DR/BDR election and network types

How are the DR and BDR chosen?

The routers choose the DR and BDR themselves. This choice is called an election. First, consider a new LAN where no DR or BDR has been chosen and all four routers are ready.

1. Check eligibility: whether a router is allowed to take either job. An interface priority of 0 excludes it from both roles. It can still exchange OSPF information as a DROther.
2. Compare the eligible routers' interface priorities. This is a number set on each connection to the LAN. Cisco's default is 1; eligible values are 1–255. The higher value wins the comparison.
3. If priorities are equal, compare router IDs numerically. A router ID is a 32-bit number written like an IPv4 address that identifies the router in OSPF. For example, with equal priorities, 3.3.3.3 wins over 2.2.2.2. The local OSPF process number is not a tie-breaker.

In our new election, R1 has priority 100, R2 has 50, R3 has 10 and R4 has 0. R1 becomes DR and R2 becomes BDR. R3 and R4 are DROthers.

Why does OSPF calculate the backup first?

The ranking above shows the final result for a fresh election. The actual procedure in RFC 2328, section 9.4 also considers the roles routers announce in their Hello messages. A router considers itself and neighbours with two-way communication, excluding priority 0.

1. Calculate the BDR. Exclude routers declaring themselves DR. Prefer those already declaring themselves BDR; compare priority, then numeric router ID within that group. If none declares itself BDR, compare the remaining eligible routers.
2. Calculate the DR. Prefer routers declaring themselves DR, using the same tie-breakers. If there are none, promote the selected BDR.
3. Repeat when the local role changes. This prevents one router from keeping both jobs. In our fresh example, R1 is first selected as the backup candidate and promoted to DR. Recalculation leaves R2 as BDR.

Once a working DR and BDR are in place, a router with a higher priority does not replace them just by joining the LAN. Raising R4's priority later does not remove R1 or R2 from their jobs.

This behaviour is called nonpreemption: the newcomer does not take a working router's role. Start order therefore matters. Our ranking assumes all candidates are present for the fresh election; it is not a promise that the highest-priority router always becomes DR on an existing LAN. Keeping established roles avoids unnecessary changes to OSPF adjacencies.

A router ID and an interface IP address have different jobs. R1's router ID is 1.1.1.1. Its IP address on this LAN is 10.0.0.1. These are numbers, not website URLs. OSPF Hello messages use the LAN interface addresses in their DR and BDR fields.

Follow one worked election and failure

Four routers share VLAN 10 through SW1. Each router uses Gi0/0; the addresses are 10.0.0.1–10.0.0.4/24. SW1 connects them at Layer 2 and does not run OSPF in this example. Follow the six steps below to see which roles change and why. Use Next step, choose a step directly, or open Full screen for a wider view.

The higher-priority step temporarily changes R4 to 255. The failure step returns R4 to its original priority 0. The final two steps return to the original healthy roles to explain update delivery; they are a separate example.

What happens when the DR fails?

Suppose R1 stops sending OSPF messages. Start with the original priorities: R1 is DR at 100, R2 is BDR at 50, R3 is a DROther at 10 and R4 is a DROther at 0. We need to separate noticing the failure from changing the roles.

1. The neighbours detect that R1 is missing

R1's last Hello does not tell the others that it is about to fail. Each neighbour counts down its Dead timer: how long it may go without another valid Hello from R1. With this lab's Hello 10 / Dead 40 settings, detection can take up to 40 seconds after the last Hello if the other routers' Ethernet interfaces remain up. A local interface going down can remove its neighbours sooner. Do not assume that R1 failing makes every router's switch connection go down.

2. The existing backup takes the DR role

Once R1 is declared down, R2 becomes DR. R2 already maintains Full adjacencies and an up-to-date link-state database: its stored description of the area's links. This preparation is why we keep a backup. After the failure is detected, the current BDR is promoted; the surviving routers do not start both jobs again from an empty election.

3. A remaining eligible router becomes the new backup

R3 becomes BDR. It has priority 10, while R4 has priority 0 and cannot hold either role. R4 remains a DROther and forms a Full adjacency with the new BDR as needed. The following snapshot shows the settled roles after detection and role changes, rather than an instant view of every database-exchange message.

4. R1 returns, but does not take its old job back

R1's priority is still 100. However, R2 is now a working DR and R3 is a working BDR, so they keep their jobs. R1 learns those roles through Hello messages and rejoins as a DROther. It rebuilds Full adjacencies with R2 and R3 and exchanges the network information it needs. Select Former DR returns in the walkthrough to compare this with the failure picture.

To check recovery, run show ip ospf neighbor on R4. After the exchange finishes, expect R2 as FULL/DR and R3 as FULL/BDR. After R1 returns, its row should normally settle at 2WAY/DROTHER from R4's viewpoint. Roles belong to this shared segment; another interface can have a different role.

Does all user traffic go through the DR?

No. A computer sends traffic for other networks to its default gateway. That gateway can be a DROther. The router then uses its routing table to decide where to send the packet next. The DR job is about sharing OSPF information, not carrying every user's traffic.

Do two routers always need a DR and BDR?

No. OSPF has a point-to-point network type for a link between two routers. With this type, the routers exchange information directly and do not choose a DR or BDR.

Having only two routers, or using a /30 subnet mask, does not automatically change an Ethernet link to this type. Configure OSPF point-to-point on both ends when that is the intended setup.

Configure priorities for the shared-LAN example

These commands use Cisco IOS-style syntax. Each router already has an IP address and OSPF running on Gi0/0. Start in global configuration mode on the named router, then enter its command block. The commands set the priorities used in our example; they do not replace a working DR or BDR.

Changing priorities or restarting a process can affect adjacencies and routes. Verify the existing roles first; resetting OSPF is not required just because a healthy election result differs from the preferred ranking.

Check the roles from R3

R3#show ip ospf neighbor
Neighbor ID     Pri   State           Dead Time   Address         Interface
1.1.1.1         100   FULL/DR         00:00:36    10.0.0.1        GigabitEthernet0/0
2.2.2.2          50   FULL/BDR        00:00:33    10.0.0.2        GigabitEthernet0/0
4.4.4.4           0   2WAY/DROTHER    00:00:38    10.0.0.4        GigabitEthernet0/0

R3#show ip ospf interface GigabitEthernet0/0
GigabitEthernet0/0 is up, line protocol is up
  Internet Address 10.0.0.3/24, Area 0
  Process ID 1, Router ID 3.3.3.3, Network Type BROADCAST, Cost: 1
  Transmit Delay is 1 sec, State DROTHER, Priority 10
  Designated Router (ID) 1.1.1.1, Interface address 10.0.0.1
  Backup Designated router (ID) 2.2.2.2, Interface address 10.0.0.2
  Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5
  Neighbor Count is 3, Adjacent neighbor count is 2

Configure and verify the separate point-to-point link

R1#show ip ospf neighbor
Neighbor ID     Pri   State           Dead Time   Address         Interface
2.2.2.2           0   FULL/  -        00:00:37    10.0.12.2       GigabitEthernet0/1

Sources: Cisco OSPF interface fields and RFC 2328 election procedure.

Test yourself on DR and BDR elections