Questions about the routing protocol (1)

In the routing selection protocol, we can divide it into many detailed types for separate explanation. Now let's take a look at some typical problems. Including Flooding, link status database, and region. Now let's take a look at more specific content from the article.

I. Flooding of Link Status)

After the neighboring relationship is established, the router starts to send the LSA to each neighbor. At the same time, each neighbor saves the received LSA and forwards it to each neighbor in sequence, in addition to the neighbor that sends the LSA, LSA is almost immediately forwarded, which is better than distance vector! Therefore, when the network topology changes, the convergence speed of the link state protocol is much faster than that of the vector protocol;

The pan-flood Diffusion Process in the routing selection protocol is the most complex part of The Link State protocol. There are several ways to make Pan-flood diffusion more efficient and reliable, for example, the unicast and multicast addresses, checksum, and active validation are used. Two processes are extremely important: sorting and aging;

1. Serial number

Assume that router C receives an LSA from router B and saves it to its own topology database, then, through vrof F, you receive the same LSA from A. vroc C finds that the LSA already exists in the Database and knows it was received from B ), should the LSA received by router C from router F be forwarded to router B? The answer is no forwarding! Because vrob B has received this LSA, because the serial number of the LSA received by vroc C from vrof F is the same as the serial number of the LSA received earlier from vrob B, therefore, router C also knows this situation and then discards the LSA;

When router A sends an LSA, the serial number in each copy is the same. The serial number and other parts of the LSA are stored in the topological database of the router, when the router receives an existing LSA from the database and the serial number is the same, the router discards the information. If the information is the same but the serial number is larger, the received information and the new serial number are saved to the database, and the LSA is extensive;

Because the serial number is carried in a fixed field in LSA, the serial number must have an upper limit. What will happen when the serial number reaches the upper limit?

1) linear serial number Space

One way IS to use a very large linear serial number so that it IS impossible to reach the upper limit, such as using a 32-bit long field (IS-IS ); if a link status route selects that the process has used up all serial numbers, it must stop restarting before re-using the lowest serial number), and waits for the LSA to be issued by it to no longer be used in all databases. If the maximum time is 1 hour or longer, this method is not feasible.

2) space of the cyclic serial number

The number in this method is used cyclically. The value 0 follows 4294967295 in a 32-bit space. It may also encounter the same problem as the linear serial number after the router is restarted!

3) lollipop serial number Space

This method is a combination of the linear serial number space and the circular serial number space. It has a linear component and a circular component. The disadvantage of the linear space is that serial numbers cannot be used cyclically, that is, the serial number is limited, the disadvantage of circular space is that there is no number smaller than all others.

2. Aging)

The LAS packet format contains an age field. When the LSA is created, the router sets this field to 0. As the data packet spreads, Each router increases the age in the notice. Of course, another option IS to start from a maximum age, then decrease, ospf is increasing, IS-IS decreasing;

The aging process increases the reliability of pan-Flood spread. This Protocol defines a MaxAgeDiff value for the network. The vro may receive multiple copies of an LSA, with the same serial number and different ages. If the age difference is less than MaxAgeDiff, it is considered that the normal latency of the network causes the age difference. Therefore, the original LSA of the database continues to store and the new LSA (older) do not spread; if the age difference exceeds MaxAgeDiff, the network is considered abnormal because the serial number of the newly sent LSA does not increase. In this case, the newer LSA will be recorded and the packets will be spread out. A typical MaxAgeDiff value of 15 min is used for OSPF );

If the LSA resides in the database, the LSA age increases. If the age of the link status record is increased to a maximum value (MaxAge)-select a protocol from a specific route-An LSA with a MaxAge value is flooded to all neighbors, the neighbor immediately deletes related records from the database;

When the LSA age reaches MaxAge, it will be deleted from all databases, which requires a mechanism to regularly confirm the LSA and reset its timer before reaching the maximum age. The link status refresh timer (LSRefeshTimer) is used for this purpose. Once the timer times out, the router will spread the new LSA to all neighbors, the received neighbor sets the router record age to the new received age. OSPF defines MaxAge as one hour and LSRefreshTime as 30 minutes.