06-23-2017 01:04 AM - edited 03-05-2019 08:45 AM
Hi everyone. A question was stack in my head and and wanted to create simple topology and test one thing about ospf. I was interested if LSU are unicast or multicast. I decided to make this test because couldn`t find proper documentation. So after test, as you see in right side of ss that LSUs are sent in different forms: Initially unicast and then became multicast.
So I need clarification for especially multicast LSUs. OK, unicast LSU were sent as answer for request and they are for synchronazing theri LSDBs. Then, why those multiacst LSUs were sent? Routers in ospf networks send LSU update to multicast on broadcast and p2p networks when only change occur. However here after they syched there were not any change
Need clarification :)
Thanks in advance! :)
Solved! Go to Solution.
06-24-2017 03:32 AM
when the connection between two neighboring routers first comes up, the router must wait for their link state database to synchronized before forwarding data traffic over the connection. In your case even though R2 already knew who the DR was, but the DB synchronization will be unicast between DR (R1) and R2 and once DB has been fully synchronized any subsequent updates sent by R2 will be sent to 224.0.0.5 .
06-23-2017 01:05 AM
PS: Sorry for image. I thought there will be option for zooming(
06-23-2017 08:42 PM
Hi Orkhan,
Then, why those multiacst LSUs were sent? Routers in ospf networks send LSU update to multicast on broadcast and p2p networks when only change occur. However here after they syched there were not any change :
I was able to look at the interface you are using to connect between the routers in your topology. If I am correct it's an Ethernet port and therefore ospf network type will be broadcast by default unless you have changed that.
If it's broadcast then it makes sense what you are seeing in wireshark. It appears that R1 is the DR and R2 is elected as the BDR, after R1 (DR for the segment) received LSUs from R2 it's re-flooding those LSUs to (allspfrouters) address which is 224.0.0.5 so other routers on the segment can update their ospf database even though R2 is the only other router on that segment.
06-23-2017 11:21 PM
I know that multicast updates are sent because of broadcast to 224.0.06 To all DRs and 224.0.0.5 to all routers. But my question is why before that multicast updates , they were sent to unicast? It is because in that stage DR and BDR had not been selected yet?
06-24-2017 03:32 AM
when the connection between two neighboring routers first comes up, the router must wait for their link state database to synchronized before forwarding data traffic over the connection. In your case even though R2 already knew who the DR was, but the DB synchronization will be unicast between DR (R1) and R2 and once DB has been fully synchronized any subsequent updates sent by R2 will be sent to 224.0.0.5 .
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide