RE: Attempts to push spam through apache
"Gite, Ashish (Security Consultancy)" <[email protected]>
| Newsgroups | gmane.comp.security.linux |
|---|---|
| Message-ID | <A059D12203FFF84A9459CD941BE494E10E2590@bgeexc04.asiapacific.cpqcorp.net> |
Hi, Following links might be related, you might want to read - Spammers use open Apache proxies http://www.apacheweek.com/issues/03-07-25#security Controlling access to your proxy http://httpd.apache.org/docs/mod/mod_proxy.html#access #Ashish -----Original Message----- From: Peter H. Lemieux [mailto:[email protected]] Sent: Thursday, August 19, 2004 7:26 PM To: [email protected] Subject: Attempts to push spam through apache My apache logs are recently full of entries like these: 211.100.24.173 - - [19/Aug/2004:21:03:48 -0400] "CONNECT 208.17.33.40:25 HTTP/1.0" 200 1844 Obviously this is an effort to pump spam through my server to 208.17.33.40. There are many other target addresses as well. If I telnet to port 80 and enter the HTTP command CONNECT 208.17.33.40:25 HTTP/1.0 the server replies with the 1844-byte home page of this site, as indicated by the "200 1844" part of the log entry. As far as I can tell, this means that these exploit attempts only get a web page in reply and are not able to push the spam through to the intended target. I don't have mod_proxy enabled or anything else that would enable proxying to work. Are these just random spammer attempts to find an open proxy? The fact that there are nearly 35,000 (!) such entries over the past few days suggests that the spammer, or the spammer's software, thinks this exploit is succeeding. How can I be sure that it's not? I've blocked the 211.100.24.0/24 subnet for now, but I'd like to be certain that others can't use the same exploit. I tried a variety of Google searches but haven't found a useful page to read on this subject. Some months ago someone used the recent mod_ssl vulnerability and managed to install an IRC proxy on this server. However I fixed those problems at the time, and there's no evidence that any unauthorized programs, e.g., proxies, are now running. (No, there are no rootkits installed, nor is the ps binary compromised, etc. I'm well aware of such possibilities.) Perhaps the machine was just added to a list of potentially vulnerable servers, and someone else is trying to take advantage of me, even though it's no longer possible? FWIW, I'm running Apache 1.3.27 on RedHat 7.3, but I'd guess these types of exploits only work if there is an open http proxy available, no? Peter
smime.p7s
(application/x-pkcs7-signature, 4.7 KB) - not displayed