77 lines
2.9 KiB

Installing Wireshark on FreeBSD/OpenBSD/NetBSD/DragonFly BSD
1. Berkeley Packet Filter (BPF) requirement
2. Running Wireshark as a non-root user
For general installation instructions, see the INSTALL file, along with
the Developer's Guide located at https://www.wireshark.org/docs/ and
in the docbook/ directory. Additional BSD specific notes and requirements
1. Berkeley Packet Filter (BPF) requirement
In order to capture packets (with Wireshark/TShark, tcpdump, or any
other packet capture program) on a BSD system, your kernel must have the
Berkeley Packet Filter mechanism enabled. The default kernel
configurations in recent versions of BSD systems have this enabled
already. To verify the bpf device is present, look in the /dev
ls -l /dev/bpf*
You should see one or more bpf devices listed similar to this:
crw------- 1 root wheel 0, 90 Aug 10 21:05 /dev/bpf0
crw------- 1 root wheel 0, 91 Aug 10 21:05 /dev/bpf1
Packet-capturing programs will pick the first bpf device that's not in
use. Recent versions of most BSDs will create bpf devices as needed, so
you don't have to configure the number of devices that will be
2. Running wireshark as a non-root user
Since the bpf devices are read-only by the owner (root), you normally
have to run packet capturing programs such as Wireshark as root. It is
safer to run programs as a non-root user if possible. To run Wireshark
as a non-root user, you must change the permissions on the bpf device(s).
If you are the only user that needs to use Wireshark, the easiest way
is to change the owner of each bpf device to your username. You can also
add the read/write ability to the group (typically wheel) and add users
that need to use Wireshark to the wheel group. Check your operating
system's documentation on how to make permanent these changes as they
are often reset upon reboot; if /dev is implemented with devfs, it might
be possible to configure devfs to create all bpf devices owned by a
particular user and/or group and with particular permissions. In
FreeBSD 6.0 and later this can be done by creating an /etc/devfs.rules
file with content such as
add path 'bpf*' {mode and permissions}
where "mode and permissions" can include clauses such as
mode {octal permissions}
to set the permissions on the device (e.g., "mode 0660" to set the
permissions to rw-rw-r--),
user {user}
to set the user who owns the device, or
group {group}
to set the group that owns the device and adding a line such as
to /etc/rc.conf. For example, an /etc/devfs.rules file with
add path 'bpf*' mode 0660 group wheel
will grant read and write permissions on all BPF devices to all users in
the "wheel" group.