Pass Cisco ENSLD 300-420 Exam in First Attempt Easily

CCNP Enterprise ENSLD (300-420) : Designing EIGRP Routing

1. SCALABLE EIGRP DESIGNS AND FAST CONVERGENCE

Is very tolerant of arbitrary technologies for small and medium-sized networks. This tolerance is both a strength and a weakness. First, let us define convergence. Convergence is the time taken for the devices in the network to recover after there is a topology change in the network. Of course we are looking for fast convergence, and EIGRP has a reputation for being one of the fastest routing protocols to reconvert. The topology of EIGRP is a lot less strict than the two-tad strict hierarchy of OSPF, but because of that, it is recommended that you are cautious with your design and include route summarization and sub routers to limit the scope of queries within the EIGRP domain. EIGRP tolerates arbitrary topology better than OSPF use. A structured hierarchical approach With EIGRP, hierarchy becomes more crucial as the size of the network increases. Scaling EIGRP GRP depends on topology and other functions such as route summarization and filtering. Limit the scope of EIGRP queries. When there is no feasible successor, you can deploy EIGRP without restructuring the network. But as the scale of the network increases, the risk of instability or long convergent times becomes greater. If you scale your network beyond a couple of hundred routers without a structured hierarchy, then you will face AIGRP performance issues. As you increase the size of the network, you will need a stricter network design. This mechanism contrasts with OSPs, where a structured design is imposed at an early stage. The counterpart to using EIGRP with an arbitrary topology would be an OSPF design that puts everything into OSPF area zero. This design works for small and medium-sized networks up to about 300 OSPF routers To scale the IDRP, you should use a structured hierarchical topology with routers with route summarization. One of the most significant stability and convergence issues with EIGRP is the propagation of EIGRP queries. When EIGRP does not have a feasible successor, it sends queries to its neighbors. The query tells the neighbour I do not have a route to this destination anymore. Do not route through me. Let me know if you hear of a viable alternative route. The router has to wait for replies to all queries that it sends out. Queries can flood through many routers in a portion of the network and increase convergence time. summarization points EIGRP's stubborn and route filtering limit EIGRP query propagation and minimise convergence time. Examples that are used in the module focus on IPV 4, but the recommendations for EIGRP concepts apply equally to both IPV 4 and IPV 6. We now have some hierarchical design recommendations for EIGRP, with this one demonstrating that we have areas, which we call zones, in different areas of the network. We are going to be using summarization to create our hierarchy and EIGRP as opposed to OSP, which uses areas. The zones are the failure domains, and the chokepoints represent the places where the zones connect into the network, and these are the places where you are going to apply your summaries. As a result, the more complex your network, the more splitting up and summarization you should use. Now there are two basic designs—two basic hierarchical designs. One is a two-tier hierarchy design; the other is a three-tier hierarchy design, and depending on the size of your network and the amount of resources that you are sharing, you will need to decide which one fits the topology of your network. EIGRP Autonomous Systems Sometimes when an EIGRP environment needs to be scaled, they will use multiple autonomous system numbers and multiple routing protocols. Now, this has its own problems because any time you use route redistribution, which will be used in this scenario right here, this can open you up to some other issues. So you can see that in this topology, rip is being used in the network in the lower portion of the graphic, and rip is being redistributed into a 200, and then router A then redistributes routes into a 100. So router B receives the route both from a S 100 and a S 200. Now, external routes are another potential problem that you might have in this diagram. The route is redistributed from Rip into S 200, and router B hears about it, and it hears about it from both as S 208,000. Because it is an external route, the administrative distance will be the same. So whatever one shows up first is the one that goes into the database. EIGRP Hierarchical Design Basics EIGRP has no areas because topology information is hidden at each one by design. In an EIGRP network, hierarchy is created through summary. EIGRP has no impulse limit on levels of hierarchy, which can be its key design advantage. The basic concepts of "zones" are topologically defined parts of the network. Zones consist of choke points, which are places where zones are interconnected, and layers, which are groups of zones that are equally distant from the core topologically defined parts of the network. Zones represent the failure domain link, and device failures should have little or no impact outside of a specific zone. Choke points represent the places where zones are interconnected. Choke points provide a place where you can aggregate reachability and topology information. Choke points are where you apply summarization. They are also a place where you aggregate traffic flows and apply traffic policies. When designing an EIGRP network, the main question is how many layers you should have. While EIGRP is not limited to several layers, typical EIGRP designs implement either two or three layers. The geographic dispersion of the network has a great effect on the number of layers. You should strive towards two layers for small, contained networks and use three layers for networks with greater reach topology depth, which is the maximum number of hops from one edge to another. also dictates the number of layers. Choose more layers for topologies with greater topology depth. The more complex the design, the more splitting the network up into zones will help the design. It is easier to do efficient traffic engineering with a two-layer design. Resource Restriction Policy Refer to Three-Layer Designs: In the end, you have to decide upon a design that balances simplicity, optimal routing, and functional separation. Now, in an EIGRP two-tired hierarchy, EIGRP can either be two- or three-tired, and in this case, the core would be where high-speed switching is happening. The aggregation layer is where we are going to do it, and in a dual layer, we are going to actually aggregate the user attachment point at the aggregation layer. Any information about the network that needs to be hidden should be either summarised or hidden using some other technique, like security policies like layer two and layer three security policies, to limit the traffic. Now, if you are going to use an EIGRP three-layer hierarchy, you have got the core of the distribution and the access layer. The core is again responsible for the high-speed switching, and summarization aggregation is happening at the distribution layer, and then the endpoint would be at the access layer. In this design, the address summarization would be applied to the choke point, and any policies would be at the access layer. Now, another thing the Tiger uses for fast convergence is a "graceful restart" or NSF. So in the past, whenever a device went down, the peers would actually send out a query if they didn't have a feasible successor, and they would look for an alternate route, any alternate route. And if I'm using a process such as Graceful Restart or NSF, they will resynchronize and recalculate more efficiently than they normally would. And this process speeds up the conversion on your network and allows the routers to forward data packets along known routes even when there is a failure and the network protocol or the routing protocol is being restored. EIGRP's fast convergence EIGRP was designed to achieve subsequent convergence. The key factor for EIGRP convergence is the presence or absence of a feasible successor. When there is no feasible successor, EIGRP uses queries to EIGRP peers and has to wait for responses. The use of queries slows down the convergence. To achieve EIGRP-fast convergence, you need to design your network properly. Summarization helps limit the scope of EIGRP queries, indirectly speeding convergence. Summarization also shrinks the number of entries in the routing table with various CPU operations. The effect of CPU operations on convergence is far less significant than the presence or actions of a feasible successor. A recommended way to ensure that a feasible successor is present is to use equal-cost routing. EIGRP metrics can be tuned using the delay parameter. However, adjusting the delay on links consistently and tuning variants are next to impossible to do well at any scale. It is difficult to put an exact boundary on the number of EIGRP neighbours that a router can support because this approach depends on the proper use of summarization, route filtering, and stub routing. A network with 500 peers can converge quickly, while a purely designed iGRP network with 20 peers might experience severe routing instability.

