home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.protocols.tcp-ip
- Path: sparky!uunet!shearson.com!snark!pmetzger
- From: pmetzger@snark.shearson.com (Perry E. Metzger)
- Subject: Stopping only incoming TCP connections (was: Firewall usage)
- Message-ID: <1992Jul28.202211.14029@shearson.com>
- Sender: news@shearson.com (News)
- Organization: /usr/local/lib/news/organization
- References: <BrsM1C.36v@cs.columbia.edu> <DRW.92Jul27143657@jordan.mit.edu> <17011@ulysses.att.com>
- Date: Tue, 28 Jul 1992 20:22:11 GMT
- Lines: 26
-
- smb@ulysses.att.com (Steven Bellovin) writes:
-
- >The *real* definition is that the connection was initiated
- >from the inside. Otherwise, the packet could be from a connection
- >initiated *from* port 513 on a dedicated attacker's machine, and to
- >some service on an inside machine. But routers don't keep track of
- >connections, they look at individual packets.
-
- I was under the impression that if you filter all the SYN packets from
- one direction that aren't SYN ACKs, bingo, you can't initiate any
- incoming TCP connections. Nice and stateless. The only flaw is that
- implementations that seperately ACK the initiating SYN and then send
- their own SYN won't be able to connect, but they are rare. Connections
- could still be hijacked by various mechanisms, but they can't be
- initiated, and there is a limited amount of damage a hijacker can do.
- I was also under the impression that some recent routers will actually
- let you do this trick.
-
- Am I egregiously wrong about this?
-
-
- --
- Perry Metzger pmetzger@shearson.com
- --
- Just say "NO!" to death and taxes.
- Extropian and Proud.
-