The operations described in this article are applicable to Macos 15.4.1, and the commands described are applicable to zsh (actually, if you use bash, it is not a big problem)
Open System Settings > VPN.
Make sure you have configured the required proxy and note the proxy server address and port (authentication proxy requires username and password)
(To see the server address, port, username, password information, please click ⓘ behind the proxy you want to configure in Terminal in System Settings > VPN
Server address in Proxies
Port in Proxies
User name in L2TP over IPSec
Password in L2TP over IPSec
user name: password, you need to make sure your User authentication is set to Password, but if you are using RSA SecurID, Certificate, Kerberos, CryptoCard, you can also try the method I will describe later, just use RSA SecurID, Certificate, Kerberos, CryptoCard as Password, but I can’t guarantee that it will succeed, because I have never used RSA SecurID, Certificate , Kerberos , CryptoCard.)
Non-authentication proxy:
export http_proxy=http://proxy server address:port
export https_proxy=http://proxy server address:port
export all_proxy=http://proxy server address:port
Authentication proxy:
export http_proxy=http://username:password@proxy server address:port
export https_proxy=http://username:password@proxy server address:port
export all_proxy=http://username:password@proxy server address:port
Input
curl https://www.google.com
If the output is <!doctype html> ... </script> </body></html>% Such a large lump means success.
If successful, you can configure it permanently.
echo "export http_proxy=http://username:password@proxy server address:port" >> ~/.zshrc
echo "export https_proxy=http://username:password@proxy server address:port" >> ~/.zshrc
echo "export all_proxy=http://username:password@proxy server address:port" >> ~/.zshrc
(For non-authenticated proxy, just write as above)
Or use nano, vim to edit .zshrc directly (file address is ~/.zshrc ).
If you write it with echo "..." >> ~/.zshrc, you need to save it with the following command.
(If you use nano, vim, please save it in the same way as nano or vim. Of course, it is also recommended to save it again with the following command, because saving in nano or vim may only save what is edited in nano or vim editor, and it is possible that the changes in .zshrc are not saved).
source ~/.zshrc
It is recommended to restart Terminal after saving with source ~/.zshrc.
Same as the non-permanent configuration test above.
Input:
curl https://www.google.com
If the output is <!doctype html> ... </script> </body></html>% such a big lump, it means success.
If your proxy is SOCKS type, the environment variable needs to use socks5:// or socks4://, for example:
export http_proxy=socks5://proxy server address:port
export https_proxy=socks5://proxy server address:port
If you want certain addresses not to go through the proxy, you can set no_proxy:
export no_proxy="localhost,127.0.0.1,*.local"
I’ll say it again.
If the proxy requires a username and password, the format is:
export http_proxy=http://username:password@proxy server address:port
export https_proxy=http://username:password@proxy server address:port
I’ll say it again.
Use the following command to confirm whether the proxy is effective:
env | grep -i proxy
curl -I https://www.google.com
Why add env | grep -i proxy What is its use?
env | grep -i proxy + curl -I https://www.google.com Function:
env | grep -i proxy ：List all environment variables containing proxy (case insensitive) in the current shell environment, such as http_proxy , https_proxy , all_proxy etc.
Used to confirm whether the proxy environment variables are set correctly and whether their specific values meet expectations (such as whether the address and port are correct).
Sample output (assuming the proxy is http://192.168.1.100:8080)：
http_proxy=http://192.168.1.100:8080
https_proxy=http://192.168.1.100:8080
all_proxy=http://192.168.1.100:8080
curl -I https://www.google.com ：Use curl to send an HTTP HEAD request to https://www.google.com and return only the response headers.
If the proxy is set up correctly, you will see a response similar to HTTP/2 200, indicating that the target website was successfully accessed through the proxy.
If the proxy is not effective or has problems, you may get an error (such as Failed to connect or Connection refused).
The default behavior of curl is to send a GET request And output the response body, that is, the content of the web page or the content returned by the interface.
curl -I sends a HEAD request to get only the response headers, not the response body. This is useful when you only want to view the server response status, type, cache information, etc.
all_proxyIf you encounter some tools (such as git, pip) that cannot recognize all_proxy, you can set http_proxy and https_proxy as a fallback.
If your proxy is SOCKS type or has complex requirements (such as bypassing specific addresses), it is recommended to refer to the previous detailed answer and use proxychains or set no_proxy.