No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

The Switchover of the NE40E on Plane B of the Bearer Network Leads to the Interruption of Services on Plane A

Publication Date:  2012-07-27 Views:  33 Downloads:  0
Issue Description
The networking diagram is as follows:
------CE1----AR1---BR1----AR1’--------CE1’-------
|         |          |           |              |                    |                   |                  
UMG ---CE2---AR2---BR2----AR2’-----CE2’--- UMG’
There are two planes on the bearer network. The traffic model of Plane A is UMG--CE1----AR1---BR1----AR1’----CE2, and the returned traffic follows the same path.
When the master and slave AR2s (NE40E) on Plane B are switched, several packets are lost on Plane A (around 1 second). Similarly, when the master and slave AR1s (NE40E) on Plane A are switched, several packets are lost on Plane B. 
 
Alarm Information
Null
Handling Process
You can adopt the following workaround: assign a specific cost value for the aggregated route in the configuration of the route aggregation to avoid route selection when a route with a small cost value is received.
ospf 1 vpn-instance 123
asbr-summary 10.0.0.0 255.0.0.0 cost 300
After the above command is run, several tests are performed to prove that there is no traffic passing the link between the CE1 and CE2 and the traffic on the master plane are always forwarded on the master plane.
In addition, if a specific route is advertised from the AR to the CE, the fault does not occur. 
 
Root Cause
The existing networking topology and routes show that the master and slave planes are independent of each other, and thus the switchover of one plane cannot affect that of the other plane.
The configuration shows that many OSPF instances on all the relevant ARs on the existing network are configured with the route aggregation commands:
ospf 1 vpn-instance 123
asbr-summary 10.0.0.0 255.0.0.0
The routes are advertised on BGP in network mode. Thus, the routes are aggregated as 10.0.0.0/8 at the remote end. According to the aggregation rules of the OSPF ABR, the cost value of the aggregated route is the largest value of the routes before aggregation (Huawei equipment also chooses the route with the maximum cost value after the aggregation of the ASBR and ABR for advertisement). Suppose there are three routes: 10.1.1.1/24 cost 10, 10.2.1.1/24 cost 100, and 10.3.1.1/24 cost 1000. Then the route after aggregation is 10.3.1.1/24 cost 1000.
After the master/slave switchover on the slave plane, the private route on the AR2 is converged again. Suppose that the AR2 receives 10.x.x.x with the cost value smaller than 200, then the cost value of the aggregated route 10.0.0.0/8 advertised from the AR2 to the PE2 is smaller than the original cost value, through the OSPF flooding of PE1. The traffic model on the master plane is UMG--CE1----CE2----AR2---BR2----AR2’----CE2’.
Because of the large scale of the network, the AR2 is not converged completely. That is, there is not a specific route corresponding to the destination for AR2, and thus the above fault occurs. 
 
Suggestions
1. You shall avoid the hidden troubles such as the checksum-error of ISIS, OSPF cost, and ISIS cost value due to the interaction of biplanes in network deployment.
2. After RFC2328 aggregates the ABR, the cost value is specified as follows:
In order to get better aggregation at area boundaries, area
address ranges can be employed (see Section C.2 for more
details). Each address range is defined as an [address,mask]
pair. Many separate networks may then be contained in a single
address range, just as a subnetted network is composed of many
separate subnets. Area border routers then summarize the area
contents (for distribution to the backbone) by advertising a
single route for each address range. The cost of the route is
the maximum cost to any of the networks falling in the specified
range.
For example, an IP subnetted network might be configured as a
single OSPF area. In that case, a single address range could be
configured: a class A, B, or C network number along with its
natural IP mask. Inside the area, any number of variable sized
subnets could be defined. However, external to the area a
single route for the entire subnetted network would be
distributed, hiding even the fact that the network is subnetted
at all. The cost of this route is the maximum of the set of
costs to the component subnets. 
 

END