I have multiple devices with the same hardware mac address (in factory reset mode) that all call a tftp server in order to get a proper mac.
On the same nic, there would obviously be no way to differentiate the devices, however, i have one device per nic. I evaluated these options to use that fact:
SO_BINDTODEVICE
Binding the tftp server on each nic separately would probably still not give me a way to respond with a package to the sender, since linux' arp table would be confused about which nic to send that package out from.
net namespaces
Linux does not allow bringing physical interfaces into a net namespace. only veths, which again would need to be bridged on the 'host' side, which will probably not work with the same mac appearing on multiple physical nics.
Am i missing an option here?
Related
I'm searching for help and an opinion-advice for a network project, in which I'm working lately. This requires a Linux machine to be a passive network appliance.
Network packets come in from one network interface and come out from another interface ( net--eth0-->Linux PC--eth1-->net) without making any modifications on data.
The application, which is going to run on the Linux system, will change only the order of the packets. It is going to be a "silly" network emulator application.
The first implementation was made with RAW sockets, where read() is called every time a packet arrives to user space and write() is called when an Ethernet packet should be sent down to the NIC.
I would like to know if there is a more practical and direct way than RAW sockets, bypassing Linux's network stack.
If what you want is to bypass the kernel, DPDK in Linux and NetMap in FreeBSD are options to do just that.
Indeed this can be done in dpdk in Linux. There are l3fw and l2fwd sample applications in the examples folder of the dpdk tree, which may inspire you. Also consider using vpp, a fd.io project hosted by Linux Foundation, which can use dpdk.
Rami Rosen
Business case:
This is year 2015, and I am using Linux kernel 3.17. My ARM Linux has two NIC, eth0 and eth1. eth0 is for configuration purpose locally for a person next to it, and eth1 is for configuration purpose from remote. So they are basically same function, except eth0 has a person to plug in the cable in the field.
Local PC ------ [eth0 My ARM Linux Computer eth1] ------ Remote PC
Using eth0, the PC software can configure eth1 IP address and port for configuration. To do that, my eth0 has a preset IP address (192.168.1.2) so the PC software can easily find it. The ARM Linux (3.17) application needs to listen on eth0 port A, and eth1 Port B (if configured), and once connected, will respond, but never initiate any activity (not even ping).
There is no connection between local PC and remote PC. They are not supposed to be bridged, or networked.
Question 1:
If eth1 is configured with same IP as eth0, will my application still able to listen on both NIC ports, and correctly respond? I will have two independent threads (maybe same function, but different parameters) to listen on the two NIC ports.
This is why I ask whether I can specify the NIC when connect, and respond, because now you can see they could have same IP address. Ideally, the OS knows where is the connect request is from, so it should know which way to reply to, even if the two NIC have same IP as long as they have different Ethernet address. But, does C++11 or Berkeley model allow me to specify eth0 or eth1 when I listen and reply?
Now, the local PC and remote PC could have same IP address as well (but different from the ARM Linux). But imagine I have two ARM Linux computers, then this configuration is completely ok. But unfortunately I only have one embedded system and one OS with eth0 and eth1.
Question 2:
If answer to question 1 is no, then can they have same subnet mask? Also, can the local PC and remote PC have same IP address?
Assumption
Setting routing table, is a solution, if works. I wanted to know whether it is feasible and also whether it is recommended not to do it.
Also, this is Linux. Similar questions were asked long time ago for Windows, and Windows XP says (no) and I am using Linux 3.17 and I have same IP address so things are different from similar questions.
At the system level, you can configure a bridge on the mediating box. This can have the same IP address on both sides of the bridge and can be locked down using iptables to only respond to particular TCP/UDP ports.
You can't have systems on both sides of the bridge with the same IP address (i.e. the local pc and remote pc can't have the same IP address). You can use ebtables to limit the access on either side of the bridge to the IP address of the bridge itself, which means that the local pc can only talk to the mediating box and the remote pc can only talk to the mediating box.
This is based on never expecting the local and remote pcs to talk to each other.
