OSPF Adjacencies

‚  < ‚  Free Open Study ‚  > ‚  

OSPF creates adjacencies between neighboring routers for the purpose of exchanging routing information. Not every neighbor becomes adjacent in a broadcast environment. The Hello protocol is responsible for establishing and maintaining an adjacency .

Hello packets are sent periodically out all router functional interfaces. Two-way communi-cation is established when the router is listed in the neighbor's Hello packet. On broadcast and NBMA networks, Hello packets are used to elect the DR/BDR.

After the two-way communication is established, the decision is made whether to form an adjacency with this neighbor. This decision is based on the neighbor state and network type. If the network type is broadcast or nonbroadcast, the adjacency is formed only with the DR and BDR routers. In all other network types, the adjacency is formed between two neighbor routers.

The first step in forming the adjacency is synchronization of the database. Each router des-cribes its link-state database in the DBD packet. Only the LSA headers are exchanged bet-ween neighbors. Master and slave election takes place during this database exchange. Each router makes a note of the LSA headers that it receives during this DBD exchange. At the end of the DBD exchange, it sends the LS request packet to request LSAs whose headers have been seen during the DBD exchange. The neighbor router then replies with the LS update packet listing the entire content of those LSAs. This LS update packet is then acknowledged by sending the link-state acknowledgment packet. At this point, all the databases are fully exchanged, and the neighbor goes into Full state.

A router can be in several neighbor states:

  • Down

  • Attempt

  • Init

  • 2-way

  • Exstart

  • Exchange

  • Loading

  • Full

The sections that follow describe the different OSPF states in more detail.

OSPF Down State

In Figure 8-29, R1 and R2 are running OSPF. The neighbor state shows DOWN. This state means that no information has been received from the neighbor yet.

Figure 8-29. OSPF Down State

graphics/08fig29.gif

OSPF Attempt State

The Attempt state is valid for neighbors on NBMA networks. If a neighbor is in this state, it means that no information is received from this neighbor, but serious effort is being made to contact the neighbor. Serious effort means that this router will constantly send a Hello packet upon every Hello interval to contact the neighbor. In Figure 8-30, R1 is sending a Hello packet that says that R1 has not seen anyone yet and doesn't know about the DR.

Figure 8-30. OSPF Attempt State

graphics/08fig30.gif

OSPF Init State

Init state is a one-way Hello. In Figure 8-31, R1 sends a Hello packet. Upon receiving this Hello, R2 declares a one-way state because R2 doesn't see itself (its router ID) in that Hello packet.

Figure 8-31. OSPF Init State

graphics/08fig31.gif

OSPF 2-Way State

An OSPF neighbor reaches the 2-way state when bidirectional communication is estab-lished. This is the beginning of an OSPF adjacency. The DR and BDR are elected in this state. In Figure 8-32, R2 sends a Hello packet that says that R2 has seen R1's Hello; the router ID of R2 is higher, so it has also elected itself as a DR.

Figure 8-32. OSPF 2-Way State

graphics/08fig32.gif

OSPF Exstart State

This state is used for initialization of the database synchronization process. Master and slave are elected in this state. The first sequence number for DBD exchange is also decided in this state. In Figure 8-33, R1 sends the first DBD packet. R2 also sends it first DBD packet. The router that has the highest router ID becomes the master. In this example, R2 has a higher router ID, so R2 is the master.

Figure 8-33. OSPF Exstart State

graphics/08fig33.gif

OSPF Exchange State

In the Exchange state, the router describes its entire link-state database through DBD packets. Each DBD sequence is explicitly acknowledged. Only one outstanding DBD packet is allowed at a time. Link-state request packets are also sent in this state to request a new instance of the LSA. Figure 8-34 shows the exchange process in action. R1 and R2 are exchanging their database information. The last arrow shows that the M bit is set to 0. This means that the master has no more data to send. At this stage R1, the slave will send whatever database is left, and R1 will also set the M bit to 0. This is the indication that both routers have exchanged the complete database.

Figure 8-34. OSPF Exchange State

graphics/08fig34.gif

OSPF Loading State

In the Loading state, LS request packets are sent to request the more recent instance of an LSA that has not been received during the exchange process. In Figure 8-35, R1 is in the Loading state and is sending LS request packets to receive a more recent instance of an LSA.

Figure 8-35. OSPF Loading State

graphics/08fig35.gif

OSPF Full State

This state means that the complete information has been exchanged between OSPF neigh-bors. In Figure 8-36, R1 and R2 have exchanged their entire database information and are in the Full state.

Figure 8-36. OSPF Full State

graphics/08fig36.gif

‚  < ‚  Free Open Study ‚  > ‚  


Troubleshooting IP Routing Protocols
Troubleshooting IP Routing Protocols (CCIE Professional Development Series)
ISBN: 1587050196
EAN: 2147483647
Year: 2002
Pages: 260

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net