Hi,
Please upgrade to newer version. I believe the issue is fixed in
version 0.14.1. See commit 78f2409d1b44b3321ed3dd6437da1b1b8994b126
which should fix the hub side issue.
Additionally the spoke should not react to multicast traffic
indications. I will probably commit soon the following as an additional
sanity check:
diff --git a/nhrp/nhrp_server.c b/nhrp/nhrp_server.c
index aa23199..df6ca6e 100644
--- a/nhrp/nhrp_server.c
+++ b/nhrp/nhrp_server.c
@@ -527,6 +527,9 @@ static int nhrp_handle_traffic_indication(struct
nhrp_packet *packet)
NULL, &dst))
return FALSE;
+ if (nhrp_address_is_multicast(&dst))
+ return FALSE;
+
/* Shortcuts enabled? */
if (packet->src_iface->flags & NHRP_INTERFACE_FLAG_SHORTCUT) {
nhrp_info("Traffic Indication from proto src %s; "
On Thu, 11 Jun 2015 18:16:01 +0300
Alex Levit <alexl.glo...@gmail.com> wrote:
> Ver is 0.13, some patches were added due to problems I reported to
> this forum in the past and the changes I think were latter submitted
> upstream as well.
> I went over the changes and did not really see the relation to the
> current problem (it was related NBMA address).
> Concerning how it breaks:
> First messages are exchanged between spoke and hub, but then spoke
> stops transmission. The counter of tx counter of mgre interface is
> not advancing when multicast is sent by Quagga.
> Another possibly important input is that I am running opennhrp inside
> a namespace implementing VRF-lite.
> Your support and recommendations are highly appreciated.
>
> Thank you!
> Alex
>
> On Thu, Jun 11, 2015 at 5:13 PM, Timo Teras <timo.te...@iki.fi> wrote:
>
> > On Wed, 10 Jun 2015 09:18:47 +0300
> > Alex Levit <alexl.glo...@gmail.com> wrote:
> >
> > > Hello,
> > >
> > > I have working HUB & Spoke setup, with RIP working over it and
> > > all is OK.
> > >
> > > I am trying to enable Phase3 feature and for this reason I am
> > > adding shortcut and redirect directives to opennhrp.conf as below:
> > >
> > > HUB
> > > ====
> > > interface mgre1
> > > holding-time 7200
> > > multicast dynamic
> > > shortcut
> > > redirect
> > >
> > > Spoke
> > > =====
> > > interface mgre1
> > > holding-time 7200
> > > map 4.1.1.6/24 2.1.6.1 register
> > > multicast 4.1.1.6
> > > shortcut
> > > redirect
> > > ======================================
> > >
> > > Multicast traffic from spoke to hub stops running and below
> > > messages are appearing in spoke log:
> > >
> > > ....
> > > Jun 9 21:36:49 SmartSwitch daemon.info opennhrp[7220]: Traffic
> > > Indication from proto src 4.1.1.6; about packet to 224.0.0.9
> > > Jun 9 21:36:49 SmartSwitch daemon.info opennhrp[7220]: Sending
> > > Resolution Request to 224.0.0.9
> > > Jun 9 21:36:49 SmartSwitch daemon.err opennhrp[7220]: No peer
> > > entry for protocol address 224.0.0.9
> > > Jun 9 21:36:49 SmartSwitch daemon.info opennhrp[7220]: Failed to
> > > resolve 224.0.0.9: protocol address unreachable (6)
> > > ....
> > >
> > > Is there anything wrong with my config ?
> > > What is possible reason for that behavior ?
> >
> > The above logs are weird. TI is not sent on multicast traffic. At
> > least should not be. What version of opennhrp you have? Is there
> > any local patches applied?
> >
> > Could you also explain a bit more how the it breaks?
------------------------------------------------------------------------------
_______________________________________________
opennhrp-devel mailing list
opennhrp-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opennhrp-devel