It's mostly a sys-adminny solution, though, there's not really an API that you could use as a developer to accomplish the same thing.
I am trying to create a setup for testing network equipment, and would like to use multiple (3) NICs in a single PC to produce traffic through an external device, i.e. a network switch. From one interface to another.
All of the tests are for IPv6 and ethernet. All NICs have link local and global IPv6 addresses with the same network/subnet prefix.
Atm. all packets are routed internally and never use the physical wire.
I see a lot of threads solving the problem for IPv4 using the net.ipv4.conf.all.accept_local parameter together with some routing table fixes. But I can't find a solution for IPv6.
Anyone know of a solution?
I have tried using several commands as well as couple of examples using C/C++ but am still not able to find a flawless method that can differentiate between physical or virtual ethernet adapters. Physical means, on that available on your board or installed externally and virtual means created by virtualization apps such as VirtualBox/VMWare/Virtual PC or VPN etc.
Any pointers?
There is no flawless method. A virtual adapter can have any MAC address, including one that might have been assigned by a constructor to a physical device. And the other way around, given that one can change the MAC address of a physical adapter. You can only make an educated guess.
You might find it easier to detect if you are running virtualized at all, rather than look for specific information about the NICs. The virt-what(1) tool looks through aspects of the running system to guess if the system is virtualized or not. (The script isn't as smart as you think, but it does have a lot of small information gathering tools in one place.)
Someone intentionally trying to bypass a license check would probably not find it difficult to defeat this mechanism.
Maybe one can use mii-tool and check if it fails, which it does for virtual:
mii-tool vmbr2
SIOCGMIIPHY on 'vmbr2' failed: Operation not supported
mii-tool eno1
eno1: negotiated 1000baseT-FD flow-control, link ok
EDIT:
What is mii-tool: view, manipulate media-independent interface status
This utility checks or sets the status of a network interface's
Media Independent Interface (MII) unit. Most fast ethernet
adapters use an MII to autonegotiate link speed and duplex
setting.
https://www.man7.org/linux/man-pages/man8/mii-tool.8.html
I need to store persistent reference to third party device on an arbitrary IP network where the IP address of the devices may be static or randomly assigned by DHCP. I don't control the devices on the network and I can't rely on DNS and other ad-hoc networking protocols existing or working with the devices.
So I have been instructed to investigate using hardware addresses and ARP. This will work but I don't want to duplicate code. The kernel must manage an ARP table. On Windows you can access it using GetIpNetTable etc.
I am hoping there is an API to answer these two questions:
How do I translate from IP to MAC address? (ARP)
How do I translate from MAC to IP address? (InARP)
If not then I may have to do it more manually:
How do I read the kernel's ARP table?
How do I add an entry if I have the determined a mapping myself?
/proc/net/arp
K
ARP tables tend to be fairly local and short-lived. If you examine the protocol, the real MAC addresses are generally only provided when the given IP address is in the local subnet.
Otherwise, the packet is forwarded to the local router, which is then responsible for forwarding it.
If you do "arp -g" on Windows or "arp -a" on UNIX, you'll see the table, but I don't think it will do you any good, due to the reasons mentioned above. That command and
That's really what DNS is for but, as you say, it may not be an option for you.
You may well have to write your own 'ARP' database at your application level.
As for ARP:
You could use system("/usr/bin/arp -option_of_choice"); and parse the output, but that's an ugly hack. -- Not my recommendation.
Take a look at /usr/include/linux/sockios.h -- At the SIOCGARP, SIOCDARP, and SIOCSARP details. Those are ioctls that you can perform to manage the ARP table on linux. Of course, you'll have to perform these ioctls on a socket fd.
Here's some examples: SIOCGARP examples
I'm sure you can find many other examples in several other languages as well. As I'm assuming that you're using C.
As for RARP:
A quote from the linux rarp manpage:
" This program is obsolete. From version 2.3, the Linux kernel no longer
contains RARP support. For a replacement RARP daemon, see ftp://ftp.demen-
tia.org/pub/net-tools"
So you'll have to install rarpd on the target system.