2. Examine EIGRP Autonomous Systems and Layered Designs

Implementing multiple EIGRP autonomous systems is sometimes used as a scaling technique. The usual rationale is to reduce the volume of EIGRP queries by limiting them to one EIGRP autonomous system. But there are issues with multiple EIGRP autonomous systems. The following is an external route for redistribution: The route is transferred from RIP to AS 200. Router A redistributes the route into an S 100. Router B receives this route from both a S 100 and a S 200. The same route is learned through separate routing processes. The route that is installed first is preferred. External route redistributing brings up a potential issue. A route is redistributed from RIP into S 200 in the diagram. Router A redistributes it into a 100. Router B hears the route prefix and advertisements from both as 208 hundred. The ad is the same because the route is external to both autonomous systems. The route that is installed into the EIGRP topology database first is placed into the routing table. When Router C sends an EIGRP query to Router A, Router A needs to query its neighbors. Router A sends a reply to Router C because it has no other neighbours in a S 100. However, Router A must also query all its neighbours in A 100 for the missing route. These routers may have to query their neighbors. Router C sends an EIGRP query to Router K. Router A needs to query all of its neighbours, including the neighbours in 100. Router.K responds on behalf of a S 200. Router A sends a query in 100. If the timing is right, Router B will have already received and replied to the query from Router C, so it would tell its 800 neighbours that it has no alternative path. The query was not stopped, it was just delayed along the way. Router A responds quickly to the query from router fee. But Router Ray still needs to wait for a response to its query in 800. Having multiple autonomous systems does not stop queries—it just delays them on the way. Using multiple EIGRP autonomous systems as an EIGRP query limiting technique does not bear fruit. To contain queries, use general scaling methods such as summarization distribution lists and stubs. Multiple EIGRP autonomous systems There are several reasons to use multiple EIGRP autonomous systems, but careful attention must be paid to limiting periods. Several valid reasons for having multiple EIGRP autonomous systems include the following migration strategy after a merger or acquisition. Although this option is not a permanent solution, multiple autonomous systems are appropriate for merging two networks. Over time, different groups administer different EIGRP autonomous systems. This scenario adds complexity to the network design, but it might be used for different domains of trust or administrative control. Organizations with very large networks may use multiple EIGRP autonomous systems as a way to divide their networks. In general, summary routes at a boundary are used to contain summarising blocks of prefixes in very large networks and to address the EIGRP query propagation issue. EIGRP two-layer hierarchy The EIGRP hierarchy is divided into two layers: core and aggregation. In an EIGRP two-layer hierarchy, the core performs high-speed switching, and the aggregation provides user attachment points. The core layer moves traffic from one topological area of the network to another. It performs high-speed switching of packets, so you could avoid the application of complex policies in the core. You should avoid reachability and topology aggregation inside the core itself. Core routers should summarise routing information toward the aggregation layer, so the fewer routes that are advertised towards the edge, the better. You can implement routing policies to control how many and which routes are accepted from the aggregation areas. The Aggregation Layer provides user attachment points. Information about the edge should be hidden from the core. using summarization and topology-adding techniques. You should place the traffic acceptance and security policies at the edge of the network. Use layer two and layer three filters to enforce the policies. EIGRP three-layer hierarchy The three-layer EIGRP hierarchy applies three layers: core, distribution, and access. In an EIGRP three-layer hierarchy, the core layer is responsible for high-speed switching. summarising and degradation are in the distribution layer, and policies are in the access layer. As in the two-layer hierarchy, the core layer moves traffic from one topological area of the network to another. It performs high-speed switching of packets, so you should avoid the application of complex policies in the core. You should avoid reachability and topology aggregation inside the core itself. Address summarization occurs at choking points between the distribution and core layers and between the distribution and access layers. Do not summarise between distribution layer routers within the distribution layer. You can implement routing policies to control how many and which routes are accepted, the access areas, and which route will be passed to the core. Aggregate traffic in the distribution layer as much as possible. You can do traffic engineering by directing traffic to the best core entry points and performing traffic filtering. The access layer provides support for users. You should place traffic acceptance and security policies in the access layer, filtering unwanted layer two and layer three traffic. You should consider configuring access layer routers as EIGRP steps deeper. Hierarchy does not change fundamental three-layer design concepts. Use the distribution layer as a blocking point for queries. Provide minimal information to the core and access layers.

