Let me quote Bruce Schneier here, as I couldn't possibly express it any clearer:
SHA-1 has been broken. Not a reduced-round version. Not a simplified version. The real thing.
Apparently the research team of Xiaoyun Wang, Yiqun Lisa Yin, and Hongbo Yu (mostly from Shandong University in China) have written a paper (which is not yet generally available) with more details on the attack.
If this really holds, this will surely have severe implications on the security and encryption world. Seems like we're slowly running out of hash functions, as MD5 unfortunately has its own problems, too.
According to their information almost all computers in the department (Linux boxes on Intel, SunOS/Solaris on SPARCs and several other architectures and OSes, I guess) have been compromised.
Someone seems to have retrieved the shadow file(s) with the encrypted user passwords. Two cases have already become known where private user data was accessed using cracked passwords, presumely from the stolen shadow file(s).
Students and researchers are advised to change their passwords as soon as possible. This also affects email, as the same password is used for IMAP/POP3 at the department. Also, one should check all computers, websites etc. where you can login without supplying a password, as those might be compromised, too.
As a preventive measure, user accounts are now handled via LDAP and user accounts where the password isn't changed soon, will be disabled.
As of now, there's no information available how the attackers got hold of the shadow file(s), i.e. which vulnerability in which service they exploited etc.
Update: The whole announcement of the admin group is now available from http://wwwrbg.in.tum.de/passwort_aendern.html (German).
John Cartwright has announced that the mailing list Full Disclosure has been compromised using a previously unpublished directory traversal vulnerability in Mailman 2.1.5. A fix is already available. All subscribers are advised to change their passwords as soon as possible.
Maximilian Dornseif from the Laboratory for Dependable Distributed Systems at the RWTH Aachen has written a blog post called Top 18 Papers in Information Security. Definately worth a read!
There's a broad range of papers from "New Directions in Cryptography" by W. Diffie und M. E. Hellman (public key cryptography) from 1976 to "Intercepting Mobile Communications: The Insecurity of 802.11" by N. Borisov, I. Goldberg, and D. Wagner from 2001.