Skip to content
Docker container for intercepting packets with scapy from a netfilter queue (nfqueue)
Branch: master
Clone or download
Type Name Latest commit message Commit time
Failed to load latest commit information.
Dockerfile initial commit Mar 2, 2017


Docker container with an example python script to listen for packets on a netfilter queue and manipulate them with scapy. You can listen on any queue number, and you can push packets into the queue from any iptables rule. This container gives you a powerful prototyping and debugging tool for monitoring, manipulating, dropping, accepting, requeing, or forwarding network packets in python. You can read from a queue on the host with --net=host --cap-add=NET_ADMIN. Or, you can run it within another container's namespace to listen for packets on an nfqueue in that container's network namespace.

This container includes a full installation of scapy and python netfilter queue (nfqueue) bindings, and an example python script to print incoming packets on the queue.

scapy: python-netfilterqueue:

How to use

Clone this repository

git clone

Build the docker container. This will take a while because it includes the full scapy install and all its dependencies. You can use any tag you want, but as an example here I'm using nfqueuelistener

cd docker-nfqueue-scapy
sudo docker build . -t nfqueuelistener


Use iptables on the host to send TCP packets destined for port 9001 to nfqueue 1:

sudo iptables -t raw \
              -A PREROUTING \
              -p tcp --destination-port 9001 \
              -j NFQUEUE --queue-num 1

Run the docker container to listen for packets and print then accept any received packets.

sudo docker run -it --rm \
                --cap-add=NET_ADMIN \
                --net=host \
                --name=nfqueuelistener nfqueuelistener

From another machine, send some packets to test:

echo "Hello" | nc -v $HOST_IP_ADDRESS 9001

You should see something like this:

miles@box:~/testing$ sudo docker run -it --rm --cap-add=NET_ADMIN --net=host --name=nfqueuelistener nfqueuelistener
Listening on NFQUEUE queue-num 1...
<IP  version=4L ihl=5L tos=0x0 len=64 id=6387 flags=DF frag=0L ttl=55 proto=tcp chksum=0x6850 src= dst= options=[] |<TCP  sport=58164 dport=9001 seq=4038873318 ack=0 dataofs=11L reserved=0L flags=S window=65535 chksum=0x67be urgptr=0 options=[('MSS', 1452), ('NOP', None), ('WScale', 5), ('NOP', None), ('NOP', None), ('Timestamp', (2615879909, 0)), ('SAckOK', ''), ('EOL', None)] |>>

Setting the queue number

The default queue number is 1. You can override this by setting the environment variable QUEUE_NUM when running the container. For example, for queue 2:

sudo docker run -it --rm \
                -e 'QUEUE_NUM=2' \
                --cap-add=NET_ADMIN \
                --net=host \
                --name=nfqueuelistener nfqueuelistener

Editing the file

One way to edit the file is to simply edit it and then rebuild the container with sudo docker build . -t nfqueuelistener. Since you are only editing the python file, building will not take as long as the first build.

You can find the documentation for the nfqueue library used at

Listening in another container's namespace

I have not tested this, but it should work.

Say you have another container $CONTAINER_ID and you want to intercept incoming packets in its namespace. You can run this docker container like:

sudo docker run -it --rm \
                --net=container:$CONTAINER_ID \
                --name=nfqueuelistener nfqueuelistener

Note that you will need to run your iptables rules to send packets to the queue from within the $CONTAINER_ID container.

Other notes

scapy is hardcoded version 2.3.2 because there is a bug in 2.3.3 causing scapy to fail on openstack deployments. The bug is actually upstream in openstack, and has been fixed, but this caused problems for me testing on where they have apparently not updated openstack yet.

You can’t perform that action at this time.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.