Educause Security Discussion mailing list archives
Re: Compromise Email Accounts
From: Steven Tardy <sjt5 () ITS MSSTATE EDU>
Date: Tue, 3 Feb 2009 17:46:28 -0600
Jeremy Mooney wrote:
Steven Tardy wrote on 2/3/09 14:10 :i had a "lightbulb" moment a few months ago. most of the compromised logins are from ip's contained in the spamhaus sbl list. 1) check every login against the spamhaus sbl list.I'm curious if there was any fallout from traveling users, especially those in other countries. We've had many legitimate logins from users traveling in specific areas of the world come from the same IP blocks/ISPs used to access compromised accounts. I haven't specifically looked for sbl listings, but when analyzing some compromises had problems expanding the search scope (IIRC even to the /24 level) to find possibly related activity (too much noise from legitimate traffic). I'm concerned that this approach may either be ineffective or have many false positives.
we've had 0 false positives using this method. the spamhaus sbl is the "spammer" list. the spamhaus pbl is the "policy" list(which includes dial-up/dynamic pools). the spamhaus sbl does NOT include the spamhaus pbl. i mentioned this for others to detect compromises sooner, at login, before spam are sent. if all you do is warn based on this, it's a step in the right direction. knowing is half the battle. (:
2) reject email's with "X-Originating-IP" in the spamhaus sbl list.Do you mean this header as inserted by your local webmail system or your edge system, or are you filtering incoming SMTP messages based on where it originated before the sending SMTP server?
both. this was an unforeseen yet beneficial side effect. currently we only do this outbound, but are investigating inbound as well. if you're not going to accept email sent from a spamhaus sbl ip, why accept email with an X-Originating-IP from a spamhaus sbl ip? i mistyped, we don't reject, we quarantine. that way if someone contacts us about emails that didn't go through, the emails can be identified and processed at a later date. -- steven tardy network services mississippi state university
Current thread:
- Re: Compromise Email Accounts, (continued)
- Re: Compromise Email Accounts Mike Porter (Jan 21)
- Re: Compromise Email Accounts Schumacher, Adam J (Jan 21)
- Re: Compromise Email Accounts Jesse Thompson (Jan 21)
- Re: Compromise Email Accounts Russell Fulton (Jan 29)
- Re: Compromise Email Accounts Sabo, Eric (Jan 29)
- Re: Compromise Email Accounts Joe Vieira (Jan 30)
- Re: Compromise Email Accounts Russell Fulton (Feb 02)
- Re: Compromise Email Accounts Daniel Bennett (Feb 03)
- Re: Compromise Email Accounts Steven Tardy (Feb 03)
- Re: Compromise Email Accounts Jeremy Mooney (Feb 03)
- Re: Compromise Email Accounts Steven Tardy (Feb 03)
- Re: Compromise Email Accounts Jeremy Mooney (Feb 04)
- Re: Compromise Email Accounts Kellogg, Brian D. (Feb 04)