3. EIGRP HUB&SPOKE AND STUB DESIGNS

Hub and spoke networks are often applied when connecting multiple branches to a single headquarters. Spokes or branches communicate with other spokes through the hub or headquarters. As a result, the hub is an ideal location for aggregating reachability and topology data. One of the most significant convergence-instability problems with EIGRP is the problem with the propagation of queries. When a router in an EIGRP autonomous system does not have a feasible successor or a viable alternative route, it sends out a query to the neighbours looking for any route to the destination network. So we want to limit the number of queries that occur in our EIGRP networks by using summarization subrouting and filtering, or by limiting query propagation and minimising convergence. So in this case, when router C sends an EIGRP query to router A, router A needs to send it to its neighbors. Router A then sends a query to router C because it has no neighbours in the autonomous system, and router A must also query all of its neighbours for the missing route, and these routers may have to query their neighbors. So you see there is a lot of querying going on, and the routes need to wait for the replies to these queries. It is no longer waiting for those responses that could cause problems. So the design can eliminate some of this convergence issue. When connecting multiple branches in a remote site, it is sometimes a good idea to convert the remote site routers into routers and a hub router at the main site. In doing so, we can summarise the spoke networks towards the hub or the core of the network and use a default route because they are only going to need to get to the hub or the core and then route accordingly. This can save on convergence time, and we also want to consider using a slash 31 on a point-to-point link to save and conserve addresses in an EIGRP hub and spoke design. Spokes only communicate through the hub. The hub advertises only the default route from the hub to the spokes. spoke networks towards the core are summarised Try to address links out of statusspace for easy summarization and, alternatively, filter link subnets from being advertised. Because the hub is the only point through which spokes can reach other networks, they advertise only the default route from the hub to the spokes. You should summarise the spoke network from the hub toward the core. When the topology is built with point-to-point links, consider using 31 subnets for the links to conserve address space. Address the links outside of the address space that is available on the spoke to allow for simple summarization. If this action is not possible, consider using a distributed list to filter link subnets from being advertised back and forth across the network. RFC 391 (three, two, one) describes the usage of 31 bet prefixes on IPV. Four point-to-point links The simplest way to explain it is that the use of a 31-bit prefix created by applying a 31-bit subnet mask to an IP address allows the all-zero and all-one IP addresses to be assigned as host addresses on point-to-point networks. Before RSC 321, the longest in common use on point-to-point links was 30 bits, which meant that all 0s and all 1 IP addresses were wasted. EIGRP scalability and hub and spoke topology depend on several factors. When spokes are connected to the hub over multiple interfaces, the processor is the primary limiting factor. With point-to-multiple topology over a single interface, the primary limiting factor is cue congestion. EIGRP has a theoretical limitation of 4000 peers per interface when they are in the same prefix. EIGRP hub and spoke scalability relies on several factors. You must configure spokes as stubs. You must minimise advertisements for spokes. You must test topology failover convergence after the primary hub sales. EIGRP is used in production environments where over 800 EIGRP neighbours are seen from one-point topologies, with over 1400 EIGRP neighbours successfully running in the lab. These numbers, however, can only be achieved with a careful design. Stubs are a must in an EIGRP, hub-and-spoke topology. You will not be able to build a network of over 100 iGRP neighbours that will convert if spokes are not configured as thumbs. Another key to scalability is summarization, which minimises advertisements so that spokes advertise either the default route or a carefully selected group of summarised networks. Remote sites can be dual-homed with two routers on the remote end, each connected to one of the hub routers. Both routers are configured as sub routers. They advertise connected and summary networks, but they do not advertise the routes that they learn from their neighbors. What happens when the link between routers B and D fails? Router C receives the 10 1 100:24 route from router D, but it does not advertise it to router A. because stub routers do not advertise the learned route network as reachable from the hub. Although one of the redundant connections between the hub and spoke is alive, a similar problem occurs in the opposite direction. Router C is a stub, so it is not advertising the default route to router D. As a result, router D has no connectivity with the hub. You can solve this issue by allowing the tutors to advertise a subset of their learned routes. You can do this with stub licking, in which you can permit stub router C to advertise the learned 101:100:24 route towards the hub, router A, and the default route towards the router D. You have established a fully redundant apology king while keeping the hub router out of the stump.

