tuntap | tuntap library supports both Linux and OS X | Runtime Evironment library

I have an application that creates, listens on and writes to a tap interface. The software will read(tun_fd,...) and perform some action on that data, and it will return data to the system as UDP packets via write(tun_fd,...).

I assign an IP to the interface, 10.10.10.10\24 so that a socket application can bind to it and so that the kernel will pass any packets for the virtual subnet to the tap interface.

The software generate frames with IP/UDP packets with the destination IP being that assigned to the interface, and a source IP existing in the same subnet. The source and dest mac address match that of the tap device. Those frames are written back to the kernel with write(tun_fd,...).

If I open said tap interface in wireshark I will see my frames/packets as I expect to, properly formatted, expected ports, expected macs and IPs. But if I try to read those packets with netcat -lvu 0.0.0.0 ${MY_UDP_PORT} I don't see anything.

Is this expected behavior?

Update 1

INADDR_ANY is a red herring. I have the problem even when explicitly binding to an interface / port as in this pseudo code:

I'm working on a simple project that listens on a tun interface and modified the packets then re-sends them to the real interface.

I have tried songgao/water, pkg/tuntap and even writing my own based on some C code floating around but no matter what I tried, I can't receive TCP packets (ICMP/UDP works fine).

I feel like i'm missing something extremely obvious but I can't figure it for the life of me...

The code:

I have been working on a program which uses a TUNTAP interface (in TUN mode) on a routing device which runs on top of the Linux kernel. It is a multicast tunneling protocol, and I am trying to send MLD joins to the kernel through my application so it can be recieved elsewhere. However, even though I have quadruple-checked my packets being sent on the interface, the linux kernel is dropping the packet before it gets passed on.

Tediously I have been tracing the path of the packet through the linuxkernel trying to figure out why it is being dropped and I think I have figured out to some extent why it is not being processed. The Hop-by-Hop options (containing the Router-Alert option which is necessary for MLD) is being parsed in net/ipv6/ip6_input.c in the ipv6_rcv function, but instead of continuing to process the packet in ip6_rcv_finish, the packet is dropped since the NF_HOOK at the end of the ipv6_rcv function is somehow interpreting the packet as being processed by something else. (NF_STOLEN instead of NF_ACCEPT)

Once the ipv6_rcv function finishes executing, something else executes ip6_mc_input, (in net/ipv6/ip6_input.c still) but from here the Hop-by-Hop options are not processed, which means when the kernel ends up processing the Layer-4 protocol, it has nothing to handle the protocol since the Hop-by-Hop options were meant to be processed beforehand. This means that the kernel drops the packet due to an unknown protocol.

What I am trying to figure out is what is calling ip6_mc_input. I have looked quite a bit on elixir for what could call it, but there are so many possibilities since it is called from a pointer in a rt6_info struct which is difficult to trace since so many things use it. Does anyone know anything that could help me in my search?

IGMP joins work fine, but the IPv4 stuff is probably quite similar so information from that context would probably be helpful too.

For reference, the linux kernel version in use is v4.4.6

I want to create one vm on a private network.

I am following this website http://www.linux-kvm.org/page/Networking to help in configuring my qemu.

I used the following command to create a vm:
sudo qemu-system-x86_64 -hda image.qcow2 -enable-kvm -netdev tap,id=tap1 -device rtl8139,netdev=tap1
and I got:
qemu-system-x86_64: -netdev tap,id=tap1: network script /etc/qemu-ifup failed with status 256

I am using arch. I create a bridge: sudo ip link add br0 type bridge

By default, I didn't have file in /etc/qemu-ifup:
So, I created my own, like recommanded on this website http://www.linux-kvm.org/page/Networking:

I am trying to access HKEY_LOCAL_MACHINE registry,

This is the manifest file:

I have an existing tap device (tap0) that I created on command line.

I am trying to create a network of virtal qemu machines and the host using a bridge on Arch Linux to test a distributed program I wrote. I have found many howtos on using the physical NIC of the host on the bridge to connect to the VMs. This works fine. However, I don't want the VMs to be visible to the outside network but rather create a virtual interface on the host to connect to the VMs. This is what I have tried so far after creating the bridge and starting the VMs on it:

In this link, https://backreference.org/2010/03/26/tuntap-interface-tutorial/, there's a code sample that uses tun/tap interface to create a TCP tunnel as below.

I'm trying to convert a system() call I have in C that calls ip tuntap add ... to C code that doesn't rely on the system() call.

Basically, it's a problem that the tunnel adapter that I use the system("ip tuntap add ...") call to bring up is left there when my application is forcibly terminated.

I've been told I can use rtnetlink to achieve my goal. I'd like to convert the following line into C code that doesn't rely on the system() call. I believe if I use rtnetlink that the tunnel adapter I bring up will be destroyed by the kernel when my application is terminated, which is the main reason I want to do this.

This is my current line in C: