Re: em0, VLAN and bpf(?) trouble w/RELENG_5

From: Mike Tancsa <mike_at_sentex.net>
Date: Mon, 08 Nov 2004 14:24:41 -0500
At 11:16 AM 08/11/2004, Robert Watson wrote:

>(1) If you run tcpdump on the em0 interface itself, does the same thing
>     happen?


I can easily reproduce this.  The first tcpdump works fine.  But when you 
stop dumping, no packets come through.  Start the tcpdump on em0 and then 
stop it works.  Then start it again on em0, the flow stops, even though 
tcpdump says the NIC is forwarding.  Stop the dump and packets flow normal.
5.3-STABLE FreeBSD 5.3-STABLE #0: Fri Nov  5



Using tcpdump

# tcpdump -i vlan0 -n
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on vlan0, link-type EN10MB (Ethernet), capture size 96 bytes
14:09:44.895401 IP 10.2.2.1 > 10.2.2.2: icmp 64: echo request seq 13312
14:09:44.895418 IP 10.2.2.2 > 10.2.2.1: icmp 64: echo reply seq 13312
14:09:45.596227 00:0b:be:69:14:9a > 01:00:0c:cc:cc:cd sap aa ui/C len=39
14:09:45.905417 IP 10.2.2.1 > 10.2.2.2: icmp 64: echo request seq 13568
14:09:45.905427 IP 10.2.2.2 > 10.2.2.1: icmp 64: echo reply seq 13568
14:09:46.915437 IP 10.2.2.1 > 10.2.2.2: icmp 64: echo request seq 13824
14:09:46.915447 IP 10.2.2.2 > 10.2.2.1: icmp 64: echo reply seq 13824
14:09:47.595903 00:0b:be:69:14:9a > 01:00:0c:cc:cc:cd sap aa ui/C len=39
14:09:47.925455 IP 10.2.2.1 > 10.2.2.2: icmp 64: echo request seq 14080
14:09:47.925465 IP 10.2.2.2 > 10.2.2.1: icmp 64: echo reply seq 14080
14:09:48.935475 IP 10.2.2.1 > 10.2.2.2: icmp 64: echo request seq 14336
14:09:48.935485 IP 10.2.2.2 > 10.2.2.1: icmp 64: echo reply seq 14336
14:09:49.595828 00:0b:be:69:14:9a > 01:00:0c:cc:cc:cd sap aa ui/C len=39
14:09:49.945493 IP 10.2.2.1 > 10.2.2.2: icmp 64: echo request seq 14592
14:09:49.945503 IP 10.2.2.2 > 10.2.2.1: icmp 64: echo reply seq 14592
^C
15 packets captured
15 packets received by filter
0 packets dropped by kernel

It seems to be sending it back, but its not.  Soon as I end the tcpdump, it 
starts sending again.  This is a test setup, so I can crash / test whatever 
you like.

         ---Mike 
Received on Mon Nov 08 2004 - 18:17:46 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:21 UTC