I have the same problem. This seems might be a kernel related problem. Plugging/unplugging a cable doesn’t seem to be noticed by the kernel (nothing in dmesg). Maybe
The only logs about the MV-88E6341 I have are the following:
[ 5.454962] libphy: mv88e6xxx SMI: probed
[ 5.458792] mdio_bus d0032004.mdio-mi: switch 0x341 probed: Marvell 88E6341, revision 0
[ 5.467432] mvneta d0030000.ethernet eth0: : detected a Marvell 88E6341 switch
[ 5.616309] Generic PHY mv88e6xxx-0:01: attached PHY driver [Generic PHY] (mii_bus:phy_addr=mv88e6xxx-0:01, irq=-1)
[ 5.700302] Generic PHY mv88e6xxx-0:02: attached PHY driver [Generic PHY] (mii_bus:phy_addr=mv88e6xxx-0:02, irq=-1)
[ 5.776342] Generic PHY mv88e6xxx-0:03: attached PHY driver [Generic PHY] (mii_bus:phy_addr=mv88e6xxx-0:03, irq=-1)
The “irq=-1” seems suspicious, but it may be normal.
Strangely, I have some interfaces (eth0, wan0@eth0, lan0@eth0, lan1@eth0)
Edit: I succeeded to have an ip manually on wan:
# ip li set up eth0
# ip li set up wan
Then, I used some dhcp on wan and it worked.
But, setting up eth0 triggered some messages from the kernel:
[ 84.485771] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
[ 86.252156] mvneta d0030000.ethernet eth0: Link is Up – 1Gbps/Full – flow control off
[ 86.260241] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[ 90.616631] dsa dsa@0 wan: Link is Down
[ 92.629555] dsa dsa@0 wan: Link is Up – 1Gbps/Full – flow control rx/tx
Finally, it’s maybe just a wrong settup from my side.