<p dir="ltr">Hi Joel,</p>
<p dir="ltr">Announcing the /24 beside the /32s from each node isn't an option? Metric could be lowered too.</p>
<p dir="ltr">Rgds, Stefan <br>
</p>
<br><div class="gmail_quote">Joel Brunenberg <<a href="mailto:ml@jjim.de">ml@jjim.de</a>> schrieb am Fr., 17. Apr. 2015 00:22:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi Hans, Hi List,<br>
<br>
thank you for your clarification,<br>
<br>
On Thu, Apr 16, 2015 at 06:27:30PM +0200, Hans van Kranenburg wrote:<br>
> Hi,<br>
><br>
> On 16/04/15 16:43, Joel Brunenberg wrote:<br>
> >[...]<br>
> ><br>
> >Another idea was a static route for the containing /24 that would be<br>
> >exported via BGP while the /32 were filtered away. That however leaves<br>
> >the /24 route in place even if there are no more /32 routes left. If<br>
> >there is a way to change that, that would be great.<br>
><br>
> Yes, I would suggest using a static blackhole route with the large aggregate<br>
> range for the BGP connection, and also export that to your local kernel<br>
> routing table to automatically drop traffic for all addresses that are<br>
> currently not in use.<br>
<br>
It seems I concentrated so much on the aspect of making the appearance<br>
of the route conditional that I missed the obvious point. What you say<br>
makes total sense.<br>
<br>
> Is there a special reason why you would want to have the /24 being removed<br>
> entirely if none of the other connections is up?<br>
<br>
I would like to retract the route in the case, no contributing routes<br>
are there so traffic is no longer attracted in that case. If that is not<br>
possible, a static blackhole route would be ok, too I guess.<br>
<br>
So its correct to assume that the following things would do?<br>
<br>
* in protocol static add a blackhole route for the aggregate network<br>
* export the aggregate route to the kernel<br>
* export the aggregate route to be BGP peers with next-hop self<br>
* NO NOT export the aggreage to the OSPF areas<br>
* Be happy and dance around the table<br>
<br>
I will have a try and see if it works. If there was a way to<br>
conditionally import a specific protocol or filter on the presence of<br>
specific prefixes in the RIB that would be awesome.<br>
<br>
Thank you so far,<br>
<br>
Joel<br>
<br>
/jbn<br>
<br>
--<br>
Joel Brunenberg - Troisdorf<br>
PGP: 0xAD25981C<br>
</blockquote></div>