Join Telegram Study Group ▷
BGP Routing
Boundary Gateway Protocol (BGP) is the routing protocol of the Internet, which is used to route traffic on the road of the Internet. For this reason, it is a very important agreement, it is also the most difficult to understand the protocol.
From our overview of Internet routing, you should realize that routing on the Internet consists of two parts: the internal fine-granularity part managed by IGP, such as OSPF, and the interconnection of these autonomous systems (AS) through BGP.
Who needs to understand BGP?
BGP is associated with network administrators of large organizations connected to two or more ISP and (ISP), an Internet service provider connected to other network providers. If you are an administrator for a small business network or end-user, you may not need to know about BGP.
BGP basics
The current version of BGP is BGP version 4, based on RFC4271.
BGP is a path vector protocol that provides routing information to an autonomous system on the Internet through its AS-Path property.
BGP is a layer 4 protocol on top of TCP. It’s much simpler than OSPF because it doesn’t have to worry about what TCP will deal with.
Peers that have been manually configured to exchange routing information will form a TCP connection and begin to say BGP. No discovery in BGP.
Medium-sized enterprises usually enter BGP, to achieve true multiposting of the entire network.
An important aspect of BGP is that AS-Path itself is an anti-cyclic mechanism. Routers do not import any of their own routes in AS-Path.
Note: BGP is a hot topic in the IT industry. If you want to know more details and you can contact SPOTO with 100% valuable study materials. Besides, SPTOP is committed to training the Cisco certification.
Why do you need to understand BGP?
When BGP is not configured correctly, it can lead to a number of availability and security problems, as Google found in 2008 that its YouTube services did not have access to most of the Internet. What happened was to ban YouTube, Pakistan Telecom from using BGP to route YouTube address blocks to black holes in its own country. However, in what is considered accidental, this routing information is somehow transmitted to Pakistan Telecom Hong Kong ISP and spread from there to the rest of the world. As a result, most of YouTube’s traffic ended up falling into Pakistan’s black hole.
A number of BGP hijacking attacks occurred in 2003, in which the modified BGP routing information allowed unknown attackers to redirect a large number of traffic in order to transmit it through a Belarusian or Icelandic router before transmitting it to a predetermined destination.
Obviously, BGP is very important. Here, we will briefly outline how BGP works, as well as the problems it solves and causes.
Autonomous systems
The first is a small term. In the world of BGP, each routing domain is called an autonomous system or AS. What BGP does is to help choose the path through Internet, usually, the route that traverses the least number of autonomous systems: the shortest AS path.
For example, if your company is connected to two large ISPs, BGP may be required. To use BGP, you need an AS number that you can obtain from the American Internet Number Registration Authority (ARIN).
When BGP is enabled, your router will extract a list of Internet routes from your BGP neighbors, in which case these routes will be your two ISPS. It then examines them carefully to find the path with the shortest AS path. These will be placed in the routing table of the router. If you only connect to a single ISP, then you don’t need BGP. This is because there is only one path to the Internet, so there is no need for routing protocols to choose the best path.)
Generally, but not always, routers will choose the shortest path to an AS. BGP only knows about these paths based on updates it receives.
Route updates
Unlike the Routing Information Protocol (RIP), a distance vector routing protocol that uses hop count as the routing metric, BGP does not broadcast its entire routing table. At start-up, your peer will hand over the entire table. After that, everything depends on the updates received.
Routing updates are stored in the routing information base (RIB). The routing table will store only one route per destination, but RIB usually contains multiple paths to the target. It is up to the router to decide which routes will enter the routing table and which paths will actually be used. If the route is revoked, another route can be obtained from RIB to the same location.
RIB is only used to track routes that may be used. If a route extraction is received and it exists only in the RIB, it is quietly deleted from the RIB. No updates were sent to peers. RIB entries never time out. They will continue to exist until it is assumed that the route is no longer valid.
BGP path attributes
In many cases, multiple routes will be routed to the same destination. Therefore, BGP uses the path attribute to decide how to route traffic to a particular network.
The easiest thing to understand is the Shortest AS_Path. This means traversing the path of the minimum number of AS "victories"
Another important attribute is Multi_Exit_Disc (multiple exit discriminator or MED).) This tells the remote AS, that if there is more than one exit point on the network, the specific exit point is preferred.
The Origin attribute specifies the origin of a routing update. If BGP has multiple routes, then the origin is one of the factors in determining the preferred route.
BGP issues
In order to really understand how BGP works, it is important to spend some time discussing the issues that haunt the Internet.
First of all, we have a lot of problems with routing table growth. If someone decides to dissolve a network that used to be a single / 16 network, they may start promoting hundreds of new routes. When this happens, every router on the Internet will get every new route. People will be forced to compress or combine multiple routes into a single advertisement. Aggregation is not always feasible, especially if you want to divide / 19 into two geographically separate / 20. Routes are shown on nearly 200000 routes, and for some time they look like exponential growth.
Second, people are always worried that someone will "advertise on the Internet". If a customer of a large ISP suddenly decides to advertise everything an ISP accepts these routes, all Internet traffic will be sent to the AS. of the small customer There is a simple solution. It is called routing filtering. Setting up filters is very simple so that your router does not accept routes from customers you do not expect, but many large ISP still accept "default" from peers that are unlikely to provide transmission.
Finally, let’s beat. BGP has a mechanism for low pressure seemingly sheet-like routing. Flaps or incoming routes are often not reliable enough to send traffic. If the route is turned over frequently, every time someone disappears and reappears, the load on all Internet routes increases due to update processing. Damping prevents the BGP peer from listening to all routing updates from the fan peer. At each flap, the amount of time under compression increases exponentially. When you have a wrong link, it’s annoying because it may take more than an hour to reach many Internet sites, but it’s very necessary.
This quick discussion of BGP should be enough to allow you to think about the right way of the protocol, but not comprehensive. If your task is to operate the BGP router, take some time to read the RFC. Your peers will be grateful.
About SPOTO
SPOTO offers Cisco CCIE certification practice tests and training to help all candidates to pursue their ideal careers. You can also find more useful tips and suggestions to help you study and master networking knowledge.
