UNIVERSITY OF CAMBRIDGE COMPUTING SERVICE

Minutes of the 138th Mail and Directory Coordination meeting held on Wednesday, 10th January 2007.

Present: P. Hazel, D.P. Carter, R.J. Dowling, F.A.N. Finch, C.J. Jardine, K.M. Jeary,
P. Mazumdar, B.K. Omotani, P. Stewart, R.J. Smith, C.E. Thompson, J.M. Wilkins

Date of next meeting: 28th February 2007 at 11:15 in C304.

1) Mailman

There are currently 2028 Mailman lists. The number of support queries involving Mailman seems to have dropped off after an initial surge.

2) Insecure Access to Hermes

IMAP and POP access to the system have been secured. TLS sessions now advertise LOGINDISABLED until encryption has been enabled. Approximately 200 email addresses are still allowed to make unauthenticated SMTP connections via smtp.hermes. This will be phased out by the end of the Lent term.

3) Bulk Email

CUSU have requested a contact list for all students in the University, citing ucam-gu-bulletin (all graduate students) as a precedent: this request has been denied. The ucam-gu-bulletin list used to be a list of contact addresses for college MCRs. During the Michaelmas term over 10,000 addresses were added to this list: this is under investigation.

4) Spamhaus RBL lists

Spamhaus have recently changed their policy regarding free use, and are requesting contributions from large institutions such as the University. There has been quite a lot of discussion on the uk-mail-managers list suggesting that a national license arranged through UKERNA is in order. The Spamhaus XBL list is the most useful single RBL list that we currently subscribe to.

Spamhaus provide a Policy Block List similar to the MAPS Dialup User List, listing addresses used by ISPs for domestic dialup or DSL systems. We plan to start using this list on a trial basis.

5) User Agents

Mulberry is still popular now that it has been released as a free product. The version of Microsoft Mail included in Windows Vista seems to work as a secure IMAP client. The built in anti-phishing component seems rather aggressive and generates false positives.

DPC 2007-01-10