we have an embedded Linux device that primarily uses the Ethernet to communicate. We also allow access via the USB port, to support this our device has a g_ether Linux gadget driver that creates a virtual ethernet port for both our device and the clients host. A DHCP server assigns the client an IP address saving them having to configure this them selves, while the device has a fix IP address. I believe this is much the same as how Android does its tethering
This all works fine until a client wants to use 2 or more devices at once. Both devices come up as distinct network interfaces on the clients machine, but both devices have the same IP address meaning you cannot target the host specifically! A smaller problem is that there is a slight possibility that both devices might assign the clients "distinct" interface the same ip address and cause a conflict there too, but I am less concerned about that as the IP address is random since the DHCP server assigns the address based on the MAC address and g_ether randomly generates this, so a retry should fix the issue (not ideal, but will do)
I would like to know if anyone else has faced a similar problem, and if so how they solved it ?
Related
Two Linux machines (embedded board and PC) are connected to each other via USB cable, and communication is established via USB gadget/Ethernet. Both sides have IP address, ping works, etc.
I need to find a way to detect USB cable disconnect on the board side, ideally with a single Bash command.
ifconfig usb0 still shows IP address and that device is "UP" and "RUNNING" even after unplugging the cable.
In my particular case it turned out to be HW+SW issue: driver (atmel_usba_udc) is using USB power (VBUS) to detect connection, since it's acting as a slave and power is provided by host. However in PCB power is provided unconditionally by the board itself for all USB ports at once.
Normally any standard method should work: ifconfig or sysfs.
One way to do this is to check the file:
/sys/class/net/usb0/operstate
If the file content is "up" then the link exists. If the content is "down", then there is no link.
I have an Android tablet and A Raspberry Pi and I want to established a connection between them automatically when the tablet sends a request to the Pi.
I followed an Android application example here and start discovering any nearby devices. (https://www.youtube.com/watch?v=qnY97iBxp30)
At the same time i run sudo wpa_cli and p2p_find 20. The Android application detects the Pi, and I try to establish connection with the Pi which will display
<3>P2P-GO-NEG-REQUEST TABLET_MAC_ADDRESS dev_passwd_id=4
Normally I would just p2p_connect TABLET_MAC_ADDRESS pbc to successfully connect them together but I find it inefficient if I were to swap to another mobile device.
Are there any other ways to connect the tablet without writing the tablet mac address? For example connecting to that specific device ssid when they send a P2P-GO-NEG-REQUEST to the pi?
TL;DR Nope.
If we look at the OSI ISO 7 layer model for network communication we can see that the Media Access Control (MAC) address is vital for identifying which device is which within a wifi network.
You could try setting up a bluetooth connection or a token-ring, but I suspect that would be more effort than you are looking for.
With IPv6 your devices could use neighbour discovery to automate past the MAC entry to the Internet Protocol, and its possible to connect between devices using their link-local address (fe80::some:thing)
Wifi carries packets of data, that have addresses. By analogy, if I tell you which town I live in, but don't write my building address on the packet, you are going to have a hard time delivering it.
We have developed a device that is connected to our clients network and we would like to be able to get its ip or set the ip without knowing its ip.
The device has a Linux OS.
We can save the device MAC Address before giving it to customers. We can program a service to broadcast the device IP and MAC Address to a certain IP or port/socket. We can listen to a certain port/socket for commands. Is this the right direction? Should we investigate in other network protocols other than TCP/IP?
We have seen this feature in hardware/device manufacturers provide a CD with a software that can locate their devices on a network even if they have been newly added to the network without network or ip configuration.
Best regards,
Hussam Kazah
Using propriety broadcasting protocol is a very common technique for detecting devices on network without knowing it's name.
However there's a better option:
UPNP, is an excellent protocol for achieving your goals.
libupnp can get you started in no time.
There's a standard protocol called DHCP which allows a network device to make a broadcast request for its IP address. This protocol is widely used by network appliances. On the other hand you may scan your local network for all connected devices using ARP (address resolution protocol) using for example arp-scan utility.
I have a server running Debian 7. The eth0 interface is configured to use the on-board ethernet card. This is basically used to connect to the internet. As it happens, I had to connect this server to some PCs through a switch, obviously on a different series of IPs. for this, I installed an external NIC in the PCI slot but, strangely it didn't seen to work. The configurations were alright. I checked them more times than I can imagine. So, I disabled my eth0 interface and connected eth1 (external NIC) to the internet. If for the same settings, the on-board card worked, so should the external one. But, it didn't. When I tried to ping some servers like 8.8.8.8, it gives me Destination Host Unreachable and on termination shows, 0 packets "transmitted" and 0 packets received, which is baffling, to say the least. The PCI slot is working because I checked if the drivers were being recognised or not. The NIC itself is working (checked with another machine running Debian 6). Any help/sugesstions would be appreciated.
P.S The NIC in question is D-Link System DGE-530T Gigabit Ethernet Adapter (rev 11)
You need to check to see if the card is being listed in lspci or not. Second, is this a virtual machine?
I would also check to see if the BIOS is handling IRQ's in auto or are they specifically assigned.
I'm using node-serialport to work with USB devices, however on Windows I can't find out how can I connect to specific device. When I use list function there is only one item with COM3 address, however that is present always, even when no other devices are connected. How can I find out what is the USB device address that can be used for setting up communication?