Slashdot Mirror
Microsoft's "Dead Cow" Patch Was 7 Years In the Making
narramissic writes "Back in March 2001, a hacker named Josh Buchbinder (a.k.a Sir Dystic) published code showing how an attack on a flaw in Microsoft's SMB (Server Message Block) service worked. Or maybe the flaw was first disclosed at Defcon 2000, by Veracode Chief Scientist Christien Rioux (a.k.a. Dildog). It was so long ago, memory is dim. Either way, it has taken Microsoft an unusually long time to fix. Now, a mere seven and a half years later, Microsoft has released a patch. 'I've been holding my breath since 2001 for this patch,' said Shavlik Technologies CTO Eric Schultze, in an e-mailed statement. Buchbinder's attack, called a SMB relay attack, 'showed how easy it was to take control of a remote machine without knowing the password,' he said."
So that's how they came up with the name 'Windows 7'
MG
...and boy are my arms tired.
P.S. I'm dead.
I mean, seriously, most of us have written it off, and it makes bad business sense too.
At work we've cancelled plans to use Win7 and WinVista and are moving to all Linux where we can, just from a staffing level perspective.
-- Tigger warning: This post may contain tiggers! --
Seven years ago, The Register devastated me with this terrible news:
Finally, I can use my favorite thrilling NTLM features without giving in and using NTLMv2!
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
Could a Windows Server Admin worth his/her salt please explain to us what SMB is, who would use it, and if there was a workaround that made the vulnerability a non-issue?
Hmm - there was an attack called C2MyAzz that was even simpler than the man in the middle attack. It would just spoof the handshake between client and server. The attacking workstation would watch for client->server message requesting authentication. The attacking workstation would send a packet back to the client before the server, asking the client to send back a clear-text password. Much easier than a man-in-the-middle attack, and it worked well. When it was released, Microsoft's official response was "most organizations use switches and routers, so this is not a problem". Originally released in 2001, IIRC.
I believe that's "*Hobbit*" ...
(jan '97)
Oh well, I guess I'd better block incoming public Internet traffic on port 139 then. That's a shame because it's been so very useful to have an Internet facing SMB share.
Like any windows server admin reads slashdot.... And the ones that do aren't going to stick their hands up and say "Oh, pick me" so we can all berate them for their choice in closed source server operating systems.
It's always been easy to take control of a machine without the password. Sit down in front of the computer. Now the only thing stopping you is yourself. Oddly enough, that's what keeps most systems up... The fact that the vast majority of people are honest, decent folk. That, and they don't know what a null pointer is.
#fuckbeta #iamslashdot #dicemustdie
According to Google, 1997. Yeah, over a decade ago.
CIFS: Common Insecurities Fail Scrutiny
- SR
I do.
You can make fun of me :)
That said, if you have a Linksys firewall in place, it usually takes care of the issue. Granted the attacks you'll get internally *can* happen, but we have managed to circumvent SMB exploitation via policy settings in Windows. It works fine for us, nice to see they finally patched it though.
The price is always right if someone else is paying.
"I've been holding my breath since 2001 for this patch"
With lungs like that he should try free-diving!
What made it worse? Taking 8 years to fix it or disclosing it before the patch was released?
Further it is not a bug at all. It is essentially badly designed protocol having a hole and instead of abandoning it and making users upgrade, MSFT left this hole open for 8 years. All the in the name of backward compatibility. Why has backward compatibility trumped security for 8 years? It not surprising no one takes MSFT's statements about its commitment to security seriously?
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
How many people were actually a victim of this exploit? Is there one documented case of an electronic break-in because of this exploit?
So does that mean we can expect a Dead Cow Level to be hidden in an included Windows 7 app, a la Flight Sim hidden in Excel 97.
None, none, none, none, none, none, none.
I guess I should get out more...
a.
Now you deserve to be made fun of.
"I use a Mac because I'm just better than you are."
It wasn't meant in seriousness, but if you want to take it and run, feel free.
Just meant that any port blocking software or hardware (as simple AS a Linksys firewall) prevents this from being anything of an issue.
Hell, even Windows built-in firewall will do the trick.
The price is always right if someone else is paying.
http://support.microsoft.com/kb/q147706/
You will see that the affected operating systems are old and that Microsoft long ago told people how to configure their systems to avoid this issue.
I've posted on the Veracode Blog about this issue for clarification purposes.
Here's the content:
With regard to the recent Patch Tuesday fix, there has been an issue fixed regarding NTLM Relaying, that has been around for more than eight years.
In 2000, I wrote an advisory about NTLM relaying (CVE-2000-0834). The problem turned out to be significantly larger than I originally suggested in the advisory. The attack extended to other NTLM-based authentications on other protocols and allowed general-purpose credential theft via a man-in-the-middle attack.
The SMBRelay tool was published in 2001 by Sir Dystic of Cult Of The Dead Cow, and that really took it to the next level. The protocol completely fell apart. It kicked off a number of other analyses of the NTLM protocol that finally resulted in this patch. Eight years after itâ(TM)s discovery.
At least they got around to it. Thanks!
--chris
(Buy my house! http://tinyurl.com/dilshouse)
My #1 beef with Microsoft is that they market it so that every small to medium business owner thinks that everything will all run together happily on one box all "plug-n-play" and snuggly whirring away on the floor of their office closet.
I have the hardest time convincing users that they cannot run their 20-user network on one SBS 2003 server, with Exchange (running OWA and OMA), running their heavily-accessed SQL database, sharepoint, anti-virus server software, backup software, and company file and printer sharing to 5 multi-function copiers and expect 5 9's of freaking uptime.
This is how it is marketed. This is what the end user expects when shopping for a Microsoft solution. You tell them that they'll need at least 3 separate boxes, Server, Exchange, SQL, etc all separate, RAID and ideally a failover system and an excellent firewall for the remote access they look at you like you're nuts. So they buy it and have it set up their way, it works like hell for a year, then they end up paying in the end to have it done again the right way (and more this time, because they have to now migrate off of their old system).
And the Microsoft money machine chugs on.
Just disrupt the deflector shield with a tachyon burst.
"Microsoft has released the specifications for the binary file formats used by pre-2007 Microsoft Office applications"
And we all know how well Microsoft maintains backwards compatibility with its office file formats...