Educause Security Discussion mailing list archives
Re: Outbound SMTP
From: Michael Sinatra <michael () RANCID BERKELEY EDU>
Date: Fri, 25 Apr 2008 11:24:07 -0700
Roger Safian wrote:
This brings me to one of my concerns. Why do we have to engineer our entire networks in one fashion? How about a research network, where port 25 was open, and an administrative network where it's not? If every time I say lets do X, you respond with but so and so needs X, we make no progress. How about we do X, where practical, and still allow so and so the use of an open network? If network security is going to make significant strides we need to quit catering to the least common denominator.
You know, I thought about this while in the shower this morning. The analogy that came to mind is that a lot of places seem to engineer the network for "bunny-slope networking"--non-technical users who don't really have much in the way of needs (email, web, youtube, IM, whatever). Maybe they make exceptions on a single-user basis either using an automated web-page or a review process, but how scalable is that? What if we instead engineered networks to offer bunny-slope all the way through double-black-diamond? You would still need a ski patrol, especially for the last category, but for people who wanted to be in a secure place where they couldn't get hurt and they could still do their jobs, you could provide that, while not affecting the power users. It does, however, imply no more firewalling at the campus border. That's precisely where "one-size-fits-all" policies originate. University of Washington does something similar with its "project 172" where it provides campus-only RFC1918 addresses, with the option of outbound NAT, for people who want security and don't need e2e transparency. What I am suggesting is probably a bit more involved, but perhaps still doable with a combination of use of RFC1918 and virtualizable firewalls (like FWSM blades) in some parts of the network and globally-routable, public vlans in other parts. It would be a way to provide both security and service (imagine that!). michael
Current thread:
- Re: Outbound SMTP, (continued)
- Re: Outbound SMTP David Lundy (Apr 25)
- Re: Outbound SMTP Jenkins, Matthew (Apr 25)
- Re: Outbound SMTP Roger Safian (Apr 25)
- Re: Outbound SMTP ken lindahl (Apr 25)
- Re: Outbound SMTP Don Nightingale (Apr 25)
- Re: Outbound SMTP Michael Van Norman (Apr 25)
- Re: Outbound SMTP Michael Sinatra (Apr 25)
- Re: Outbound SMTP Joel Rosenblatt (Apr 25)
- Re: Outbound SMTP Basgen, Brian (Apr 25)
- Re: Outbound SMTP Jason S. Cash (Apr 25)
- Re: Outbound SMTP Michael Sinatra (Apr 25)
- Re: Outbound SMTP Michael Sinatra (Apr 25)
- Re: Outbound SMTP Joe St Sauver (Apr 25)
- Re: Outbound SMTP Scholz, Greg (Apr 25)
- Re: Outbound SMTP Halliday,Paul (Apr 25)
- Re: Outbound SMTP John Kristoff (Apr 28)
- Re: Outbound SMTP Tim Cantin (Apr 28)
- Re: Outbound SMTP Mike Porter (Apr 28)
- Re: Outbound SMTP Valdis Kletnieks (May 05)