4. Describe EIGRP Convergence Features

The process of network convergence relies on the speed with which a device on the network can detect and react to a failure of one of its own components or a component in a routing protocol. Beer-layer-two failure detection times can vary widely depending on the physical media of the intervening devices. For example, the Ethernet switch can hide layer-two failures from the routing protocol peers. Bi-Directional Forwarding Detection, or BFD, can provide fast failure detection times for all media types. encapsulates topologies and routing protocols. In the best-case scenarios, it can provide fast failure detection in sub-50 milliseconds. BFD is a liveliness detection protocol. It does not determine the correct reaction to a detected failure. BFD can be used at any protocol layer, but it is particularly used as a routing protocol by routing protocols like EIGRP, OSPF, the Intermediate System, and the Border Gateway Protocol. There is one BFD session per client protocol. If a BFD device fails to receive a BFD Control package within the detect timer, it informs its client protocol about the failure. The client protocol determines the appropriate response. Routing Protocol Peering Session Termination verifies connectivity between two systems. BFD control packets are always sent as unicast packets to the BFD peer. The Cisco BFD implementation encapsulates VFD control packets in UDP packets using destination port 3784 EIGRP informs the BFD process of the IP address of the neighbour that it needs to monitor. BFD does not discover its peers dynamically. It relies on the configured routing protocols to tell it which IP addresses to use and which peer relationships to form. BFD on each router forms a BFD control packet. These packets are sent at a minimum of 1-second intervals until every session is established. After the remote router receives a BFD Control Packet during the session initiation phase, it copies the value of the "My discriminator" field into its own "Your discriminator" field and sets the "I hear you" bit for any subsequent BFD Control package that it transmits. The session is established when both systems see their own discriminators in each other's control packets. Both systems send at least 1-second intervals until they see the appropriate discriminators in each other's BFD control packets. When the BFD session is established, BFD timers are negotiated. These timers can be renegotiated at any time during the session without causing a session reset. BFD timers can be negotiated asynchronously: one peer may be sending BFD control packets at 50 millisecond intervals in one direction, while the other peer is sending its BFD control packets every 150 milliseconds in the other direction. As long as each BFD pier repair receives a BFD Control Package within the detected timer period, the BFD session remains up. Any routing protocol that is associated with BFD maintains its adjacencies. If a BFD peer does not receive a control package within the detect interval, it informs any routing protocol in this BFD session about the failure. It is up to us to determine the appropriate response to this information. The typical response will be to terminate the routing protocol peering session and reconverge, bypassing the failed peer. EIGRP. Graceful Restart and Cisco NSF Fundamentals When a networking device restarts all routing, peers associated with the device detect that the device is down and routes from this pair are removed. The session is reestablished when the device completes the restart. This transition results in the removal and reinsertion of routes, which could spread across multiple routing domains. Router A is using RouterB as the successor for several routes in this example. Router C is the feasible successor. Normally, router B would not tell router A if the EIGRP process on router B was going down. Router A would have to wait for its entire period to expire. Packets that are sent during this time would be lost with a graceful shutdown. The goodbye message page was broadcast. when an EIGRP routing process shuts down. It informs adjacent peers to synchronise and recalculate more efficiently than would normally occur.

