Model :

Hardware Version :

Firmware Version :


I have built a stack with 3 units of T1700G-28TQ over a line topology. The switch hardware and firmware versions are as follows.

Hardware Version: T1700G-28TQRev2
Firmware Version: 2.0.1 Build 20180206 Rel.60432(Beta)

The stack has been set up in a LACP configuration between the stack itself and a single T1700G-28TQ at ports 1/0/28/ and 3/0/28, and ports 1/0/27 and 1/0/28, respectively, using 4 x SFP+. The 3/0/28 has been disconnected initially.

DHCP snooping security has also been enabled and these ports are configured as trusted ports in the respective LAGs.

The single unit has been powered up first. Then the unit 2 of the stack has been powered on an hour later. The stack has been successfully reconfigured for the unit 2 to join the stack and the network has been working fine.

The unit 3 is powered on 30 minutes later and the stack has been working fine.

However, when the LAG is trying to established after the 3/0/28 is connected manually. The entire network goes haywire. It is observed that the LEDs of the switches flash at a much higher frequency than normal.

Before the unit 3 has been powered on, any connected device can connected to the default gateway IP and each other. The network is considered down. If either the 1/0/28 or 3/0/28 is disconnected from the stack, the network is working fine.

Any clue on why the LAG will fail.