ShadowSocks got picked up by DPI (they have FortiGuard). I have added the v2ray plugin to my server, and routed it through CloudFlare.
But now I need to add the v2ray plugin to OpenWRT/gl-ss. Should I ditch the gl-ss and instead install manually the shadowsocks-libdev and v2ray plugin? I’d love to be able to keep the gl-ss installation and GUI front end, and only add the executable v2ray plugin to the config.
Copy the file to /tmp (this is FLASH memory), as there is not enough memory available on the main storage to install otherwise.
Run the command opkg install v2ray-plugin_*_ mips_24kc .ipk
The binary file is copied to /usr/bin
When you go to Tools | Advanced of OpenWRT will take you to the generic LuCI screens, where settings for Shadowsocks are more complete. You can enter plugin and plugin-opts here. When you enter the details, and start the SOCKS5 proxy, it loads the program like this;
ss-local –c -l 1080 -f /var/run/ss-local..pid
LuCI stores its configuration in /etc/config/shadowsocks.
The generated configuration is performed during the service start up. You can see how that is performed by looking in the /etc/init.d/shadowsocks file.
You can run ss-local as a daemon from the command prompt;
ss-local is a normal SOCKS5 proxy. You need to configure clients to use it.
ss-redir is a transparent proxy. That means you don’t need to configure the clients to use it. As soon as traffic arrives at the router, it transports it via the proxy (transparently)
On Windows 10 Chrome; the ‘System Proxy’ must not be SOCKS5 (possibly SOCKS4?). To load Chrome with SOCKS5, start it like this;