BGP intro - WIP

bgp-intro/README.md

@@ -0,0 +1,97 @@
+BGP
+===
+
+Blablabla, work in progress here.
+
+The next routing protocol to get some attention would be BGP.
+
+BGP is not an alternative to OSPF. It's used for different things.
+
+To start off quickly, here's the bare essentials:
+
+![BGP network, barebones](/bgp-intro/bgp-heythere.png)
+
+When routers talk BGP to each other, they just claim that some network ranges are reachable via them. Voila.
+
+## OSPF vs. BGP
+
+Ok, a bit less simplified:
+
+![BGP network, less simplified](/bgp-intro/bgp-hey2.png)
+
+ * OSPF: routes in the network are originated by putting ip addresses on a network interface of a router, not manually defined. These are addresses and subnets that are actually in use.
+ * BGP: publish "umbrella" ranges, there is no actual proof that the addresses are actually in use.
+
+So, OSPF is a network in detail, and BGP can connect that network to a network of another external someone, just telling them your aggregated ranges. Other end does not have any knowledge about the internal topology of your network, and changes in there do not result in updates to external networks.
+
+ * Between routers: small subnet, like an IPv4 /30 or /31, which only contains the two routers.
+ * Often direct links, in this bird/openvswitch tutorial just use a vlan
+ * A complete network under control of somebody has an AS number, an Autonomous System
+ * Where OSPF shortest path deals with knowledge about all separate routers, paths and weights, BGP just looks on a higher level, the shortest path, considering a complete network being a step.
+
+![BGP network, three ASses](/bgp-intro/bgp-hey3.png)
+
+But, later. First of all, do it with bird, build full example.
+
+## BIRD BGP
+
+![BGP and OSPF network](/bgp-intro/bgp-ospf.png)
+
+Whoa, such network.
+
+Hopsa, clone some containers, copy paste configuration
+ * already provide bird config with ospf for internal network
+
+Zoom in on R3, R10 and add BGP configuration
+ * add ebgp protocol
+ * meh, umbrella route is not in bird routing table
+ * how to add it? static proto, but adding it will result in kernel proto pushing it into FIB
+ * keep route administration of ebgp peer in separate table. whoa, bird tables and protocols
+
+## Intermezzo: BIRD tables, protocols, import, export
+
+BIRD concepts:
+ * table: multiple routing tables (not like the linux kernel, it's just a collection of routes). e.g. there may be multiple paths to the same network in a table present.
+ * protocol: move routes around between tables (pipe) or into / out of BIRD (ospf, bgp)
+ * import: draw routes closer to BIRD
+ * export: push routes away from BIRD to the outside world
+ * table "master": central route table, import = pull to it, export = push away
+
+picture:
+
+    master <-- p_master_to_ebgp_r10 <-- t_ebgp_r10 <-> ebgp_r10
+                                                 \ <-- originate_to_r10
+
+ * t\_ebgp\_r10 holds route information about routes originates to r10 and routes that are learned from r10
+ * originate\_to\_r10 is a collection of routes that is advertised by this network
+ * pipe to master forwards all routes that are learned from the ebgp session
+ * master can just export all to kernel
+
+## Let's do it
+
+ * show BIRD config for pipe and table and ebgp
+ * reload, show proto, show route export p_blah etc
+ * see that routes are imported from the other end
+ * see that they end up in master table and in ip r
+
+## Connecting the internal network, not only edge
+
+ * now ebgp router knows route to external network
+ * how to tell the ospf network?
+ * export the bgp routes to ospf? no!
+ * explain ibgp -> tell non-edge-routers about external connectivity
+ * ibgp: all routers must have ibgp session with edge routers
+ * KEEP BGP IN BGP
+ * do it! configure ibgp session between R0, R3
+ * R0 can just import the info and put it into kernel table
+ * wait what... bgp nexthop? -> first router *outside* the network
+ * BGP needs OSPF to provide internal nexthop towards external router -> recap: BGP trusts on OSPF behaviour
+ * small link nets for ebgp sessions need to be in IGP!
+ * make it work!
+
+## What about the third network?
+
+ * build R1,R2,R3 with 3 containers
+ * show AS-path
+ * disable link between routers, see traffic being redirected
+ * just put some address from the umbrella range on loopback, yolo