Stagnent Ip address routed through Mikrotik

RouterOS Forwarding Protocols
1

Hi - I was hoping someone here may be able to help me out with a curious problem.

I have a publicly routable IP range (195.x.0.0/22) that is published via BGP from a perimiter CCR1009 router and a secondary CCR1009. These two have been separated into two data centres and will be configured for high availability in the future. for the time being the first router is acting as a perimeter router and is publishing the IP range via BGP. The second router is publishing any attached subnets to the first router via OSPF. First router IP is 195.x.1.1 and the second router IP is 195.x.1.2. The second router has the subnet 195.x.0.0/28 published to the perimeter router.

The problem that I am experiencing is that one IP address (195.x.0.1) appears to be locked to a device attached to the second router. If I deassign this IP address and assign it to a new device I am not able to access it from the internet at all however, both Mikrotik routers can successfully ping it and the ARP table reflects the new MAC address. This new device is able to route to the internet. I need to decommission the old device and put in a new one with the same IP.

I have tested by adding a second IP address to the old device 195.x.0.2. I test this from the internet back to the old device and all works. I then deassign this IP and assign it to the new device and I test from the internet and it successfully reaches the new device.

There is an interesting anomaly when running a traceroute to these IP addresses even while both of them are assigned to the same device. There appears to be some hops missing with the IP address that is stagnant. Other than this I am not really sure where to look.

traceroute to 195.x.0.1 (problem IP address)

1 41.x.x.9
2 41.x.x.1
3 * * *
4 196.x.x.101
5 196.x.x.205
6 * * *
7 196.x.x.98
8 196.x.x.171 - Upstream Provider
9 195.x.0.1 - Old Device

traceroute to 195.x.0.2

1 41.x.x.9
2 41.x.x.1
3 * * *
4 196.x.x.101
5 196.x.x.205
6 * * *
7 196.x.x.98
8 196.x.x.171 - Upstream Provider
9 196.x.x.164 - Mikrotik 1
10 195.x.1.2 - Mikrotik 2
11 195.x.0.2 - Old Device


Any suggestions for possibilities will be much appreciated.

2

One thing about the traceroute that seems odd - why is the device’s IP the very first one after the provider IP?
That should be investigated…

Check the routing table on both Mikrotiks whenever the old device is disconnected, but the new one is also not connected.
See if any strange routes show up.

Check the routing tables again when it IS connected.


Try this command:
/ip route print where 195.x.0.1 in dst-address

It will show every route that applies to the address, and of course the most specific route is the one that will actually get taken.

So long as the device itself has the correct IP/mask/gateway on it, and the routes point inward correctly, it should work.
Otherwise, there’s a firewall/mangle/arp/other stranger, deeper low-level issue going on.