Re: Regression in vr - not receiveing multicast



--On December 9, 2008 20:47:23 +0900 Pyun YongHyeon <pyunyh@xxxxxxxxx> wrote:

On Tue, Dec 09, 2008 at 10:40:17AM +0100, Goran Lowkrantz wrote:
> Hi,
>
> in July, vr had this problem and was fixed:
> <http://www.FreeBSD.org/cgi/query-pr.cgi?pr=125010>
> <http://www.FreeBSD.org/cgi/query-pr.cgi?pr=125024>
>
> but now it's back again!
>

There was just one bug fix since then and I guess the fix is not
related with your issue.

> On a system with the following:
> 7.1-PRERELEASE FreeBSD 7.1-PRERELEASE #3: Thu Oct 16 12:31:04 UTC 2008
>
> I have to set all vr interfaces in promisc to get routing info.
>
> Using Quagga
> # pkg_info -Ix uagga
> quagga-0.99.10_3 Free RIPv1, RIPv2, OSPFv2, BGP4, IS-IS route
software >
> on an inner network using RIPv2
> # ifmcstat -i vr0
> vr0:
> inet xxx.xxx.xxx.xxx
> group 224.0.0.9
> igmpv2
> mcast-macaddr 01:00:5e:00:00:09 refcnt 1
> group 224.0.0.1
> mcast-macaddr 01:00:5e:00:00:01 refcnt 1
>
>
> On the same box, we have some em devices also and they work without
any > problems.
>

There is fundamental differences between em(4) and vr(4). The
vr(4) for VT6105M takes advantage of perfect multicast filtering
feature which is not present on all em(4) interface. Perfect
multicast filtering can reduce unwanted multicast traffics such
that it could save a lot of CPU cycles. The downside is that vr(4)
cannot accept multicast frames for a multicast group without
joining the multicast group first.
For multicast routing purpose I guess 'options MROUTING' kernel
option should be enabled to accept all multicast frames.
Does your kernel have that option?

No it has not. I will create such a beast and return with stories.


/glz


_______________________________________________
freebsd-stable@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscribe@xxxxxxxxxxx"



Relevant Pages

  • Re: [Linux] Multicast Routing?
    ... > Kann Linux sauber Multicast forwarden und wie wird ... > das igmp gehandelt etc. ... Der richtig kompilierte Kernel und z.B. pimd. ... IP Multicast funktioniert in einer Broadcastdomain out of the box. ...
    (de.comp.os.unix.networking.misc)
  • Re: [Lse-tech] Re: [Patch 7/7] Generic netlink interface (delay accounting)
    ... Assume several apps in user space and a target in the kernel (this ... The response is unicast; the ... effect of what they affected may cause an event which may be multicast. ... Suppose C issued a GET to find details about a specific pid, ...
    (Linux-Kernel)
  • Re: [Lse-tech] Re: [Patch 7/7] Generic netlink interface (delay accounting)
    ... Assume several apps in user space and a target in the kernel (this ... The response is unicast; the ... effect of what they affected may cause an event which may be multicast. ... Suppose C issued a GET to find details about a specific pid, ...
    (Linux-Kernel)
  • Re: IP_MAX_MEMBERSHIPS
    ... This limit in no way affects multicast routing, ... break your kernel for modules compiled without such a change. ... I think you might be hitting the IGMP limit. ... The 244.0.0.5/6 multicast address is joined using IGMP and there is a kernel variable that needs to get bumped up higher to peer with more than 20 neighbors. ...
    (freebsd-stable)
  • Multicast problems between 2.4.20 and 2.4.21?
    ... Was there ever a resolution to the multicast issues with kernels beyond ... up in the kernel mailing lists so far (i.e. ... This substitution gets the machines multicasting but I've had ... In particular 224.0.0.x gets no IGMP ...
    (Linux-Kernel)