Hi everyone,

 

I could really use some help diagnosing a frustrating issue with my TP-Link setup. I recently purchased the ER605 and the EAP610 AX1800. My goal was to create a more secure network using VLAN separation for internal and guest traffic.

 

 

🔧 Setup Overview

 

• Internet connection:

  Provided by Starlink, using the Starlink Ethernet Adapter

  → Connected via LAN port from Starlink router to WAN port on ER605

 

• LAN setup:

  • ER605 Port WAN/LAN1 → PoE injector → EAP610

  • All devices managed via Omada Cloud Controller

 

 

🧱 Network Configuration

 

• SSIDs on EAP610:

  1. Internal

  2. Guest

 

• VLANs (created in Omada Controller):

  1. VLAN20 – Internal devices (192.168.20.0/24)

  2. VLAN30 – Guest network (192.168.30.0/24)

 

• Each SSID is correctly assigned to its corresponding VLAN

• Port WAN/LAN1 on ER605 is configured as a Trunk port, carrying VLAN 20 and VLAN 30 to the EAP610

• DHCP is handled by the ER605 for both VLANs

 

 

✅ What Does Work

 

• Devices in each SSID get proper IP addresses:

  • Internal → 192.168.20.x

  • Guest → 192.168.30.x

• Traffic is correctly separated

• Internet works briefly after reboot/config changes

 

 

❌ The Problem

 

After about 30–60 minutes, all clients lose internet access, even though the ER605 still seems “connected” to the Starlink router physically and the lights are flickering.

 

• EAP610 still broadcasts SSIDs

• Clients stay connected to Wi-Fi, but can’t reach the internet

• Omada Cloud Controller shows no useful logs or warnings just that the devices are no longer reachable

• Resetting ER605 temporarily restores connectivity

• Running the EAP610 in standalone mode, connected directly to Starlink (192.168.1.x), works flawlessly – no disconnects at all

 

 

🔎 Additional Notes

 

• All firmware is updated to the latest version

• The Starlink router is not in bridge mode (since Starlink doesn’t natively support it without workarounds)

• Double NAT is likely present, but shouldn’t cause a total loss of outbound access

• This seems specific to when the ER605 is in the setup, which leads me to believe something in its WAN lease renewal, routing table, or VLAN trunk config is failing over time

 

 

❓ Questions

 

1. Could the ER605 be failing to renew its WAN DHCP lease from Starlink, causing a silent drop?

2. Is it problematic to have VLAN trunking over WAN/LAN1, which is also involved in routing?

3. Should I reassign the VLAN trunk to a dedicated LAN port (e.g., LAN1), and only use the dedicated WAN port for Starlink?

4. Are there known issues with Starlink DHCP leases or incompatibility with Omada-managed ER605 setups?

 

 

🧩 Goal

 

I’d like to keep using VLANs for separation, but currently this setup isn’t stable due to the recurring internet drop. Any help or similar experiences would be appreciated — I’m close to returning the ER605 if no solution is found.

 

Thanks in advance!
Johannes