<div dir="ltr">I tried some additional debugging with this setup :<div>(Note : All interfaces are tap interfaces)</div><div><br></div><div>eth-0 (<a href="http://10.0.1.11/24">10.0.1.11/24</a>) - ospf and bgp configured</div><div>eth-1 (<a href="http://10.0.1.12/24">10.0.1.12/24</a>) - bgp</div><div>eth-2 (<a href="http://10.0.2.11/24">10.0.2.11/24</a>)</div><div>eth-3 (<a href="http://10.0.2.12/24">10.0.2.12/24</a>) - bgp </div><div><br></div><div>Result : OSPF is up and running, BGP is in established state</div><div><br></div><div>Changed OSPF to be on eth-1. It stays in alone state. (Packet capture shows hello packets being received from neighbor. Neighbor adds us(bird) and changes to init state but bird seems to just not receive packets . Doing a 'netstat -an' , the receive and send queue are empty inspite of packets being received on that interface.</div><div><br></div><div>Then when we change eth-0 to other subnet (<a href="http://10.0.3.11/24">10.0.3.11/24</a>) , OSPF on eth-1 immediately comes up , bgp on eth-1 is also in established state. But BGP on eth-0 times out.</div><div>I think the kernel does not like the interfaces in same subnet ? (just a guess here). </div><div>In my device protocol , i am just scanning interfaces every 10 seconds. I have a custom forwarding plane and so, dont use kernel protocol( that does not seem to be the issue here though).</div><div><br></div><div>Thanks,</div><div>JM</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Sat, Oct 10, 2015 at 6:15 AM, Ondrej Zajicek <span dir="ltr"><<a href="mailto:santiago@crfreenet.org" target="_blank">santiago@crfreenet.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">On Thu, Oct 08, 2015 at 06:48:19PM -0400, Jigar Mehta wrote:<br>
> Hi, I wanted some help regarding a problem I was seeing when i had multiple<br>
> interfaces with same ip address . I have multiple routing domains setup so<br>
> i have valid cases to have same ip address on different interfaces (so i<br>
> have seperate routing tables for each instance). When I do a show ospf<br>
> interface for each ospf protocol instance, it seems to correctly dump the<br>
> interface . When I did a packet capture, it shows incoming/outgoing hello<br>
> packets, but when I enabled debug, i can see that it never prints a "hello<br>
> packet received" message(but prints "hello packet sent") although, i can<br>
> see the peer sending ospf hello packets.<br>
> This seems to be a problem with only one instance and not both.<br>
<br>
</span>Hi<br>
<br>
I don't know. I just tested it, works for me. Does the other side see<br>
sent hello packets? (Should be in Init state.) If you disable the other<br>
interface with the same address, does the problematic iface start to work?<br>
<span class="HOEnZb"><font color="#888888"><br>
--<br>
Elen sila lumenn' omentielvo<br>
<br>
Ondrej 'Santiago' Zajicek (email: <a href="mailto:santiago@crfreenet.org">santiago@crfreenet.org</a>)<br>
OpenPGP encrypted e-mails preferred (KeyID 0x11DEADC3, <a href="http://wwwkeys.pgp.net" rel="noreferrer" target="_blank">wwwkeys.pgp.net</a>)<br>
"To err is human -- to blame it on a computer is even more so."<br>
</font></span></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature">Jigar Mehta<br><br><br></div>
</div>