1 netsniff-ng's known issues:
2 ///////////////////////////
4 Q: When I perform a traffic capture on the Ethernet interface, the PCAP file is
5 created and packets are received but without 802.1Q header. If I use
6 tshark, I get all headers but netsniff-ng removes 802.1Q headers. Is that
8 A: Yes and no. The way how VLAN headers are handled in PF_PACKET sockets by the
9 kernel is somewhat problematic [1]. The problem in the Linux kernel is that
10 some drivers already handle VLAN, others not. Those who handle it have
11 different implementations, i.e. hardware acceleration and so on. So in some
12 cases the VLAN tag is even stripped before entering the protocol stack, in
13 some cases probably not. Bottom line is that the netdev hackers introduced
14 a "hack" in PF_PACKET so that a VLAN ID is visible in some helper data
15 structure that is accessible from the RX_RING. And then it gets really messy
16 in the user space to artificially put the VLAN header back into the right
17 place. Not mentioning about the resulting performance implications on that
18 of /all/ libpcap tools since parts of the packet need to be copied for
19 reassembly. A user reported the following, just to demonstrate this mess:
20 Some tests were made with two machines, and it seems that results depends on
24 * ethtool -k eth0 gives "rx-vlan-offload: on"
25 -> wireshark gets the vlan header
26 -> netsniff-ng doesn't get the vlan header
28 * ethtool -K eth0 rxvlan off
29 -> wireshark gets twice the same vlan header (like QinQ even though
31 -> netsniff-ng gets the vlan header
34 * ethtool -k eth0 gives "rx-vlan-offload: on"
35 -> wireshark gets the vlan header
36 -> netsniff-ng doesn't get the vlan header
38 * ethtool -K eth0 rxvlan off
39 -> wireshark gets the vlan header
40 -> netsniff-ng doesn't get the vlan header
42 Even if we would agree on doing the same workaround as libpcap, we still
43 will not be able to see QinQ, for instance, due to the fact that only /one/
44 VLAN tag is stored in this kernel helper data structure. We think that
45 there should be a good consensus on the kernel space side about what gets
46 transferred to the userland.
48 [1] http://lkml.indiana.edu/hypermail/linux/kernel/0710.3/3816.html
50 Q: When I start trafgen, my kernel crashes! What is happening?
51 A: We have fixed this ``bug'' in the Linux kernel under commit
52 7f5c3e3a80e6654cf48dfba7cf94f88c6b505467 (http://bit.ly/PcH5Nd). Either
53 update your kernel to the latest version, e.g. clone and build it from
54 git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git or don't
55 start multiple trafgen instances at once resp. start trafgen with flag -A
56 to disable temporary socket memory tuning! Although trafgen's mechanism is
57 written in a correct manner, some probably Linux internal side-effects
58 cause the tigger of the BUG macro. Why tuning? In general, if not otherwise
59 specified, the netsniff-ng suite tries to get a good performance on default.
60 For instance, this includes things like tuning the system's socket memory,
61 enabling the BPF JIT compiler, migrating the NIC's interrupt affinity and
62 so on. If you don't want netsniff-ng to do this, look at the relevant cmd
63 line options that disable them with ``--help'' and explicitly specify them