Why are you binding an interface to a multicast address? That is not
allowed in IP so it shouldn't be done. If you want to receive
multicast you just need to open a socket on that address and listen.
(make sure you have multicast support compiled into the kernel).
I don't think the Linux IP stack should allow binding an interface to
a class D (multicast) address.
> eth0:0 Link encap:Ethernet HWaddr 00:20:AF:6A:09:1E
> inet addr:224.168.228.1 Bcast:224.168.228.255 Mask:255.255.255.0
> UP BROADCAST RUNNING MTU:1500 Metric:1
> RX packets:0 errors:0 dropped:0 overruns:0
> TX packets:0 errors:0 dropped:0 overruns:0
...
> First, I run tcpdump on the host to read UDP packets (see below "HOST B, WIND
> OW
> 1"). Second, I run (and strace) netcat listening to the same UDP port (see
> "HOST B, WINDOW 2"). Third, on another host on the same LAN I run netcat
> pointed to the multicast address configured for HOST A's eth0:0, and type in
> a
> few words to STDIN. Each word gets packetized and sent out to the multicast
> address.
>
> Notice that tcpdump (see "HOST B, WINDOW 1") caught each of the packets. But
> netcat did not (see "HOST B, WINDOW 2"). Netcat caught only the *first* pack
> et
> (then I hit Ctrl-C). You can see the strace below in "HOST B, FILE
> "ncstrace"".
>
> Is this a bug in netcat, or in the Linux networking stack?
-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to [EMAIL PROTECTED]
