Slashdot Mirror
MS Monthly Patch Omits Word Zero-Days
bungee jumper writes "Microsoft released four bulletins with patches for 10 vulnerabilities but there are no fixes for known MS Word zero-day flaws that are under active attack, eWeek.com reports. The January batch covers critical bugs in Excel, Outlook, and Windows. The first confirmed Windows Vista flaw, a denial-of-service issue that was publicly released on an underground hacker site in Russia, also remains unpatched." eWeek notes that Microsoft originally scheduled eight bulletins for release, but pulled four last Friday without explanation.
The patches caused more harm than good so they decided to pull them?
Damn them for not releasing patches that make a more unstable system! Damn them I say!
Local elevation of privilege is now considered a DoS attack on Vista? I guess even submitters don't have to RTFA here anymore to get published. I did read the article though since I was worried about any DoS attack for Vista and wanted to see what ports, processes, etc. it was using. All that was there though was a local only elevation of privs (where an authenticated user logged on to the box can get admin rights). Not good of course, but far from a DoS...
"Zero-day" is an exploit classification.
It goes like this. Software has bugs. These bugs can cause security vulnerabilities, which are then published and patches issued to fix the vulnerabilities. Hopefully, all this happens before the black hats can take advantage of -- or exploit -- these vulnerabilities.
An exploit of a vulnerability is the virus, worm, SQL injection, hack attempt, etc. itself. An exploit can be labelled "zero-day" when an in-the-wild exploit has been detected on the same day that the vulnerability was made known to the security industry. Most often, "zero-day" means "we learned there was a vulnerability when we found this exploit". This is rather like finding out the locks on your doors don't work when a thief has already been and gone. Zero-day exploits then will have a maximal timeframe to affect vulnerable systems since no work has been done on fixing the vulnerability (presumably).
The Slammer worm, for example, was an [i]exploit[/i] of MS SQL Server 2000. SQL Server 2000 had a buffer overflow vulerability which was the subject of Slammer. Slammer was not zero-day, however, since this security vulnerability had been known about for many months and MS had already issued patches for it (six months prior to Slammer).
The vast majority of exploits are *not* zero-day, but uninformed reporters for computer news services (like CNet, or anything Ziff Davis owns) are now using "zero-day" as a synonym for "new vulnerability" instead of the proper "new exploit to unknown vulnerability".
The road to tyranny has always been paved with claims of necessity.