Forum Moderators: DixonJones
In looking in my logs I saw that i was getting a lot of referrals from a site on which has no link to mine, and after getting into the raw data I see that it is a single page on that site which it sending 'get' commands to my images. The thing is, when I go to the page in question, I see that it is a pop-up which has only an image, a close button and a link back to its own home page.
So how could this pop-up from another site appear in my logs as a referral to mine, and/or how could it get images from my site? Thanks for any and all input into the matter.
Take care,
Josefu.
<img src="http://www.your-domain.com/images/your-image.gif">
Here is a helpful thread [webmasterworld.com].
The page could could save the image somewhere, email it to someone, or just grab it and not used and not do anything with it (maybe someone is trying to drain your bandwidth).
Hope this helps. Let me know if you have any questions.
***.18.157.108 - - [09/Dec/2003:07:25:51 -0600] "GET /salon/boutique/antiques/images/bar/barpops/journalb.gif HTTP/1.1" 200 646 "http://www.*******.com/portecle/pages/pc101.htm" "Mozilla/4.0 (compatible; MSIE 4.01; Windows 98)"
...repeating thus until all of the images contained in one of my pages is loaded... into the other site's pop-up?
Here's another thought: is it possible to 'spoof' what goes into a site's logs? Also, is it possible, if a site is being used as a proxy, that the logs would show odd information like that? Still that doesn't explain the 'linkless' pop-up connecting to my images. Grrr.
Thanks again, guys : )
is it possible to 'spoof' what goes into a site's logs?
Yes. The HTTP_REFERER value can be set arbitrarily. If the IP doesn't matchethe IP for the host in the HTTP_REFERER it may be spoofed. It's also possible that that server is showing different versions of the page to different visitors. See the Apache forum here or Google for Apache mod_rewrite for more information. Another possibility is that a script on that server is issuing a GET request for your file, but the HTTP_REFERER is spoofed and doesn't actually reside where it claims. Programs like Curl allow such requests to be made and give the user full control over what headers are sent (and a lot more). If you have Curl installed (it's pre-installed on many modern Linux distros) do "curl --help" for more info.
Cheers, and thanks again!
Josefu.
[added]Double thanks as, as I am on Panther OS X, your command line is directly accessable through the terminal and all is quite quite clear now ; ) Looks like someone (a competitor) is in line for a spanking - the IP owner, not the referral[/added]