ccie blog

Hello Protocol On Each OSPF Network Type

Point-to-Point Network

– OSPF hellos are sent to the AllSPFRouters multicast address – 224.0.0.5

– Default timers are Hello 10 secs, Dead 40 secs

– No DR/BDR election (point-to-anything does NOT require a DR)

I have configured an ospf point-to-point network below to show the OSPF hello packets that are sent:

I’ve taken a packet capture that you can view online here: OSPF Point-to-Point Packet Capture.  The capture shows that packets are sent to the 224.0.0.5 AllSPFRouters address.  It also shows the DR/BDR ip address as 0.0.0.0, indicating that no DR/BDR election has occurred.  The command below shows the default timers on a point-to-point network.

R2#sh ip ospf int s1/0
Serial1/0 is up, line protocol is up
  Internet Address 24.0.0.1/24, Area 0
  Process ID 1, Router ID 2.2.2.2, Network Type POINT_TO_POINT, Cost: 64
  Transmit Delay is 1 sec, State POINT_TO_POINT
  Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5
    oob-resync timeout 40
    Hello due in 00:00:05

 

Broadcast Networks

– OSPF hellos are sent to the AllSPFRouters multicast address – 224.0.0.5

– Default timers are Hello 10 secs, Dead 40 secs

– DR/BDR election occurs

In the diagram below you can see that four routers are connected over the same Ethernet segment.  You can immediately see the benefit of broadcast networks on this segment when you look at the amount of hello’s sent.  If hellos were unicast to each neighbor, each router would have to send 3 hello’s.  Because, in broadcast and point-to-point networks, the hellos are sent as multicast packets, it means each router only has to send one hello.

 

 

I’ve taken a packet capture that you can view online here: OSPF Broadcast Network Packet Capture. You will see within the OPSF Hello Packet that each router sends a hello to 224.0.0.5 & lists the router-id’s it knows about.  For example, in the first capture, when the source 10.0.0.3 sends a hello to 224.0.0.5, it lists 1.1.1.1, 2.2.2.2, & 4.4.4.4 as active neighbors. This is actually the verification process that occurs in the 2-way state when the OSPF adjacency is forming.  So when routers R1, R2 & R4 receive this multicast, they can see their router-id’s in the hello packet that was sent from R3. The command below shows the default timers in an OSPF broadcast network type.

R4#sh ip ospf int
FastEthernet0/0 is up, line protocol is up
  Internet Address 10.0.0.4/24, Area 0
  Process ID 1, Router ID 4.4.4.4, Network Type BROADCAST, Cost: 1
  Transmit Delay is 1 sec, State DR, Priority 1
  Designated Router (ID) 4.4.4.4, Interface address 10.0.0.4
  Backup Designated router (ID) 3.3.3.3, Interface address 10.0.0.3
  Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5

 

 Point-to-Multipoint (broadcast) Network

– OSPF hellos are sent to the AllSPFRouters multicast address – 224.0.0.5

– Default timers are Hello 30 secs, Dead 120 secs

– No DR/BDR election

In the diagram below, I’ve configured a point-to-multipoint (broadcast) network.  On this type of network, each link is treated as a point-to-point network.  So actually, R2 & R3 do not become adjacent.  Instead, R1 injects itself as the next hop between these routers in the routing table.

 

The packet capture can be seen online here: OSPF Point-to-Multipoint Packet Capture.  The capture shows that only hellos sourced from R1 include both R2 (2.2.2.2) & R3 (3.3.3.3) as an active neighbor. Hello’s sourced from R2 show only R1 (1.1.1.1) as an active neighbor. The command below shows the default timers for this kind of network:

R2#sh ip ospf int s1/0
Serial1/0 is up, line protocol is up
Internet Address 10.0.0.2/24, Area 0
Process ID 1, Router ID 2.2.2.2, Network Type POINT_TO_MULTIPOINT, Cost: 64
Transmit Delay is 1 sec, State POINT_TO_MULTIPOINT
Timer intervals configured, Hello 30, Dead 120, Wait 120, Retransmit 5

 

 Point-to-Multipoint non-broadcast Network

– OSPF hellos are sent unicast to each neighbor

– Default timers are Hello 30 secs, Dead 120 secs

– no DR/BDR election

The topology is the same as the the previous example, however this is a point-to-multipoint non-broadcast network.  This means I had to statically configure the neighbors under the OSPF process on R1, or the hello’s would not be sent. In all non-broadcast networks, the hellos are sent as unicast rather than multicast. The reason for this is because non-broadcast networks do not send either broadcasts or multicasts.  Again, each link is treated like a point-to-point network, so R2 & R3 do not become adjacent: and R1, again, injects itself as the next hop between R2 & R3 in the routing table.

The packet captures for this network can be found here: OSPF Point-to-Multipoint non-broadcast Network.  As you can see, the destination in each hello is  the actual neighbors interface IP address.  The default timers for this type of network are shown below:

R1#sh ip ospf int s1/0
Serial1/0 is up, line protocol is up
  Internet Address 10.0.0.1/24, Area 0
  Process ID 1, Router ID 1.1.1.1, Network Type POINT_TO_MULTIPOINT, Cost: 64
  Transmit Delay is 1 sec, State POINT_TO_MULTIPOINT
  Timer intervals configured, Hello 30, Dead 120, Wait 120, Retransmit 5

Please note, the output shows point-to-multipoint.  This is just how it displays this network type even though it’s point-t0-multipoint non-broadcast.  One last thing on this network type.  The reason point-t0-multipoint non-broadcast was created was so that the OSPF cost could be assigned on a per neighbor basis, rather than per-interface basis, giving us more control on the way our routing works on multipoint interfaces.

 

Non-broadcast Network

– OSPF hellos are sent unicast to each neighbor

– Default timers are Hello 30 secs, Dead 120 secs

– DR/BDR election occurs

The topology is the same as the previous diagram, however the the network type is simply just a non-broadcast network (default for frame-relay).  R1 is the DR and, again, injects itself as the next hop between spokes. R1 was the only router to be configured with the neighbor commands under the OSPF process.  The reason for this is because the spokes reply with hellos upon receiving a hello, so there is no need for neighbor statements to be configured under their OSPF process. The output below shows the default timers in this network type:

Serial1/0 is up, line protocol is up
  Internet Address 10.0.0.1/24, Area 0
  Process ID 1, Router ID 1.1.1.1, Network Type NON_BROADCAST, Cost: 64
  Transmit Delay is 1 sec, State DR, Priority 1
  Designated Router (ID) 1.1.1.1, Interface address 10.0.0.1
  No backup designated router on this network
  Timer intervals configured, Hello 30, Dead 120, Wait 120, Retransmit 5

I’ve taken a packet capture that you can view online here: OSPF Non-Broadcast Packet Capture. It can be used to verify any of the information above.

 

Table to show the default OSPF timers

OSPF Timers
no Broadcast non-broadcast point-to-point point-to-multipoint point-to-mutlipoint non-broadcast
Hello Interval 10 30 10 30 30
Dead Interval 40 120 40 120 120
Poll Interval 120 120 120 120 120
Retransmit-Interval 5 5 5 5 5
Transmit-Link-Delay-Interval 1 1 1 1 1

3 Comments

p.rameshFebruary 11th, 2014 at 4:43 am

sir/madam

yr blog is excellent

QuebecoApril 25th, 2014 at 2:02 am

Great explanation
Thank you for the clarity

KarthikSeptember 1st, 2014 at 2:54 pm

Very well explained with topology and packet captures.

Leave a comment

Your comment