ADBlock is blocking some content on the site

ADBlock errore
  •   curl
  •  | 
  •  | 
  •   ( words)


How do I make curl ignore the proxy? Setting $NO_PROXY doesn't seem to work for me.


I assume curl is reading the proxy address from the environment variable http_proxy and that the variable should keep its value. Then in a shell like bash, export http_proxy=''; before a command (or in a shell script) would temporarily change its value.

(See curl's manual for all the variables it looks at, under the ENVIRONMENT heading.)


If your curl is at least version 7.19.4, you could just use the --noproxy flag.

curl --noproxy "*" http://www.stackoverflow.com

From the manual.

I ran into the same problem because I set the http_proxy and https_proxy environment variables. But occasionally, I connect to a different network and need to bypass the proxy temporarily. The easiest way to do this (without changing the environment variables) is:

curl --noproxy '*' stackoverflow.com

From the manual: "The only wildcard is a single * character, which matches all hosts, and effectively disables the proxy."

The * character is quoted so that it is not erroneously expanded by the shell.

Long shot but try setting the proxy to "" (empty string) that should override any proxy settings according to the man page.

First, I listed the current proxy setting with

env | sort | less

(should be something like http_proxy=http://wpad.local.machine.location:port number)

Then I tried setting

export http_proxy=";" 

which gave this error message:

curl: (5) Couldn't resolve proxy ';'


export http_proxy="" && curl http://servername:portnumber/destinationpath/ -d 55

and it worked!

PS! Remember to set http-proxy back to its original settings with

export http_proxy=http://wpad.local.machine.location:port number

I have http_proxy and https_proxy are defined. I don't want to unset and set again those environments but --noproxy '*' works perfectly for me.

curl --noproxy '*' -XGET
  "status" : 200,
  "name" : "Medusa",
  "cluster_name" : "elasticsearch",
  "version" : {
    "number" : "1.5.0",
    "build_hash" : "544816042d40151d3ce4ba4f95399d7860dc2e92",
    "build_timestamp" : "2015-03-23T14:30:58Z",
    "build_snapshot" : false,
    "lucene_version" : "4.10.4"
  "tagline" : "You Know, for Search"

This works just fine, set the proxy string to ""

curl -x "" http://www.stackoverflow.com

You should use $no_proxy env variable (lower-case). Please consult https://wiki.archlinux.org/index.php/proxy_settings for examples.

Also, there was a bug at curl long time ago http://sourceforge.net/p/curl/bugs/185/ , maybe you are using an ancient curl version that includes this bug.

Add your proxy preferences into .curlrc

proxy =
noproxy = .dev,localhost,

This make all dev domains and local machine request ignore the proxy.

Lame answer but: Remember to make sure no proxy is set in a ~/.curlrc file (...).

In my case (macos, curl 7.54.0), I have below proxy set with ~/.bash_profile

$ env |grep -i proxy |cut -d = -f1|sort

With unknown reason, this version of curl can't work with environment variables NO_PRXY and no_proxy properly, then I unset the proxy environment variables one by one, until to both HTTPS_PROXY and https_proxy.

unset https_proxy

it starts working and can connect to internal urls

So I would recommend to unset all proxy variables if you have in your environment as temporary solution.

unset http_proxy https_proxy HTTP_PROXY HTTPS_PROXY

My curl was not ignoring the proxy on Ubuntu 12.04 until I set the "no_proxy" (lowercase) environment variable. The --noproxy option was not available.

Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow