This article describes how to set up a Wi-Fi Access Point, sometimes mentioned as AP mode, using open-source software Hostapd in Embedded Linux.
If this is not the information you are looking for, please, expand the menu below for more WiFi related information.
First of all, make sure that the Wi-Fi adapter supports Access Point mode. You can check which Toradex modules have built-in Wi-Fi in the Wifi\BT enabled Toradex SoMs article and for other modules in the WiFi\BT adapters article.
This setup was tested using the following hardware and software:
Instructions may vary when using other Wi-Fi adapters.
Steps are provided for configuring AP mode.
Check current mode and available interfaces, you should see a single interface in AP mode, like this:
$ iw dev phy#0 Interface <IF_NAME> ifindex 5 wdev 0x2 addr d0:c5:d3:33:cd:31 type **AP** txpower 0.00 dBm
Considering the test environment from Prerequisites section, the name of the interface is uap0.
Set the Access Point interface IP and SSID in Hostapd configuration file. Change the following variables in
interface=<IF_NAME> # Access Point interface name ssid=access-point # SSID hw_mode=g # default channel=1 # default own_ip_addr=192.168.8.1 # IP address of the Access Point interface #wpa=1 # no wpa
To provide the Access Point interface with a static IP address and DHCP, it is necessary to create the file
[Match] Name=<IF_NAME> [Network] Address=192.168.8.1/24 DHCPServer=yes
The Hostapd service starts concurrently with the network manager and it fails. To work around this, edit the property After from file
# After=network.target After=multi-user.target
Reload the systemd configuration:
systemctl --system daemon-reload
At last, activate the following services:
$ connmanctl enable wifi $ systemctl enable hostapd $ systemctl start hostapd
Check the Access Point interface:
$ ip a s dev <IF_NAME>
Now you can connect to this Access Point from other Wi-Fi devices. Access Point will automatically start after reboot.
Once AP Mode is configured, it's a common scenario to share an internet connection from another interface. This section relies on the Internet Sharing article from the Arch Wiki.
Create a file to add the packet forwarding rules:
net.ipv4.ip_forward=1 net.ipv6.conf.default.forwarding=1 net.ipv6.conf.all.forwarding=1
The changes will take effect after a reboot.
NAT relies on kernel configuration and userspace tools. This section is thus split into two.
For the example from this article, we have to make sure that the following kernel parameters are enabled. You have to evaluate if additional parameters are required for your use case:
You will find those configs in Networking support > Networking options > Network packet filtering framework (Netfilter). See an illustration (you may click it to see in higher resolution):
It may be possible to compile those configs as modules - check the kernel documentation. For instructions on how to build the kernel or modules, read the article Build U-Boot and Linux Kernel from Source Code.
After the new kernel is deployed, you can verify if it has the correct configs:
zcat /proc/config.gz | grep IP_NF_IPTABLES zcat /proc/config.gz | grep NETFILTER_XT_MATCH_CONNTRACK
You might have to enable some kernel modules, depending on your configuration.
You can use
iptables to enable NAT and a systemd service to make it start on reboot. First, enable the rules for the current session:
iptables -t nat -A POSTROUTING -o eth0 -j MASQUERADE iptables -A FORWARD -m conntrack --ctstate RELATED,ESTABLISHED -j ACCEPT iptables -A FORWARD -i uap0 -o eth0 -j ACCEPT
Save the current config to
mkdir /etc/iptables/ iptables-save > /etc/iptables/iptables.rules
Create a systemd service file
iptables.service. The example is modified from the iptables package from Arch Linux:
[Unit] Description=IPv4 Packet Filtering Framework Before=network-pre.target Wants=network-pre.target [Service] Type=oneshot ExecStart=/usr/sbin/iptables-restore /etc/iptables/iptables.rules ExecReload=/usr/sbin/iptables-restore /etc/iptables/iptables.rules RemainAfterExit=yes [Install] WantedBy=multi-user.target
Notice that in the example above we don't have the rule to stop the service. Carefully evaluate if you need it and the best method for stopping the service, where a good starting point is the iptables package aforementioned.
Reload the systemd services and enable:
systemctl --system daemon-reload systemctl enable iptables
Some utils can enable both DHCP and DNS servers. You might want to use dhcpd or dnsmasq.