OSPF protocol and export filters for E2 routes
Hi guys, I'm wondering if someone knows the answer to this, here is the setup... R1 => R2 => R3 - R1 advertises a route (in this case 0.0.0.0/0) to R2 (from static or kernel) - R2 accepts the route with "import all" - R2 has a filter for the route with "export my_filter;" and in that filter there is a "reject;" - R3 should not receive the route, correct? What I am seeing is this works 100% as expected in RIP. When I try the same thing with OSPF, I am not seeing it being rejected on R2. Infact, I am not seeing the E2 route at all pass through the export filter. But I am receiving it on R3. I just want to make sure this is expected or unexpected behavior? My use case is if R1 advertises a default route via OSPF, I was looking at filtering that default route out on R2 so it is not advertised to R3. -N
On Wed, Dec 09, 2020 at 05:48:18PM +0000, Nigel Kukard wrote:
Hi guys,
I'm wondering if someone knows the answer to this, here is the setup...
What I am seeing is this works 100% as expected in RIP.
When I try the same thing with OSPF, I am not seeing it being rejected on R2. Infact, I am not seeing the E2 route at all pass through the export filter. But I am receiving it on R3.
I just want to make sure this is expected or unexpected behavior?
Hi It is expected behavior. There is a fundamental difference between vector/path distance routing protocols (RIP, BGP) and link-state routing protocols (OSPF, IS-IS). The former pass around routes (and can decide what is passed in each step), the later pass around network topology description and can control propagation only on boundaries (like between OSPF areas). -- Elen sila lumenn' omentielvo Ondrej 'Santiago' Zajicek (email: santiago@crfreenet.org) OpenPGP encrypted e-mails preferred (KeyID 0x11DEADC3, wwwkeys.pgp.net) "To err is human -- to blame it on a computer is even more so."
participants (2)
-
Nigel Kukard -
Ondrej Zajicek