I have an issue with a new device refusing to assign IP address as I have configured it. The device is CCR2004-1G-12S+2XS, the one with 12 SFP ports.
Below you can find the exported configuration.
I’ve hidden the public IP with the xxx.xxx.xxx.xxx signs.
Basically, I have two subnets (10.100.10.0/24 and 10.100.20.0/24) which each need to be on different interfaces. Interfaces 1-3 are bridged into one and interfaces 4-10 into a second bridge.
I created two DHCP servers, two IP pools and two networks with the above mentioned IP ranges and assigned them each to the DHCP servers which each assign local IPs to their respective bridges.
The detailed configuration is above.
Problem is, when I plug in a device into port 3 for example, it doesn’t get an IP. The leases table is empty and the laptop doesn’t detect a connection.
A bit of a trick is that the network is not even in my country, so I don’t really have direct physical access to it but I have someone on the site helping me.
Nevertheless, I mirrored this setup at my own site with a small rb2011uias Mikrotik and it works flawlessly. Difference is only the device.
Is it maybe because of the SFP ports? Do they need some extra configuring? What else am I missing here?
I would be very thankful for any advice!
Okay, an update: I tried disabling the bridges and only assigning a DHCP server to a single port with some pool. When I’ve done that it works properly, the computer connected gets a proper IP.
The question is, why is it not working with bridged interfaces?
SFP12 is used as the WAN port, it’s connected via an SFP module for a regular patch cable.
I have a laptop on the ether1 (management) port connected which I RDP into to configure the router.
And a test laptop on SFP3 which is also connected over an SFP module with a regular patch cable.
What exactly do you mean by ‘status of the cages’?
As I said, when the DHCP is set only for one port it works properly, only when the interfaces are bridged it stops. I’ve read a lot about this specific issue and mostly the solution is to set the bridge protocol to ‘None’ from MSTP. I did that but it didn’t work for me.
Yep, all the interfaces which are bridged I turned the auto-negotiation off for and suddenly it started working. I set them all to 10Gbps since that’s actually the max throughput for these SPF+ ports.
Another pair of questions not related to the thread per se.
Now I am working on establishing a connection to the outside. This router is in a datacenter behind a Cisco router, which is connected to our router via port SPF12. The datacenter people say that the connection on their router is set as passthrough from the outside to our router (port SPF12).
Question 1: This should mean that our Mikrotik should be accessible directly via the public IP, which I have been provided?
Problem is, I created the public IP I was provided under IP->addresses and assigned it to port SPF12, which also created a route which said it was reachable, but I still couldn’t ping 8.8.8.8 from the router.
Then I’ve tried creating a DHCP client on port SPF12 which DID get an IP address but it was a local one - 192.168.x.x .
Question 2: What the heck does this mean? I concluded that the datacenter people actually are wrong in assuming it’s set up as a passthrough, since their router is obviously assigning our router a local IP address.
