I will make some other test, but i can't test on a real hardware.<br><br><div title="signature"><p><span style="font-size: small;"><strong>
</strong></span></p><strong><hr>
Luca Belforte</strong><p></p></div><div class="gmail_quote">On Mon, Aug 16, 2010 at 10:43 PM, Ben Greear <span dir="ltr"><<a href="mailto:greearb@candelatech.com">greearb@candelatech.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div class="im">On 08/16/2010 01:21 PM, Luca Belforte wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I'm using xorp.ct, i get on the git, no sure about the version but it<br>
was downloaded 3 days ago.<br>
</blockquote>
<br></div>
Thanks..as long as it's relatively current, it should be fine.<div class="im"><br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
The route on all routers are correct, traceroute block on the 2 hop. (If<br>
i'm pinging R1 to R8-R4 interface it stop on R2) but if i do a<br>
traceroute on R2 it works. All tables are correct and show the correct<br>
next-hop. Neighbors are all well established<br>
What is more amazing is that sometimes it works for let's say 5-10<br>
traceroute and after it stop again to work.<br>
</blockquote>
<br></div>
You might try: ip monitor (or similar command if you are not on Linux)<br>
and see if routes are flapping.<br>
<br>
Also, I'd start some sniffers on R2..if it receives a pkt, and has<br>
proper route, then it should forward it regardless of any other xorp<br>
behaviour, as xorp is NOT in the packet forwarding path for normal<br>
IP traffic.<br>
<br>
You might also sniff all interfaces to make sure it forwards out the<br>
correct one.<div class="im"><br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
The others routers have similar configuration.<br>
<br>
Well maybe is the virtualbox internal network who have problems. In any<br>
case for my test this is not really problematic.<br>
</blockquote>
<br></div>
It would be good to verify with real hardware or some different virtualization<br>
technique.<div><div></div><div class="h5"><br>
<br>
Thanks,<br>
Ben<br>
<br>
<br>
-- <br>
Ben Greear <<a href="mailto:greearb@candelatech.com" target="_blank">greearb@candelatech.com</a>><br>
Candela Technologies Inc <a href="http://www.candelatech.com" target="_blank">http://www.candelatech.com</a><br>
<br>
</div></div></blockquote></div><br>