Relatively self-taught user over here, so not sure if this is basic, but Google is indexing our site via https. Which, when you click to get to the home page, is not rendering a section that has a javascript on it. All the text is laid on top of each other, rather than rotating through.
However, if you go to our home page with the address http:// the java is working fine. Any help correcting this would be great.
Thanks!
You have to serve your JS over https as well, because when your https page refers to http content it is considered as insecure data, and your browser won't load it.
It looks like you are using absolute urls for your scripts. or cdn without https. if yes, use relative urls for your local script and use https:// for cdn script paths. You can simply define //cdn path without http: or https: so that the script will load according to the site path.
Related
I recently switched my http site to https.
Since I made this change, one of my projects no longer works.
In this random generator, the background using the Three.js library does not appear when the url is in https
But in http , it works beautiful and well.
Does anyone know any solution to this problem?
Thank you for your consideration in this matter.
You can download the three.min.js file and save it on to your https server.
And instead of loading the script from http server, provide the path of your https server where you put the three.min.js file.
<script src="https://[path to your server]/js/three.min.js"></script>
Open your developer tools and look at the console, you'll find the explanation:
What this means is that you're not allowed to have an https:// site that loads some content via http://. This is called "mixed content" and it's a security vulnerability. If you're serving a secure site, make sure all your content is using the HTTPS protocol.
Here are the links to the page in question:
http version
https version
The https version of the page doesn't render the video embedded on the top of the article. I inspected the source code and found that HTTPs is missing an entire block of code, as you can see in the images below:
I wonder how this happens? Isn't http(s) just the protocol to communicate with the server? Why do I get different code by using different protocols?
From Chrome Developer Toolkit:
The page at 'https://www.eyeviewdigital.com/blog/eyeview-launches-addressable-tv-ads-with-cablevision-dish-programmatic-tv-w-clypd-and-wideorbit/' was loaded over HTTPS, but requested an insecure script 'http://pshared.5min.com/Scripts/PlayerSeed.js?sid=281&width=480&height=401&playList=519141523'. This request has been blocked; the content must be served over HTTPS.
You typically need to load your content in either http or https. Mixing them together results in the error above.
On line 252 replace
http://pshared.5min.com/Scripts/PlayerSeed.js?sid=281&width=480&height=401&playList=519141523
with
https://delivery.vidible.tv/aol?sid=281&width=480&height=401&playList=519141523.js
Long story short the content was never loaded so the page looked different.
You're right that there should be no difference based on http/https as it is just the transport layer. There must be something in the server's code that is producing the html, that behaves differently based on the url used to make the request. I suggest you provide information on the web server code to help analyse the issue.
This is most likely caused because you're loading elements from insecure sources. Google Chrome for example, loading an image over http when your page is loaded over https can result in the image not loading at all. This is all for security purposes, of course. Just go through all of your code and make sure all sources are accessed through secure https so that they load in properly.
I have a web application which has root html and this html (say index.html) loads some java script. This application is accessible through https and I want to load one java script which is exposed over http.
https://mydomain/index.html contains this line of code:
<script src="http://unsecure/custom.js" type="text/javascript"/>
When I try to run my application thorugh IDE everything works fine but problem happens when I bundle my application in war file and run it.
It fails to load the included java script by saying:
[blocked] The page at https: //mydomain/ ran insecure content from http: //unsecure/custom.js.
Is there any way to load this unsecure javascript or do I need to publish this unsecure javascript through a secure way and then access it (can change this included java script protocol from http to https)?
All content which is used by a site, which is accessed through HTTPS, must use HTTPS as well, otherwise you get this warning.
The reason for this: If not all content of a HTTPS site is HTTPS, the browser can't tell that the site is "secure" and therefore gives the user a warning.
You could either do below two things:
1) Download it over your local and create war. Then you would be using relative path.
2) Place in some https location.
If its a third party library and you do not have control on the frequent changes that would happen for this library, you could ask them to put it in https. Majority of the times hosted JS would be both http and https too.
I have just moved my site from http to https and IE-9 started showing non-secure content warning at home page. This warning is understandable because i have one http call to googleapi for getting jquery script. But when I login and enter the inner pages there is no warning from IE despite the fact that most of the images are coming from other servers through http protocol.
So the question: Is getting image over http is fine when accessing site over https? Does only css and js matters? or shall I have to get all the data through HTTPS? If so how is my scenario justifiable (getting images over http from other server on https page without warning)?
If you load CSS and JS over HTTP then an attacker can inject executable code. Unfortunately IE will execute JavaScript within CSS. The problem with loading images over HTTP from the same domain is that the browser will likely spill the session id in plain text which is a violation OWASP a9.
You can use the protocol-relative URL on all your urls to avoid this issue in IE.
Basicaly, instead of linking to a js/image/css by using its full path with the protocol, you instead link to it by leaving out the protocol bit and just using a double slash, //.
This will have the effect of all the above links inheriting the protocol from the parent page.
Of course this depends on you having valid SSL certs on the domains you're serving the different files form.
One other thing to note also is that images in your pages or CSS that are done using data URI could also cause mixed content warnings in IE.
To find out what files are causing issues, I recommend using Fiddler
There is also another tool that a fellow SO user, Eric Law wrote:
Install it from http://www.bayden.com/dl/scriptfreesetup.exe and you will get a different mixed content prompt which shows the exact URL of the first insecure resource on the page. That tool is basically a prototype and you should uninstall it when you're done with it. It works on IE8 and you should install it as admin.
If your page gets called using https protocol, any relative path to an external css will be called using https protocol as well. Are you really need to encrypt/decrypt css contents? :D However, if you use absolute path referring to an external css, you can specify the protocol to use, generally http rather than https. Maybe I am wrong!
Please let me know whether or not my mental model is totally wrong.
Yes you are correct, relative paths will use the same protocol you're currently on, fully qualified paths can include whatever protocol they want.
You typically want the relative behavior to act like it does, otherwise you'll get a insecure-content warning if you're loading http:// resources on a https:// page in most browsers.
As others have answered, your model is correct. A really fun feature that many people are not aware of is protocol relative urls. "//www.foo.com/bar.html" is a perfectly valid URL and will preserve the current protocol.
This is very handy if you are outputting HTML which must work on http and https but needs some host names (e.g. to support static.foo.com / images.foo.com).
Yes, css/js specified by relative path will be loaded using HTTPS. No, you won't have to decrypt the contents, the browser will do it for you. Yes, you can refer to css/js by using absolute path, specifying the protocol, including HTTP. One thing you have to be careful is that some browsers will prompt a warning about a secure site loading insecure contents.