Re: OSPF ADV Router is not-reachable

Hi Hemanth,

thanks for your reply...

this is a lab so no real gear but just gns3.

I've relabbed this again and it's clear that SW1 sees R3 as not-reachable
because I've shutdown the interfaces on R3.
Actually, (just to make sure) I've isolated this nssa area from every other
area (there were 2 exit point).
I've clear all ospf processes but despite that I still see the R3 in
not-reachable state.
I got to reboot the router in that area to fix this.. at this point I think
it's just a GNS3 issue. I cannot lab it with real gear at the moment.

thanks for your help.

On Thu, Dec 13, 2012 at 3:19 PM, HEMANTH RAJ <hemanthrj_at_gmail.com> wrote:

> Hey
>
> There are few reasons in OSPF that you can state when u see ADV router is
> no reachable.
>
> 1. ur forwarding address ( present in TYpe 5 or Type 7 ) may be reachable
> via a n external route.
>
> 2. Network Type mismatch
>
> 3. One side is a Unnumbered link and another is a numbered link
>
> 4. Wrong subnet on Point to Point links ( It can still form neighborship
> even though they r in different subnets as the rule is relaxed in Point to
> Point links )
>
> So in ur case, lets see that u r reloading or doing a switchover or
> restarting the ospf process on R3 would surely kill the OSPF session with
> ur peer. So ultimately u should not see any OSPF neighbor
>
> Are u running any NSF ( Graceful Restart ) with ur neighbor. So for
> example when u enable this, then may be ur neighbor may hold down the
> routes in FIB and mark the entries as stale and then wait for the Stale
> timer to expire to flush the routes from FIB.
>
> If u enable NSR ( Non Stop Routing ) , then Grace LSAs will be sent to the
> neighbor when it comes back again to ensure that the database is in syn
> after the reload.
>
> Which router are u using and did u check the Routers Docs or how they have
> implemented it
>
> Ideally when u shut the link, the ISM ( interface state manager ) will
> inform its clients ie OSPF to shut the peer on this interface and
> ultimately its flushes the routes for that peer.
>
> I am still wondering is there any way that it will retain the route and
> then gives this message "ADV is not reachable"
>
> Can u verify it once more and let me know
>
>
>
> On Thu, Dec 13, 2012 at 5:25 PM, ccie99999 <ccie99999_at_gmail.com> wrote:
>
>> why by design it's not?
>>
>
>
>
> --
>
> Regards
> Hemanth Raj
> CCIE#28593 (R&S)
>
>
>

-- 
@ccie99999
Blogs and organic groups at http://www.ccie.net