I'm operating the http://gcc-melt.org/ website (a static website on MELT, a domain specific language to customize the GCC compiler), which should be of interest only to software developers. It is a Linux VPS (which I am renting from OVH in France) with lighttpd as the web server.
In the access.log
file (which is not accessible thru HTTP), I have lines like
46.118.226.141 gcc-melt.org - [27/Dec/2015:09:10:13 +0100] "GET / HTTP/1.1" 200 21170
"http://gardena.ru/" "Mozilla/4.0 (compatible; MSIE 6.0; Windows XP)"
46.118.153.165 gcc-melt.org - [26/Dec/2015:18:53:05 +0100] "GET / HTTP/1.1" 200 21170
"http://brendbutik.ru/" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1)"
193.90.12.90 gcc-melt.org - [27/Dec/2015:09:18:11 +0100] "GET / HTTP/1.1" 200 7226
"http://burger-imperia.com/" "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/37.0.2062.120 Safari/537.36"
(Of course I am splitting the lines in two to show easily the refering site, such as "http://gardenia.ru/"
for the first one)
I quickly checked, and the referring URL above "http://gardenia.ru/"
, "http://brendbutik.ru/"
, "http://burger-imperia.com/"
are completely unrelated to software development, so I don't understand why would unrelated websites refer to my web site?
I do know that it is quite easy to fake such a referrer by putting a bogus HTTP Referer header field in the HTTP request, but I don't understand why people (or perhaps bots) are doing so...
PS. It is on purpose that I am leaving the real Referer
URL. They might useful to others, so please don't edit my question to hide them. Of course, be cautious if browsing them!
NB. I have manually "migrated" (by copy/pasting) this question from Unix&Linux. And that question on webmasters is close to mine.