examples/bgpv4/BgpIpv6Basic: use addRemoteRoutes instead of manual on-link routes

Now that the configurator can install the directly-connected routes without
the precomputed remote routes, drop the hand-written on-link <route>
workaround: run the configurator with addRemoteRoutes=false so it
auto-generates the on-link (direct) routes and the host default routes,
leaving the inter-AS routes to be learned via BGP.

Only the graphical (tyf) fingerprint changes; tplx (all events), ~tNl
(on-wire traffic) and ~tND (packet data) are byte-identical, i.e. the
simulation trajectory is unchanged. hostA still receives 20/21 ping replies
once BGP converges (the first is lost during convergence).

Author: avarga

examples/bgpv4/BgpIpv6Basic/BgpIpv6Basic.ned

@@ -24,9 +24,9 @@ network BgpIpv6Basic
             parameters:
                 @display("p=100,100;is=s");
                 config = xmldoc("Ipv6Config.xml");
-                // no auto static routes: routers get only the on-link routes configured
-                // manually below, so the inter-AS routes can only come from BGP
-                addStaticRoutes = false;
+                // auto-generate the on-link (direct) routes and the host default routes,
+                // but NOT the inter-AS (remote) routes -- those can only come from BGP
+                addRemoteRoutes = false;
         }
         rA: Router {
             parameters:

examples/bgpv4/BgpIpv6Basic/Ipv6Config.xml

@@ -9,13 +9,7 @@
     <interface hosts="rB" towards="hostB" address="2001:db8:0:20::1"/>
     <interface hosts="hostB" address="2001:db8:0:20::100"/>
 
-    <!-- on-link (directly-connected) routes only; the inter-AS routes are left for BGP -->
-    <route hosts="rA" destination="2001:db8:0:1::"  prefixLength="64" interface="eth1"/>
-    <route hosts="rA" destination="2001:db8:0:10::" prefixLength="64" interface="eth0"/>
-    <route hosts="rB" destination="2001:db8:0:1::"  prefixLength="64" interface="eth0"/>
-    <route hosts="rB" destination="2001:db8:0:20::" prefixLength="64" interface="eth1"/>
-
-    <!-- hosts default-route to their gateway router -->
-    <route hosts="hostA" destination="::" prefixLength="0" interface="eth0" gateway="2001:db8:0:10::1"/>
-    <route hosts="hostB" destination="::" prefixLength="0" interface="eth0" gateway="2001:db8:0:20::1"/>
+    <!-- No manual routes: the configurator adds the on-link (direct) routes and the host
+         default routes automatically; the inter-AS routes are left for BGP (the configurator
+         is run with addRemoteRoutes=false) -->
 </config>

examples/bgpv4/BgpIpv6Basic/README.md

@@ -15,15 +15,15 @@ What it demonstrates / verifies:
 - IPv6 reachability is exchanged using the **`MP_REACH_NLRI`** path attribute
   (AFI=2 / SAFI=1), per RFC 4760 — rA advertises `2001:db8:0:10::/64`, rB advertises
   `2001:db8:0:20::/64`, each via a `<Network>` element in `BgpConfig6.xml`.
-- The learned routes are actually **installed and used**: the configurator adds only
-  on-link routes (`addStaticRoutes = false` + manual `<route>` entries), so the inter-AS
-  routes can come *only* from BGP. `hostA` pinging `hostB` therefore succeeds only once
-  BGP has converged.
+- The learned routes are actually **installed and used**: the configurator is run with
+  `addRemoteRoutes = false`, so it adds the on-link (direct) routes and the host default
+  routes automatically, but leaves out the inter-AS (remote) routes — those can come
+  *only* from BGP. `hostA` pinging `hostB` therefore succeeds only once BGP has converged.
 
 Notes:
 
 - The BGP Identifier is a 4-octet router id even for IPv6 (RFC 4271/6286); it is set
   explicitly via `bgp.routerId` since an IPv6 router has no IPv4 address to derive it from.
-- The manual on-link `<route>` entries are a workaround for a configurator limitation
-  (`addDirectRoutes` is only applied when `addStaticRoutes` is true); see
-  `plan/pending/configurator-decouple-direct-routes.md`.
+- `addRemoteRoutes` (and the matching `addManualRoutes`) let the configurator install the
+  directly-connected routes without the precomputed remote shortest-path routes that would
+  otherwise mask the protocol under test; see `plan/pending/configurator-decouple-direct-routes.md`.

tests/fingerprint/examples.csv

@@ -31,7 +31,7 @@
 /examples/bgpv4/BgpUpdate/,          -f omnetpp.ini -c General -r 0,                30s,             9552-83f7/tplx;3a06-a4f0/~tNl;62be-5688/~tND;0573-e0b4/tyf, PASS, ospf EthernetMac Ipv4
 /examples/bgpv4/BgpAndOspf/,         -f omnetpp.ini -c General -r 0,                1000s,           1e22-36e3/tplx;2b66-00b2/~tNl;430e-7064/~tND;de89-afd7/tyf, PASS, ospf EthernetMac Ipv4
 /examples/bgpv4/BgpAndOspfSimple/,   -f omnetpp.ini -c General -r 0,                1000s,           dc15-38aa/tplx;dacd-0cbd/~tNl;46d3-bcb3/~tND;8a59-2347/tyf, PASS, ospf EthernetMac Ipv4
-/examples/bgpv4/BgpIpv6Basic/,       -f omnetpp.ini -c General -r 0,                30s,             7ed5-3294/tplx;8d57-7c0c/~tNl;d6e7-ba7e/~tND;2e97-8361/tyf, PASS, EthernetMac Ipv6
+/examples/bgpv4/BgpIpv6Basic/,       -f omnetpp.ini -c General -r 0,                30s,             7ed5-3294/tplx;8d57-7c0c/~tNl;d6e7-ba7e/~tND;2e97-8771/tyf, PASS, EthernetMac Ipv6
 
 /examples/clock/,                    -f omnetpp.ini -c General -r 0,                10ms,            35c3-8325/tplx;0000-0000/~tNl;0000-0000/~tND;a4b7-bff5/tyf, PASS,