Now and again there’s a need to run more than one routing protocol and have more than one routing area: multivendor shops, relocation starting with one protocol then onto the next, versatility issues of a solitary protocol, political or individual inclination, creation versus test systems, mergers, and acquisitions.
How to Reach Devices in Other Domains with IGP Route Redistribution
Redistribution is the way toward passing routing data starting with one routing protocol then onto the next to have reachability for gadgets that live in various routing spaces. Each routing protocol will contribute one of a kind data into the routing tables inside its space, however there can be a craving or need to achieve gadgets in another area. Redistribution is done on at least one limit routers between a source routing area or protocol into a target space or protocol.
There are three options to get full reachability between spaces:
- Default routes from a limit switch. You can pass a default course from a switch that touches all routing spaces (limit switch) to those routers that exclusive take part inside one area (internal switch). That would cover obscure routes from any area that the internal routers are unconscious of and have the internal routers forward to the limit switch, which would have a total routing table since it would take part in all the routing areas. This procedure works best if there is just a single purpose of contact between the routing areas.
- One-path redistribution, with a default. At least one limit routers pass a default course into one space, yet redistribute into another area. Ordinarily, you would pick a center protocol to redistribute into and alternate protocols that get the default course would be taken a gander at as edge protocols. One-way redistribution is utilized to scale up to bigger quantities of routes, for example, in a substantial multinational organization. The center protocol could be BGP (Border Gateway Protocol) and the edge protocol(s) could be any IGP (Interior Gateway Protocol, for example, OSPF, EIGRP, RIP or IS-IS, or even various occurrences of the same IGP. It functions admirably with acquisitions and mergers, since the “new” some portion of the organization doesn’t need to run an indistinguishable routing protocol from the rest, or need to change for the time being. It just adds an association with the center.
- Two-way or shared redistribution passes a few or the majority of the routing data of one protocol into another. This is the most complex option, particularly if there is more than one purpose of contact between the routing areas. It ought to be utilized when there are goals that should be reachable starting with one space then onto the next. Be that as it may, particular approach must be connected to show how the movement achieves those goals or how the activity must be dealt with in light of security strategies. The regular worries with two-way distribution are routing loops, asymmetric routing and suboptimal routing.
- Asymmetric routing is the place the sending path is not quite the same as the arrival path. Issues can emerge if there’s a security strategy set up for how activity is sent or if there are an arrangement of firewalls set up. Stack balancers can likewise be disturbed by asymmetric routing. Stack balancers, which appropriate load to particular gadgets in light of a common address, expect the sending and return paths to be predictable.
- Suboptimal routing is the place the most favored path in a sending table is not by any means the most direct course. This happens when the limit switch “listens” about routes from the starting protocol and furthermore through another routing protocol as an external course. On the off chance that the regulatory separation for the external course protocol is more dependable than the beginning protocol, the switch will lean toward the external course over the local course. The settle is to control the authoritative separation of the routes being referred to. This is not the most straight-forward process and differs from stage to stage, even inside a solitary merchant’s product offering.
- Routing loops, or an input loop, can happen when the routing data is redistributed into one protocol at one purpose of contact and after that redistributed once again into the beginning protocol at another purpose of contact. With a specific end goal to settle a routing loop, you need to make an input channel. The channel would deny the routes beginning in the target protocol from being promoted once more into that same protocol. You need to fabricate the channel for course. For instance, on the off chance that you have OSPF and EIGRP areas associated at least two focuses, you would construct one channel for all the OSPF routes and channel them on the redistribution from EIGRP into OSPF. Another channel would be worked for all the EIGRP routes, sifting them on the redistribution from OSPF into EIGRP. This separating must be done on all limit routers between the two protocols to be viable. You can coordinate on the prefixes or you can utilize labels, which is my inclination. The labels can be doled out as a component of the way toward redistributing the routes into the target protocol, then you can search for the labels to channel on. You need to make a strategy that first searches for the labels and denies them and on the off chance that they’re not there, then labels the routes to recognize the source protocol. This is accomplished for bearing, so two polices would need to be made. All routing protocols, including RIPv2, can bolster labels.
We should take a gander at IGP course redistribution on Cisco gadgets. When redistributing from one protocol into another, there are a couple of things to recall:
- The redistribution procedure pulls from the routing table, not the protocols database. On the off chance that you will redistribute RIP into OSPF, then the procedure searches for those routes named as RIP in the routing table. There is one special case: the associated routes that the protocol is running on.
- On the Cisco routers for IPv4, the associated routes will naturally redistribute too. This is valid the length of you don’t redistribute when associated into that same target protocol, which would bring about the element to stop.
- On the Cisco routers for IPv6, the redistribution procedure does not redistribute those associated routes that the protocol is running on, unless you include the incorporate associated option the redistribution line.
Some Cisco OS requires an arrangement connected to the redistribution command for routes to be passed starting with one protocol then onto the next. When redistributing into a protocol, you need to supply metrics for the routes so they’re in the right organization in respect to the target protocol. The metric for one protocol doesn’t really search amend for another. There is a seed metric that must be attached to the external routes going into the target protocol. The table in Figure 1 shows every protocol with some slight varieties.
|Source||into RIP||into EIGRP||into OSPF||into IS-IS||into BGP (MED)|
|Connected||1||Interface metric||20 (E2)|
|Static||1||Interface metric||20 (E2)|
|RIP||Infinite||20 (E2)||IGP metric|
|EIGRP||Infinite||Other process metric||20 (E2)||IGP metric|
|IS-IS||Infinite||Infinite||20 (E2)||IGP metric|
Figure 1: Protocol Variations
On the off chance that the seed metric is unending, the course is not useable. You need to supply the seed metric while redistributing the source protocol into the target either on the redistribution line or through the default metric command under the target routing protocol. The seed metric is in the arrangement for that target protocol: jumps for RIP, cost for OSPF and IS-IS, and the composite metric for EIGRP (data transfer capacity, delay, unwavering quality, load and MTU).
One final thought—if the source is BGP, then just the external BGP routes will be redistributed into the IGP. This is a loop counteractive action system. In the event that you have to redistribute the internal BGP routes, then arrange under the BGP procedure (not the target protocol) bgp redistribute-internal command.
Along these lines, in case you’re running more than one routing protocol and you require full or incomplete reachability, you’ll need to redistribute between those protocols. There are a couple of things to consider and get ready for before you begin arranging. Redistribution can be exceptionally basic (one sets of protocols, one purpose of contact) and can be extremely unpredictable.