on the router would enable my PC to load everything and access the internet no problem the whole time when it was switched on. After shutting it down and booting back up again, the problem came back, but solved itself by switching the port on the router.

I have already tried all that @XynthCode tried with two additional resets to factory defaults of the router. I strangely found out that I was only able to set my MTU to 1360 to net get the MTU-missmatch-message altough I’m also on Telekom.

I have now also tried your suggestions @Povilas and they haven’t had any effect on my problem.

5

Hi Glorfendor,

Thanks for the report — based on your description, this seems a bit different from my Samsung case (that one was limited to a single service being blocked via CG-NAT), while you’re losing connectivity on the entire Ethernet side, so I suspect a different root cause.

Possible causes:

  1. Stuck bridge FDB (forwarding database) or ARP entry — this fits best with the “moving the cable to a different port temporarily fixes it” symptom. It’s possible the router “remembers” your device as being on a different physical port than it actually is, causing outgoing packets to be forwarded incorrectly.
  2. Hardware Flow Offloading still active — @Povilas asked you to disable only Software Flow Offloading, but RutOS has a separate Hardware Flow Offloading option (NAT acceleration in the switch chip). If that’s still enabled, it could be “pinning” sessions to a specific port and producing exactly the behavior you’re seeing.
  3. The lower PMTU on your side (having to go down to 1360 to avoid the mismatch warning) is likely a separate issue and less likely to be the main cause of the connection drops — but it might be worth trying to set MTU to 1440 (which works stably for me) and see if that also improves things for you, though I’m not sure your network path will allow it.
  4. A partial hardware failure on one of the CAP700’s switch ports can’t be ruled out either — that could produce a similarly unstable picture.

Solutions to test (in this order):

  1. Check in the WebUI/CLI whether Hardware Flow Offloading is separately disabled (not just Software) — Network → Firewall → Settings.
  2. When the issue occurs (before moving the cable), SSH in and check:
    • bridge fdb show | grep <your MAC> — whether the bridge points to the correct physical port,
    • ip neigh show — ARP state.
      If the port in the FDB doesn’t match the physical port you’re actually connected to, that would confirm cause #1.
  3. Check ethtool -S <port> for CRC errors / link flaps on the specific port — this will help confirm or rule out a hardware fault.

Evaluating results:

It would be most useful if, during the next occurrence, you could collect the data from steps 2 and 3 before moving the cable — that would give Povilas/Justinas concrete evidence to diagnose further, instead of another round of guessing.

Let me know what those commands show.

Regards,

XynthCode