HCNP Study notes OSPF protocol principle and configuration 9-OSPF troubleshooting

How to locate the cause of the fault and correct troubleshooting, all need to be based on the understanding of the operation of the agreement.

This article is about the idea and operation of OSPF troubleshooting.

The command to use:

Display IP routing-table//See route first, determine fault location

display OSPF peer//based on neighbor relationships, further identify the point of failure

display OSPF error//Huawei, China 3 router proprietary, Cisco not supported, very powerful

Display Current-configuration//deprecated, large amount of configuration data in the current network

Display Current-configuration configuration OSPF

display OSPF peer brief//view status, excluding errors

Intra-zone routing failures

Look at the routing table first, if there is a route, usually due to the upper layer configuration, such as firewalls.

If there are no related entries, usually due to a neighbor establishment error, first display OSPF error locating errors According to the command, and then check the relevant configuration of OSPF based on the error, and finally fix the configuration error

display OSPF error, as shown in

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M02/70/9A/wKioL1W6ABLDRwgHAAH48F66cnw231.jpg "title=" 1.jpg " alt= "Wkiol1w6abldrwghaah48f66cnw231.jpg"/>

It can be seen that the cause of the failure is the router ID conflict, after the fault point, with the relevant configuration command to solve.

The following table lists the causes of common errors

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M02/70/9D/wKiom1W5_sazd0ySAAL7mieUrDo564.jpg "title=" 2.jpg " alt= "Wkiom1w5_sazd0ysaal7mieurdo564.jpg"/>

Two-Zone fault

Inter-zone failures need to be ruled out in conjunction with the declaration process of OSPF routes.

At its most basic point, all non-backbone routes are routed through AREA0, and this is done to prevent loops.

To see an example, the topology is as follows:

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M01/70/9D/wKiom1W6ANGCVFxjAACpaZmxPKY183.jpg "title=" 1.jpg " alt= "Wkiom1w6angcvfxjaacpazmxpky183.jpg"/>

In this example, three routers through the serial port interconnection, RTA use 1.1.1.1 as router ID;RTB use 2.2.2.2 as router ID;RTC use 3.3.3.3 as router ID. Network segment 10.1.1.0/30 in area 0, the network segment 10.2.1.0/30 in area 1, the network segment 10.3.1.0/30 in area 2, all links use the same bandwidth, so each router arrives at the end of the network segment should have two equivalent paths. But the actual routes are as follows:

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M02/70/9D/wKiom1W6AQKCPbGcAAGDlVP9xfQ232.jpg "title=" 2.jpg " alt= "Wkiom1w6aqkcpbgcaagdlvp9xfq232.jpg"/>

Reason:

Routing information between regions can only be released through area 0 and cannot be published directly between non-backbone regions.

In this example, RTC cannot publish the routing information to 10.2.1.0/30 directly to RTA, and only the ABR (RTB) of area 1 publishes the routing information that arrives at 10.2.1.0/30 to RTA, so in the RTA routing table, The next hop is only 10.1.1.2 (RTB).

Similarly, in the routing table of RTB, in the route to the network segment 10.3.1.0/30, there is only one next hop, and the next hop is 10.1.1.1 (RTA).

Solution:

A virtual link is configured between RTB and RTC, and the transit area used by the virtual connection is area 1. RTC can post information in the backbone area

A virtual link is configured between RTA and RTC, and the transit area used by the virtual connection is area 2.

After configuring the virtual connection between RTB and RTC, area 1 is the transit zone, and the RTC will only release the routing information for areas 2 to RTB via Region 1, and will not release the routing information for Area 1 to RTA by the location 2, so a virtual connection with RTA is required. The routing information in Area1 is published through AREA2 to Area0.

Three-route aggregation failure
When the routing aggregation is configured in multiple ABR regions, all the ABR should be configured to converge, not only on the part of the ABR to configure aggregation.

Example, the topology is as follows:

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M00/70/9D/wKiom1W6AevwOXdZAADYuqJ1uRM097.jpg "title=" 1.jpg " alt= "Wkiom1w6aevwoxdzaadyuqj1urm097.jpg"/>

All network segments are configured on the RTC in area 1, the routing aggregation is configured on the ABR of area 1, and the network segments 20.1.1.1/32 and 20.1.1.2/32 converge into 20.1.1.0/24, so that when the ABR advertises to the backbone, it only advertises 20.1.1.0/ 24, suppress detail routing 20.1.1.1/32 and 20.1.1.2/32.

Fault: Look at the RTA routing table, find that the route is incorrect, go to the detail segment of the route using the RTD as the next hop, the route to the Converged network segment uses RTB as the next hop, both exist simultaneously.

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M00/70/9A/wKioL1W6A__yXTAVAAEsuMWEzgY331.jpg "title=" 2.jpg " alt= "Wkiol1w6a__yxtavaaesumwezgy331.jpg"/>

Failure reason analysis: Because the routing to the detail segment uses the RTD as the next hop, the RTD only advertises the detail route, does not advertise the aggregation route, and may not even do the routing aggregation; The aggregation route uses RTB as the next hop, stating that RTB correctly advertises the route.

This article from "Day Up" blog, please be sure to keep this source http://4625416.blog.51cto.com/4615416/1680271

HCNP Study notes OSPF protocol principle and configuration 9-OSPF troubleshooting