WebShield scans file types I have excluded

  1. I have exceptions for “IMAGE/GIF”, “IMAGE/PNG” and “IMAGE/JPG”
  2. I turn on the “Show detailed information”
  3. I shows that ALL images of these types on web-pages are still scanned

Why? Since the WebShield slows down quite a bit (earlier discussion) I don’t want it to scan images, but only other content (like HTML-code, applets, executables etc). But the exclusions don’t seem to work :frowning:

Lars-Erik,
webshield compares the content-type with the information sent by the server. It is possible, that the information sent about the image is somewhat different than you think. For example jpeg files are most probably described as : image/jpeg. I’m sure you have some pages with image/jpg as well, but you should try to add this string to the exclusion list.

Lukas.

I tried adding the “'*.jpg” to the URL exception list, and that works quite good, except when the image is not a static file, but a result of a script or something (then it doesn’t end with “jpg”). I’ll try your suggestion and see.

Didn’t help adding IMAGE/JPEG :frowning:
Still had to add “*.jpg” and I guess that is not that safe (script etc could call them self .jpg right).
Does the case have anything to say (IMAGE/JPG vs image/jpg)?
Any other reason it does not work (I use Firefox)?

Lars-Erik, please post here the image URL and I’ll tell you what is its content-type. Lukas

OK, check the images on these ones:

http://www.vg.no/nyheter/
http://www.dagbladet.no/nyheter/
http://www.aftenposten.no/
http://pub.tv2.no/nettavisen/

Hello Lars-Erik,

I’ve checked: http://www.vg.no/nyheter/

Most of the images on this page have content-type of “image/jpeg”, some of them are gifs, so they have “image/gif”. I have not seen any “image/jpg”.

You might also see in the yellow WebShield bar (if show detailed info is on) listed some URLs with .jpg extension. I have analyzed those, and these are not JPG images! Have you checked that the urls your are reporting are indeed images?

An example can be this document:

[b]URL[/b]: http://www.vg.no/nyheter/uploaded/image/bilderigg/2008/01/15/1200385682690_267.jpg [b]Content-type[/b]: [b][i]text/html[/i][/b]; charset=iso-8859-1

Contents:

302 Found

Found

The document has moved here.


Apache/2.0.52 (CentOS) Server at www.vg.no Port 80

To sum up, I have tested WebShield in the latest version and I have not found a single issue where Content-Type exclusions are not working.

Cheers.
Lukas

What kind of files are those “*.jpg” files then?
Why would a serious newspaper use JPG on non-jpg files?

I have just posted the contents of the file together with its content type. It is a HTML document redirecting you to the actual location.