Differences

This shows you the differences between two versions of the page.

--- documentation:examples:multicast_with_pim-sm [2019/11/08 10:55] – [4. Check VM3 correctly notice this mcast subscriber] olivier
+++ documentation:examples:multicast_with_pim-sm [2019/11/08 18:55] – [Checking NIC drivers and Bhyve compatibility with multicast] olivier
@@ Line 130: / Line 130: @@
 </code>
+===== Checking NIC drivers and Bhyve compatibility with multicast =====
+Before to star with advanced routing setup, just start to test simple multicast between 2 relatives host: Some NIC (vtnet) or some hypervisors network setup aren't compliant with very simple multicast.
+On VM1, start a mcast generator (client emitting mcast):
+<code>
+[root@VM1]~# iperf -c 239.1.1.1 -u -T 32 -t 3000 -i 1
+------------------------------------------------------------
+Client connecting to 239.1.1.1, UDP port 5001
+Sending 1470 byte datagrams, IPG target: 11215.21 us (kalman adjust)
+Setting multicast TTL to 32
+UDP buffer size: 9.00 KByte (default)
+------------------------------------------------------------
+[  3] local 10.0.12.1 port 46636 connected with 239.1.1.1 port 5001
+[ ID] Interval       Transfer     Bandwidth
+[  3]  0.0- 1.0 sec   131 KBytes  1.07 Mbits/sec
+[  3]  1.0- 2.0 sec   128 KBytes  1.05 Mbits/sec
+[  3]  2.0- 3.0 sec   128 KBytes  1.05 Mbits/sec
+[  3]  0.0- 3.5 sec   446 KBytes  1.05 Mbits/sec
+[  3] Sent 311 datagrams
+(...)
+</code>
+On the direct connected VM2, start to check if in non-promiscious mode it sees mcast packets comming:
+<code>
+[root@VM2]~# tcpdump -pni em0 -c 2
+tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
+listening on em0, link-type EN10MB (Ethernet), capture size 262144 bytes
+:22:32.517270 IP 10.0.12.1.33482 > 239.1.1.1.5001: UDP, length 1470
+:22:32.528668 IP 10.0.12.1.33482 > 239.1.1.1.5001: UDP, length 1470
+packets captured
+packets received by filter
+packets dropped by kernel
+</code>
+=> VM2 is receiving mcast packets from 10.0.12.1 to mcast group 239.1.1.1.
+Now on VM2 start a mcast listener (server receiving), it should receive multicast flow
+<code>
+[root@VM2]~# iperf -s -u -B 239.1.1.1%em0 -i 1
+------------------------------------------------------------
+Server listening on UDP port 5001
+Binding to local address 239.1.1.1
+Joining multicast group  239.1.1.1
+Receiving 1470 byte datagrams
+UDP buffer size: 41.1 KByte (default)
+------------------------------------------------------------
+[  3] local 239.1.1.1 port 5001 connected with 192.168.100.149 port 35181
+[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
+[  3]  0.0- 1.0 sec   129 KBytes  1.06 Mbits/sec   0.038 ms  107/  197 (54%)
+[  3]  1.0- 2.0 sec   128 KBytes  1.05 Mbits/sec   0.054 ms    0/   89 (0%)
+[  3]  2.0- 3.0 sec   128 KBytes  1.05 Mbits/sec   0.021 ms    0/   89 (0%)
+[  3]  3.0- 4.0 sec   128 KBytes  1.05 Mbits/sec   0.025 ms    0/   89 (0%)
+[  3]  4.0- 5.0 sec   128 KBytes  1.05 Mbits/sec   0.024 ms    0/   89 (0%)
+[  3]  5.0- 6.0 sec   129 KBytes  1.06 Mbits/sec   0.024 ms    0/   90 (0%)
+[  3]  6.0- 7.0 sec   128 KBytes  1.05 Mbits/sec   0.024 ms    0/   89 (0%)
+(...)
+</code>
+=> Notice the mcast receiver is correctly receiving at 1Mb/s.
+Here is a non working example (here because source interface not given):
+<code>
+[root@VM2]~# iperf -s -u -B 239.1.1.1 -i 1
+------------------------------------------------------------
+Server listening on UDP port 5001
+Binding to local address 239.1.1.1
+Joining multicast group  239.1.1.1
+Receiving 1470 byte datagrams
+UDP buffer size: 41.1 KByte (default)
+------------------------------------------------------------
+(...)
+</code>
+=> Here it doesn't receive traffic and stay in "waiting" mode forever.
 ===== Checking pimd behavior =====
@@ Line 460: / Line 534: @@
 </code>
-VM3 correctly learn that there is a subscriber to group 239.1.1.1 on interface vif1 (toward VM4).
+VM3 correctly learn that there is a subscriber to group 239.1.1.1 on interface vif1 (toward VM4) and the source is on vif0 (toward VM2).