kwe@bu-cs.BU.EDU (kwe@bu-it.bu.edu (Kent W. England)) (05/02/89)
You may recall the thread of a discussion I started regarding security on TCP/IP internets. I think it was called "IP authentication of hosts" or something similar. Well, Steve Bellovin of Bell Labs told me about an article he had written and was soon publishing that I should read. I did. I recommend it to your attention. It is in ACM Computer Communication Review Vol 19, No. 2, April 1989 pg 32 available on your news stands now. It is entitled "Security Problems in the TCP/IP Protocol Suite". Steve covers these problem areas: TCP Sequence Number Prediction Source Routing RIP attacks EGP attacks ICMP based attacks The RFC 931 Authentication Server Information dissemination services (finger, e-mail, ...) DNS FTP Network Management Remote Booting snooping and spoofing on a LAN TFTP Privileged Ports and comprehensive defenses based on authentication and encryption.
pearce@tycho.yerkes.uchicago.edu (Eric C. Pearce) (05/02/89)
> Well, Steve Bellovin of Bell Labs told me about an article he >had written and was soon publishing that I should read. I did. I >recommend it to your attention. > > It is in ACM Computer Communication Review Vol 19, No. 2, >April 1989 pg 32 available on your news stands now. It is entitled >"Security Problems in the TCP/IP Protocol Suite". Does anybody know if this paper available on-line anywhere? -- - Ecp.
smb@ulysses.homer.nj.att.com (Steven M. Bellovin) (05/02/89)
In article <PEARCE.89May1131943@tycho.yerkes.uchicago.edu>, pearce@tycho.yerkes.uchicago.edu (Eric C. Pearce) writes: > > It is in ACM Computer Communication Review Vol 19, No. 2, > >April 1989 pg 32 available on your news stands now. It is entitled > >"Security Problems in the TCP/IP Protocol Suite". > > Does anybody know if this paper available on-line anywhere? Given that the paper has been published, I tend to prefer that people go to the original -- if I just wanted to send out copies, I'd post all my musings to netnews. (Actually, I do that as well....) Journals do exist for a reason, after all. And I'll use this as an occasion to plug ACM in general, and SIGCOMM in particular... However -- if for some reason you can't get hold of the printed copy, send me email and I'll mail you a PostScript version. (Actually, the PostScript version is the original; no pieces of mashed tree pulp were ever sent by me to the editor, Craig Partridge.) Or rather, I will as soon as the routing tables here are a bit more stable (to say nothing of more accurate); our move to an NSF regional network went smoothly, but not perfectly.... --Steve Bellovin smb@ulysses.att.com