Only specified routes are propagated from the remote stub device. A device that is configured as a stub with the EIGRP Stub command shares connected and summary routing information with all neighbor devices by default. With under commands, the configured device act as a stub that advertises connected and summary routes:. The EIGRP stub command is used with the connected and static keywords to configure the device as a stub that advertises connected and static routes.
This option is enabled by default and is the most widely practical stub option. Redistributing static routes with the redistribute static command is still necessary.
Summary routes can be created manually with the ip summary-address command or automatically at a major network border router with the auto-summary command enabled. This option is enabled by default. The parameters of this command can be used in any combination, with the exception of the receive-only keyword. After receiving this information only then will R5 send the reply back up to R4 and then with the last reply from R4 sent up to R1.
After this has completed, R1 realises that there is no backup path to 1. As you can see this process is quite long and depending on the topology some steps are redundant.
As you can see below, R6 ended up receiving the query. As the network grows this process could become too long and the route may become stuck in active. At which point the adjacency between the converging router and the routers queried will be torn down.
When you configure a router as a stub it will do two things. Firstly, It will tell the neighbours to suppress any queries to this neighbour and secondly it will influence what updates are sent to the neighbours. For example, I will now configure R2 to become a stub router since it doesn't connect to anything else.
Upon executing this command R2 will reset it's neighbours. As a result R1 will now no longer send query messages over to R2. You can issue the show ip eigrp neighbors detail command to see detailed information about neighbours; including whether or not queries are suppressed and what sort of stub configuration has been applied. Equally, now if we flap the loopback interface on R1, R2 will only receive an update message about the loopback going down and will not get queried about whether or not it has a backup to that prefix.
With this we could implement a similar configuration on R3 and R6 and then be done however that would mean that as the core expands the query domain is still increasing largely. Instead we could segment the query domain in half by using a transit router such as R4 and configuring that as an EIGRP stub router.
This would require some extra configuration to complete this. See the image below for more details on the different syntax options with the EIGRP stub command that can be stringed together.
Connected - This means that the router can advertise directly connected networks that have been included into the EIGRP process. Leak-Map - A leak map allows you to fully customise which prefixes can be advertised to neighbours. Typically this is used in designs where a stub router is in transit for some traffic. Receive-Only - This means that the router has no additional networks it knows about and should only receive updates. This is typically used on IDS appliances that connect to a single port on a switch only.
Since R4 is acting as a transit router between R5 and the rest of the network we will need to implement a leak-map configuration to ensure that all the prefixes are advertised accordingly. To do this, we can create a prefix-list of all the prefixes we wish to allow and then reference that inside a route-map which will be applied as a leak-map. The configuration of this can be seen below:.
As a side note, the leak map configuration will apply in both directions and therefore means that R4 will ONLY advertise what is listed inside the leak-map as well as directly connected and summary routes. To access Cisco Feature Navigator, go to www.
An account on Cisco. This type of configuration is commonly used in WAN topologies, where the distribution device is directly connected to a WAN. The distribution device can be connected to many remote devices, which is often the case. In a hub-and-spoke topology, the remote device must forward all nonlocal traffic to a distribution device, so it becomes unnecessary for the remote device to have a complete routing table.
Generally, the distribution device need not send anything more than a default route to the remote device. Only specified routes are propagated from the remote stub device. Any neighbor that receives a packet informing it of the stub status will not query the stub device for any routes, and a device that has a stub peer will not query that peer.
The stub device will depend on the distribution device to send proper updates to all peers. The stub routing feature by itself does not prevent routes from being advertised to the remote device. In the above example, the remote device can access the corporate network and the Internet only through the distribution device. Having a complete route table on the remote device would serve no functional purpose because the path to the corporate network and the Internet would always be through the distribution device.
The large route table would only reduce the amount of memory required by the remote device. Bandwidth and memory can be conserved by summarizing and filtering routes in the distribution device.
The remote device need not receive routes that have been learned from other networks because the remote device must send all nonlocal traffic, regardless of the destination, to the distribution device.
If a true stub network is desired, the distribution device should be configured to send only a default route to the remote device. In most cases, the network administrator will need to configure summarization on distribution devices. When configuring the distribution device to send only a default route to the remote device, you must use the ip classless command on the remote device.
Without the EIGRP stub routing feature, even after routes that are sent from the distribution device to the remote device have been filtered or summarized, a problem might occur. If a route is lost somewhere in the corporate network, EIGRP could send a query to the distribution device, which in turn would send a query to the remote device, even if routes are being summarized.
If there is a communication problem over the WAN link between the distribution device and the remote device, an EIGRP stuck in active SIA condition could occur and cause instability elsewhere in the network.
The EIGRP stub routing feature allows a network administrator to prevent queries from being sent to the remote device. In addition to a simple hub-and-spoke network, where a remote device is connected to a single distribution device, the remote device can be dual-homed to two or more distribution devices.
This configuration adds redundancy and introduces unique issues, and the stub feature helps to address some of these issues. A dual-homed remote device will have two or more distribution hub devices.
However, the principles of stub routing are the same as they are with a hub-and-spoke topology. The figure below shows a common dual-homed remote topology with one remote device: however, or more devices could be connected on the same interfaces on distribution Device 1 and distribution Device 2.
0コメント