tcpdump Mailing List
Covers the classic tcpdump text-based network sniffer and its libpcap sniffer library component.
List Archives
- Jan–Mar
- Apr–Jun
- Jul–Sep
- Oct–Dec
- 2026
- 1
- –
- –
- –
- 2025
- 10
- 42
- 25
- 51
- 2024
- 19
- 34
- 22
- 36
- 2023
- 77
- 8
- 20
- 16
- 2022
- 21
- 47
- 37
- 17
- 2021
- 90
- 5
- 32
- 16
- 2020
- 57
- 69
- 72
- 61
- 2019
- 46
- 47
- 37
- 45
- 2018
- 41
- 58
- 149
- 32
- 2017
- 57
- 54
- 24
- 59
- 2016
- 33
- 73
- 63
- 48
- 2015
- 112
- 79
- 62
- 108
- 2014
- 133
- 84
- 69
- 112
- 2013
- 127
- 157
- 55
- 107
- 2012
- 176
- 84
- 53
- 144
- 2011
- 177
- 234
- 187
- 215
- 2010
- 217
- 131
- 85
- 141
- 2009
- 220
- 182
- 186
- 145
- 2008
- 233
- 140
- 139
- 269
- 2007
- 154
- 118
- 251
- 226
- 2006
- 200
- 147
- 71
- 162
- 2004
- 392
- 374
- 377
- 208
- 2003
- 315
- 283
- 259
- 304
- 2002
- –
- –
- –
- 319
Latest Posts
activities report for December 2025
Denis Ovsienko (Jan 01)
December 2025
=============
The accounted activities in December stand for 120:10 working hours and
40 commits (5 in tcpdump, 20 in libpcap and 15 in tcpdump-htdocs).
There are 310 new tests in libpcap.
Most notably, tcpdump 4.99.6 and libpcap 1.10.6 have been released.
This libpcap release fixes two minor vulnerabilities (CVE-2025-11961
and CVE-2025-11964), a more detailed account of other improvements is
available in the change logs. In...
Re: capture and inject device capabilities in libpcap
Guy Harris (Dec 19)
My inclination is to have libpcap supply devices regardless of whether they have no-capture or no-inject set, and have
the caller choose what to show.
That way, it can be changed at the application level if the existing application behavior is an issue. For example, a
sniffer could start out not showing no-capture devices and, if people ask why the XXX interface isn't showing up, and
it turns out that it's a device that doesn't...
Re: Improving DPDK support
Stephen Hemminger via tcpdump-workers (Dec 12)
Re: Improving DPDK support
Denis Ovsienko (Dec 12)
Yes. How much of the existing code to preserve, if any, would be up to
anyone who can get pcap-dpdk.c into shape.
One of the directions of work done in libpcap in recent few years has
been removing of modules that could/should not be maintained anymore
(e.g. AirPcap, Enetfilter, NIT, Septel, STREAMS NIT, TurboCap and
[Tru64/Ultrix] Packet Filter). Another was fixing of various bit rot
in the remaining modules where practicable (DAG, DLPI,...
Re: Improving DPDK support
Guy Harris (Dec 12)
As this is talking about libpcap, presumably you mean "which would have a program using libpcap, such as tcpdump or
Wireshark, running as a seondary process"; they might be the *primary* users of libpcap, but they're not the *only*
users (Snort and scapy both use it).
Presumably this would still allow it to be used as a sniffer, as that's what tcpdump and Wireshark are, and as that's
the primary purpose of libpcap (to...
Improving DPDK support
Stephen Hemminger via tcpdump-workers (Dec 11)
Re: BPF ISA web page
Vadim Goncharov (Dec 06)
It should contain not only "mnemonics" (BTW, A and X registers are uppercase)
but also also layout of instruction and actual codes such as BPF_K or BPF_IND.
This will allow reader to understand potential of extensions. When I tried to
design next step of BPF ISA (alternative to eBPF and comaptible to classic
BPF), I've even drawed them like RFC-style diagrams:
https://github.com/nuclight/bpf64/blob/main/bpf64spec.md
Re: dlt_choice table in pcap.c
Guy Harris (Dec 04)
The full fun story is at https://gitlab.com/wireshark/wireshark/-/issues/2010
Re: dlt_choice table in pcap.c
Guy Harris (Dec 04)
I think the capture code in tcpdump, which I think later became the separate libpcap library, originally just supported
the BPF capture mechanism, which used DLT_ values to indicate link-layer types. Thus, libpcap used DLT_ values in its
APIs; there *were* no LINKTYPE_ values.
Unfortunately, when they added new link-layer types to BPF, various OSes that picked up BPF sometimes chose values such
that the same numerical value corresponded to...
dlt_choice table in pcap.c
Michael Richardson (Dec 04)
Guy, we have this lovely table in pcap.c:
static struct dlt_choice dlt_choices[] = {
DLT_CHOICE(NULL, "BSD loopback"),
DLT_CHOICE(EN10MB, "Ethernet"),
DLT_CHOICE(EN3MB, "experimental Ethernet (3Mb/s)"),
DLT_CHOICE(AX25, "AX.25 layer 2"),
DLT_CHOICE(PRONET, "Proteon ProNET Token Ring"),
...
I feel like it ought to be indexed by LINKTYPE instead....
Re: serving git:// from git.tcpdump.org
Michael Richardson (Dec 03)
Denis Ovsienko <denis () ovsienko info> wrote:
> git.tcpdump.org has been available for git cloning as both https:// and
> git:// for a while. I consider introducing Debian 13 to provide this
> service, but this version of this Linux distribution made it more
> difficult to serve git:// because "The git-daemon-run and
> git-daemon-sysvinit packages have been removed from trixie due to
>...
serving git:// from git.tcpdump.org
Denis Ovsienko (Dec 03)
Hello all.
git.tcpdump.org has been available for git cloning as both https:// and
git:// for a while. I consider introducing Debian 13 to provide this
service, but this version of this Linux distribution made it more
difficult to serve git:// because "The git-daemon-run and
git-daemon-sysvinit packages have been removed from trixie due to
security reasons." [1] I do not know which security reasons these are
besides the obvious lack...
activities report for November 2025
Denis Ovsienko (Dec 02)
November 2025
=============
The accounted activities in November stand for 164:10 hours and 62
commits (3 in tcpdump, 44 in libpcap and 15 in tcpdump-htdocs). There
are 2171 new tests in libpcap.
In libpcap the main improvements are as follows:
* "(host|net) <IPv4 address>", "(arp|rarp) host <IPv4 address>",
"(ip|ip6) multicast", "ip broadcast" and the index operation now
compile to...
Re: Accurate ECN support in tcpdump/libpcap
Scheffenegger, Richard via tcpdump-workers (Nov 24)
Re: linktype files should repeat the DLT value
Michael Richardson (Nov 22)
I brain-farted and typed DLT in the email when I should have typed LINKTYPE!
Guy Harris <gharris () sonic net> wrote:
>> I don't think that these files were generated, but I wanted to be sure before
>> I hand-edit them to include a number.
> They're not generated, except by copy-and-paste-and modify.
thanks for confirming.
>> I think that it should be an h3 header after the h2 title....
More Lists
Dozens of other network security lists are archived at SecLists.Org.
