[javascript] Refused to load the script because it violates the following Content Security Policy directive

When I tried to deploy my app onto devices with Android system above 5.0.0 (Lollipop), I kept getting these kind of error messages:

07-03 18:39:21.621: D/SystemWebChromeClient(9132): file:///android_asset/www/index.html: Line 0 : Refused to load the script 'http://xxxxx' because it violates the following Content Security Policy directive: "script-src 'self' 'unsafe-eval' 'unsafe-inline'". 07-03 18:39:21.621: I/chromium(9132): [INFO:CONSOLE(0)] "Refused to load the script 'http://xxx' because it violates the following Content Security Policy directive: "script-src 'self' 'unsafe-eval' 'unsafe-inline'".

However, if I deployed it to mobile device with Android system of 4.4.x (KitKat), the security policy works with the default ones:

<meta http-equiv="Content-Security-Policy" content="default-src 'self' data: gap: https://ssl.gstatic.com 'unsafe-inline' 'unsafe-eval'; style-src 'self' 'unsafe-inline'; media-src *">

Then I thought, maybe, I should change to something like this:

<meta http-equiv="Content-Security-Policy" content="script-src 'self' 'unsafe-eval' 'unsafe-inline'; object-src 'self'; style-src 'self' 'unsafe-inline'; media-src *">

Basically, both options don't work for for me. How can I solve this issue?

The probable reason why you get this error is likely because you've added the /build folder to your .gitignore file or generally haven't checked it into Git.

So when you Git push Heroku master, the build folder you're referencing don't get pushed to Heroku. And that's why it shows this error.

That's the reason it works properly locally, but not when you deployed to Heroku.

We used this:

<meta http-equiv="Content-Security-Policy" content="default-src gap://ready file://* *; style-src 'self' http://* https://* 'unsafe-inline'; script-src 'self' http://* https://* 'unsafe-inline' 'unsafe-eval'">

Try replacing your meta tag with this below:

<meta http-equiv="Content-Security-Policy" content="default-src *; style-src 'self' http://* 'unsafe-inline'; script-src 'self' http://* 'unsafe-inline' 'unsafe-eval'" />

Or in addition to what you have, you should add http://* to both style-src and script-src as seen above added after 'self'.

If your server is including the Content-Security-Policy header, the header will override the meta.

Full permission string

The previous answers did not fix my issue, because they don't include blob: data: gap: keywords at the same time; so here is a string that does:

<meta http-equiv="Content-Security-Policy" content="default-src * self blob: data: gap:; style-src * self 'unsafe-inline' blob: data: gap:; script-src * 'self' 'unsafe-eval' 'unsafe-inline' blob: data: gap:; object-src * 'self' blob: data: gap:; img-src * self 'unsafe-inline' blob: data: gap:; connect-src self * 'unsafe-inline' blob: data: gap:; frame-src * self blob: data: gap:;">

Warning: This exposes the document to many exploits. Be sure to prevent users from executing code in the console or to be in a closed environment like a Cordova application.

For anyone looking for a complete explanation, I recommend you to take a look at Content Security Policy: https://www.html5rocks.com/en/tutorials/security/content-security-policy/.

"Code from https://mybank.com should only have access to https://mybank.com’s data, and https://evil.example.com should certainly never be allowed access. Each origin is kept isolated from the rest of the web"

XSS attacks are based on the browser's inability to distinguish your app's code from code downloaded from another website. So you must whitelist the content origins that you consider safe to download content from, using the Content-Security-Policy HTTP header.

This policy is described using a series of policy directives, each of which describes the policy for a certain resource type or policy area. Your policy should include a default-src policy directive, which is a fallback for other resource types when they don't have policies of their own.

So, if you modify your tag to:

<meta http-equiv="Content-Security-Policy" content="default-src 'self' data: gap: https://ssl.gstatic.com 'unsafe-eval'; style-src 'self' 'unsafe-inline'; media-src *;**script-src 'self' http://onlineerp.solution.quebec 'unsafe-inline' 'unsafe-eval';** ">

You are saying that you are authorizing the execution of JavaScript code (script-src) from the origins 'self', http://onlineerp.solution.quebec, 'unsafe-inline', 'unsafe-eval'.

I guess that the first two are perfectly valid for your use case, I am a bit unsure about the other ones. 'unsafe-line' and 'unsafe-eval' pose a security problem, so you should not be using them unless you have a very specific need for them:

"If eval and its text-to-JavaScript brethren are completely essential to your application, you can enable them by adding 'unsafe-eval' as an allowed source in a script-src directive. But, again, please don’t. Banning the ability to execute strings makes it much more difficult for an attacker to execute unauthorized code on your site." (Mike West, Google)

To elaborate some more on this, adding

script-src 'self' http://somedomain 'unsafe-inline' 'unsafe-eval';

to the meta tag like so,

<meta http-equiv="Content-Security-Policy" content="default-src 'self' data: gap: https://ssl.gstatic.com 'unsafe-eval'; style-src 'self' 'unsafe-inline'; script-src 'self' https://somedomain.com/ 'unsafe-inline' 'unsafe-eval';  media-src *">

fixes the error.

The self answer given by MagngooSasa did the trick, but for anyone else trying to understand the answer, here are a few bit more details:

When developing Cordova apps with Visual Studio, I tried to import a remote JavaScript file [located here http://Guess.What.com/MyScript.js], but I have the error mentioned in the title.

Here is the meta tag before, in the index.html file of the project:

<meta http-equiv="Content-Security-Policy" content="default-src 'self' data: gap: https://ssl.gstatic.com 'unsafe-eval'; style-src 'self' 'unsafe-inline'; media-src *">

Here is the corrected meta tag, to allow importing a remote script:

<meta http-equiv="Content-Security-Policy" content="default-src 'self' data: gap: https://ssl.gstatic.com 'unsafe-eval'; style-src 'self' 'unsafe-inline'; media-src *;**script-src 'self' http://onlineerp.solution.quebec 'unsafe-inline' 'unsafe-eval';** ">

And no more error!

Adding the meta tag to ignore this policy was not helping us, because our webserver was injecting the Content-Security-Policy header in the response.

In our case we are using Ngnix as the web server for a Tomcat 9 Java-based application. From the web server, it is directing the browser not to allow inline scripts, so for a temporary testing we have turned off Content-Security-Policy by commenting.

How to turn it off in ngnix

• By default, ngnix ssl.conf file will have this adding a header to the response:

  #> grep 'Content-Security' -ir /etc/nginx/global/ssl.conf add_header Content-Security-Policy "default-src 'none'; frame-ancestors 'none'; script-src 'self'; img-src 'self'; style-src 'self'; base-uri 'self'; form-action 'self';";

• If you just comment this line and restart ngnix, it should not be adding the header to the response.

If you are concerned about security or in production please do not follow this, use these steps as only for testing purpose and moving on.