Thinking on the route control method of China-ccie IGP

Information location: http://www.china-ccie.com/ccie/lilun/igp/igp.html

Through the analysis can be learned that the reason for the suboptimal path is: (OSPF ad value is less than RIP) (in fact, the use of two-way redistribution, resulting in a more complex network, the generation of suboptimal path, this thing really no other way to use.) ）

Since router R1 in the RIP Zone distributes the external route 100.1.1.0 into the RIP, R2 again distributes 100.1.1.0 into the OSPF area, routing to R4, because the ad value of OSPF is 110, less than RIP, resulting in R4 reaching 100.1.1.0 from farther OSPF paths, When R4 re-distributes OSPF into RIP, the METIRC value is 1, less than R1 will be 100.1.1.0 is distributed into the RIP value, so R3 also chooses to go from the OSPF path to 100.1.1.0/24.

Direction 1: You should not configure the OSPF redistribution into RIP, and then return the 100.1.1.0/24 to rip (let the incoming do not yield)

Direction 2: Simply redistribute RIP to OSPF without 100.1.1.0/24, but there are several ways to solve the problem. (Never Let Go)

My thinking: According to the topology, the sub-optimal path is generated because of these two. Different topologies, the reasons for generating suboptimal paths are different. When analyzing a problem, try to find out the cause of the suboptimal path before you resolve it. The following 5 methods, just provide ideas, the specific use of which, according to the specific situation. As the subject shows, I will summarize the pros and cons of these methods.

I think: When configured, you can't distribute it back from one protocol to another. Is that right?

Method 1: Filter by distribute-list (direction 1)

Configuring Distribute-list filtering on R4, OSPF on R4 rejects the route from 100.1.1.0/24 (from the OSPF domain), preventing suboptimal paths.

R4 (config) #access-list 4 Deny 100.1.1.0

R4 (config) #access-list 4 Permit any

R4 (config) #router OSPF 1

R4 (config-router) #distribute-list 4 in s1/0

Method 2: Modify the Ad Value control path direction (based on the protocol itself, from the modification protocol parameters to solve.) Difficult)

Since the ad value of 100.1.1.0/24 in R4 from RIP is 120, and the ad value learned from OSPF is 110, it causes R4 to go from OSPF to 100.1.1.0/24, so by modifying the ad value of OSPF larger than RIP, you can control the path direction, It is also possible to control the ad value to control the path of the R3 toward 100.1.1.0/24. (It is not recommended to modify the ad value, it will complicate the problem, or how to get the problem of the route out of it.) And it needs to be configured on all routers on the path to the route to be controlled. Feeling or only on the border router to do the configuration is better, after all, it is done redistribution caused by this)

R3 (config) #access-list permit 100.1.1.0

R3 (config) #router rip

R3 (Config-router) #distance 109 13.1.1.1 0.0.0.0 33

R4 (config) #acc ess-list permit 100.1.1.0

R4 (config) #router OSPF 1

R4 (Config-router) #distance 121 2.2.2.2 0.0.0.0 44

Method 3: Filter routes via Route-map (direction 2)

When you redistribute rip into OSPF on R2, filter out 100.1.1.0/24

R2 (config) #access-list permit 100.1.1.0

R2 (config) #route-map NET100 deny 10

R2 (Config-route-map) #exit

R2 (config-route-map) #match IP address 22

R2 (Config-route-map) #exit

R2 (config) #route-map NET100 Permit 20

R2 (Config-route-map) #exit

R2 (config) #router OSPF 1

R2 (config-router) #redistribute RIP

R2 (config-router) #redistribute RIP subnets Route-map NET100

Description: Filter out 100.1.1.0/24 when RIP is re-distributed into OSPF via ROUTE-MAP on R2

Method 4: Inter-process filtering (also used for distribution lists) (direction 1)

When R4 the OSPF route into RIP, it filters out 100.1.1.0/24 in the process to return RIP: Similar to Method 1, Method 1 is applied to the physical interface of OSPF (out of direction), Method 4 is applied in the RIP protocol (from OSPF) in the direction.

R4 (config) #access-list 4 Deny 100.1.1.0

R4 (config) #access-list 4 Permit any

R4 (config) #router rip

R4 (config-router) #distribute-list 4 out OSPF 1

Disadvantage: Because the filter is in effect on other routers, the path of the R4 remains unchanged. (Valid for R3, R4 Invalid) Method 1 is more effective relative to this.

Method 5: Configure the Tag filtering technology to control the path trend (ROUTE-MAP)

Tag filtering technology is the routing of the respective protocol tag, in the redistribution of the route back to the original routing protocol, by matching the tag to reject the return route (recommended: Do not need to configure the address segment, unlike other only to resolve the 100.1.1.0/24,12,13 network segment is also back to normal)

R4 (config) #route-map r2o deny 10

R4 (CONFIG-ROUTE-MAP) #match Tag 110

R4 (Config-route-map) #exit

R4 (config) #route-map r2o Permit 20

R4 (config-route-map) #set tag 120

R4 (Config-route-map) #exit

R4 (config) #route-map o2r deny 10

R4 (config-route-map) #match tag 120

R4 (Config-route-map) #exit

R4 (config) #route-map o2r Permit 20

R4 (CONFIG-ROUTE-MAP) #set Tag 110

R4 (Config-route-map) #exit

R4 (config) #router rip

R4 (Config-router) #redistribute OSPF 1 metric 1 Route-map o2r

R4 (config) #router OSPF 1

R4 (config-router) #redistribute RIP subnets Route-map r2o

In the bidirectional redistribution, the OSPF own route to Tag 110, rip its own route to tag 120, and by matching the tag to reject the original route back to return, that is, OSPF redistribution into the RIP, the rejection Tag 120 of the route returned, When RIP is re-distributed into OSPF, the route to deny Tag 110 is returned because they are inherently their own routes.

Disadvantage: Configuration commands are more, in R2, R4 have to be configured. The command is not good to understand, not to write.

Summary: The OSPF is distributed into RIP and is configured under the RIP protocol.

This is what I can think of, there may be unexpected or wrong place, I hope you point out, learn together.

This article is from the "Snow Moon Studio" blog, please be sure to keep this source http://xueyue8.blog.51cto.com/4650249/1621195

Thinking on the route control method of China-ccie IGP