Forum Moderators: open
Target UK site hosted in UK.
Tie-in of IPs etc made through web logs, trap logs and the form itself, which reports source info (ip, time, ua etc) in its hidden fields.
1. Seek form (time: 12.41.54)
UA: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; (R1 1.5); .NET CLR 1.0.3705; FDM)
Bot from softlayer on 75.126.23.nnn proxies through indonesian IP 222.124.208.nnn (apparent broadband) - IP hosts web page with suspicious content including winbox download and telnet instructions. Referer obviously fake (google.fr from a search that could not and did not resolve to the targetted domain). IP rejected (all softlayer rejected) with a form for feedback by real people (or form-filling bots!). Target was a guestbook URL in a mode commonly associated with scammers/scrapers on this site.
2. Form filled and submitted (time: 19:38:28)
UA: Mozilla/4.0 (compatible; MSIE 6.0; Windows 98; Win 9x 4.90; Creative; FunWebProducts)
Form submitted 7 hours later with obvious formspam content by IP 61.18.170.206 (hong kong cable). Apart from spam in form's only free-entry field the rest of the submitted form included the original rejection IP, time, UA etc.
3. Lose response? (time: 19:38:30)
UA: Mozilla/4.0 (compatible; MSIE 6.0; Windows 98; Win 9x 4.90; Creative; FunWebProducts)
The returned Thanks page was read by another IP in the range 61.18.170.198. It's possible that these IPs are part of an aol-type multiple IP system.
Spam:
The form contents included the common formspam triple repetition of the URL with the link text being variations on the theme: "national widgets".
The URL was for a "widget investigators" and included today's date in the directory format /2009/03/14/
[edited by: incrediBILL at 9:56 pm (utc) on Mar. 14, 2009]
[edit reason] removed specifics [/edit]
On the other note just picked up a client (xcart) that has a Gardening site, they were working on the content for the past several years, over 1400 pages with really interesting info about Plants and such. Few months ago their developer installed/activated Review Product plug-in. With in days site was bloated with comments posted by bots. Site tanked in SERP. I have deleted over 20 Megs of spam comments from DB. Really a shame.
The form submission system for complaints is something new on the server. It used to display an unlinked URL but a customer got feisty about his customers being too thick to understand how to complain (they use some very paranoid access methods and can't seem to understand simple instructions).
A few hours ago I had five submissions together, all from various internetserviceteam / NETDIRECT IPs - now there's a surprise!
These were not formspam - they were immediate submissions, not delayed like the earlier one - and had no content in the free-entry field. I'm rather at a loss as to what they were doing.
Oddly, they didn't invoke the Thanks page, which got me wondering. I think it's because there is a redirect within the form parser to send the browser to a different script - browsers are ok with this but some robots seem upset by it; although not the form spammer.
NETDIRECT IPs are, of course, all blocked but the point of a complaints tool in this scenario is that you have to let the jerks through. Well, maybe. Soon as I get time I'm adding a trap to the trap. :)
78.159.112.nnn
69.46.16.nnn
89.149.253.nnn
212.95.63.nnn
212.95.63.nnn
All five postings were from the one original form access from 78.159.112.nnn, all posted within 8 seconds of the original access and all with the same (deformed and obsolete) UA:
Mozilla/4.0 (compatible; MSIE 5.0; Windows 98; DigExt\)
The site gets a lot of scrape and injection attempts.
In this case the lack of protection is deliberate since it's designed as a method whereby people blocked from my server can bring legitimate complaints, so it has to be open - although not as open as it currently is: this is an opening from a previously closed system which for some reason complainers found difficult to comprehend.
This is only incidentally about forms: the complaints are about being blocked from sites in general, not from normal forms, which are seldom a problem since they have their own protection.
This is partially experimental anyway, and is producing some peculiar results and interesting insights.