Hide

CCNP Enterprise ENSLD (300-420): Designing OSPF Routing

1. Designing OSPF Routing

Hello and welcome to designing OSPF. This section contains a variety of in-depth best practises to assist you in designing your OSPF network in such a way that it can grow and provide skillets while also reducing network and routing. The first topic is about scalability and backing off adjacencies, with a brief discussion on link state advertisers and the Link state database, as well as the positioning of the designated router in various topologies. Now we move to the area and domain routing design where we discuss how scalability is impacted by several considerations regarding the number of prefixes, the number of adjacent neighbors, the stability of connections, the number assigned to a router, and the number of routers contained in an area. An explanation of the short-first Dijkstra's algorithm follows. In addition to the relationship between event processing and LSA, there are three Ltimers to throttle and regulate the number of SPF calculations that are performed when dealing with lability challenges. The section will tell you to keep in mind that there are three resources that must be considered. Those are memory, CPU, and bandwidth. The next topic revisits the impact that area routers have on scalability regarding the amount of information flooded into an area in the LSAs, and you will read about the maximum transmission unit (MTU) rule relating to fragmentation. The section transitions to how good backbone design can alleviate single points of failure, and this second part is the discussion of configuring the area border routers. The means to reduce the update traffic are provided by the two- and three-layer hierarchical topologies in terms of the core distribution and access. There are recommendations regarding the configurations and placement of area border routers as well as the incorporation of VGP to connect different Do OSPF domains. Some mechanics are described with reference to root summarization and how LSAs are transported between areas and the backbone. There is also information on how to configure the most effective addressing scheme, the discussion lanes, and how external nonospf roots should be injected and summarised into the OSPF network. The next topic takes you through the point-to-point and point-to-multipoint hub and spoke topology design, into full mesh and partial mesh recommendations, and utilising Isto is to counter full mesh flooding. You will read about how reducing update traffic impacts scalability by configuring Stubby, Totally Stubby, and Nostubby areas, as well as configuring the hub and spoke book interfaces to be specific network types. Now, these recommendations relate to the point-to-point and point-to-multipoint topics. You have read that convergence is critical to the operational uptime of your network. to complement and boost Kenneth. Another protocol that can be configured to work with OSPF and reduce failure detection time is called bidirectional forwarding detection. BFD provides sub second detection of Alink failure and reduces CPU resources. Now, Cisco has also implemented an exponential-back algorithm in the router iOS, along with tuneable parameters. Additional parameters included in the SPF standard are flood reduction, which eliminates the periodic refresh of unchanged LSAs. Another feature that reduces excessive LSA is OSPF database overload protection. Now, I trust that you can quickly see that this section provides a rich and robust group of design recommendations along with exposure to a number of features that can assist you in designing your OSPF network in such a way to allow your work to grow and provide scalability. That is an overview of designing OSPF routing. Thank you.

Hide