I finally narrowed this down to a problem with all CRS210-8G-2S+IN Routerboards that I have (around 40 of them).

With a small percentage of devices, Port 1, and ONLY Port 1 (PoE port) will refuse to register at the full 1Gbps and drops back to 100Mbps FDX after a few seconds of on/off negotiation.

I can plug the same offending laptop into any other of the ports 2-8 and it doesn't have a problem immediately registering 1Gbps FDX. Laptop/device/router doesn't have any other known issues with any other device.

Doesn't matter if I'm using PoE or direct DC power, same issue.

Very very consistent, and the number of devices that have the issue with it are steadily growing.

I just wanted to check here if anyone else had noticed this.

Also, if by any chance it's a default configuration error let me know.

This is on the latest RouterOS I had loaded recently 6.30.2, but also occurs on 6.27 on the same device (upgrade, downgrade, etc, always the same failure).

I have the same problem. Port 1 only connects to 100Mbit after about 10 sec. Other ports connect to 1gbit with same device and cable.

CRS210, 6.34.1

there was a bugfix in the latest firmware for this

i did had a problem with the crs-112, i had packet loss on port 1, i cant check if it is solved with the new firmware since it is installed at a customer.

Recently I bought CRS210-8G-2S+IN device, for about 2 weeks was running well as router (port 1 for routing) and the reaming ports as bridge.
The setup was:
A) Switch
- port1 used for routing
- ports 2-4 switch, MTU 1500 and L2 MTU 1588
- "bond1 - ESXi" with ports 5&6, MTU 1500 and L2 MTU 8996 (limited by my storage device)
- "bond2 - QNAP" with ports 6&7, MTU 1500 and L2 MTU 8996
- Bridge1 (ports 1 to 4 , bond 1 and bond 2)

B) Server
- My ESXi server with 4 network cards, 2 of them connected to "bond1-ESXi" and the other 2 on the free ports , one card is Intel Gigabit dual port (one port in bond, another port in port4)

C) Storage
- QNAP with his 2 ports connected to "bond2-QNAP"

It was working well for 2 weeks and one day I was at office suddenly I get disconnected from my remote VM (I thought it's a network issue on my Internet provider).
I went home, I connected from LAN and it was OK but after a while I start to get disconnected , a lot of packets lost.

I reset the switch to factory default settings and I reconfigured port1 for routing and the remaining ports for bridge (I didn't configure bonds anymore because the performance on my storage was lower than using a dedicated port for direct connection with my server = iSCSI), but the switch was unstable (packets lost, QNAP showing the LAN is disconnected on the remaining port connected to switch). I updated to the last OS version but I had the same issue; in the end I re-used my old Linksys router and my LAB is up and running).

It's anybody else facing connectivity issue with CRS210-8G-2S+IN?

same here - consistent port flap against i350-t4
6.37. and 6.40.1 too
port goes 1G and down every 2-3 seconds