|
An email harvester!! What would be fun is to poison the well for this particular IP. Not too long ago, I had that problem with a personal site being crawled by a harvester. I created a program to generate bogus emails. It will poison the harvester's database and the person who is using that database for spam will have a few thousand (or more) bad addresses and have to throw out the list. When that particular IP address shows up again, they won't get the forum, but a generated bogus list. Jeffrey Flaker Senior Programmer/Analyst Linens 'N Things 6 Brighton Rd Clifton, NJ 07015 Phone: 973-249-4384 Fax: 973-249-4901 http://www.lnt.com "A good player makes himself look good; a great player makes the team look good." Author unknown -----Original Message----- From: web400-bounces@xxxxxxxxxxxx [mailto:web400-bounces@xxxxxxxxxxxx] On Behalf Of Bob Cozzi Sent: Friday, September 09, 2005 9:45 AM To: 'Web Enabling the AS400 / iSeries' Subject: RE: [WEB400] RE: Socket Accept() error occurred in QHTTP(classicserver) Scott, What it looks like is that someone from a specific IP address, was going through our RPG Community Forum messages, one by one, and following those links to the individual replies and hunting for email addresses. So I've done two things. (1) I removed the function/page that allows users of the forum to specify their email address and therefore the link to the page that would allow sending directly to those email address. And (2) I've blocked that IP address. Although I'm sure tomorrow a new one will be used. Thanks for the tips. -Bob Cozzi www.RPGxTools.com RPG xTools - Enjoy programming again. -----Original Message----- From: web400-bounces@xxxxxxxxxxxx [mailto:web400-bounces@xxxxxxxxxxxx] On Behalf Of web400@xxxxxxxxxxxxxxxx Sent: Thursday, September 08, 2005 2:23 PM To: Web Enabling the AS400 / iSeries Subject: Re: [WEB400] RE: Socket Accept() error occurred in QHTTP (classicserver) > Is it really though Bob? If I look at my logs that is the period of > time when I get most DOS and other similar hack attacks. Is it > possible that some attempt at a buffer over-run exploit or something is triggering it? Sounds more like a SYN flood than a buffer overrun! To establish a TCP connection, a client computer sends a SYN (Synchronize) packet to the server. This tells the server that a new connection is coming, so it sends back a SYN/ACK (Synch Acknowledge) packet and the accept() API creates a new socket descriptor for the connection. A SYN flood is where someone writes a program that generates lots of SYN packets from fake IP addresses and sends them as fast as possible to a server. This causes the server to think that lots of simultaneous connection attempts are coming in, and it'll open up descriptors and send back SYN/ACK packets. Eventually, they time out and the descriptors close, but if you send the SYN packets fast enough, the server won't be able to recover and will run out of descriptors. That's what the "Too many files open" message means -- that you've run out of descriptors. IBM could fix the problem by modifying the TCP/IP stack to be more careful in uses as few resources as possible until the complete session is set up, or they could drop SYN packets when they come in faster than a certain rate, there's lots of different strategies. More info on SYN flooding can be found here: http://www.cert.org/advisories/CA-1996-21.html I don't know for sure that this is what's happening, it's just a guess on my part.
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.