Home page logo
/

basics logo Security Basics mailing list archives

Security and EOL issues (was RE: WMF Exploit Patch released)
From: "Donald N Kenepp" <don () videon-central com>
Date: Sun, 8 Jan 2006 21:32:32 -0500

Hi Matthew,

  Perhaps I was too harsh in my response.  I understand your argument that
since Microsoft does not give others their source code, they should be fully
responsible for maintaining the code.  At the same time, they have decided
that rather than continuously patching the same piece of software, they are
going to do a major rewrite every three to five years and sell it as a new
system.

  Some bugs are design flaws that simply cannot be patched.  This is why no
matter whether we choose closed or open-source, we always have to apply a
certain amount of maintenance to our software.  With open source, we are
largely responsible for ourselves from day one.  While we assume major flaws
will be repaired, there is no guarantee that a fix will ever be released.
With closed source, we can demand someone else fix it for a reasonable in a
reasonable timeframe for a reasonable amount of time.  Eventually the time
and effort involved in patching something rather than replacing it is simply
not efficient, and in both cases, that someone else will stop doing so.  The
only relief with open source is that if no one but you cares, you can still
try to do it yourself.

  You are correct that I should not be addressing this as being directly
about the age of the OS.  This is not about patching vulnerabilities due to
the age of the OS, but rather due to the changes since this version of the
OS and other advancements in that time.  Compare AIX to Windows, and AIX
version X to Windows version X.  I do not know the AIX system model, but it
seems to have obsolete versions as well.  IBM ended software support for AIX
Base Operating System 4.3.3.  Perhaps IBM still does release fixes for
security flaws in 4.3.3 even though they have declared its EOL.  Does it
still release fixes for 3.2?  I would not fault them for no longer
supporting their older kernel versions past a reasonable transition.
Eventually it becomes too much overhead that is better spent elsewhere.

  Every vendor releases new builds and patches, and inevitably there are
versions that should not be used because they are known to be vulnerable and
have been replaced.  I believe you will find that any operating system, just
like any other software program, has a point at which they are no longer
supported.  Should Adobe still be held responsible for fixing security
issues in Acrobat 3?

  It may be that developers or third-party vendors stop supporting the older
OS version.  In the case of open-source, a version is basically no longer
supported when the vast majority of the open-source community simply stops
trying to fix the old material and works on something new.  I am sure there
are kernel bugs and security flaws that will never be found, exploited, or
patched for the Linux 2.0.0 kernel.  Linus has passed off everything prior
to 2.6.x to other maintainers, and most new software expects you to use
2.4.x or 2.6.x.

  At what point should we start over rather than fixing fatal flaws?  Should
Microsoft be putting time and effort into fixing fatal flaws in Windows NT
3.x as well?  At what point would you rather IBM focus on future development
than spending time to fix a fatal flaw in AIX 4.3?

  If we are still using NT 4.0 on mission critical machines, we need to ask
why we are doing so.  What is our obligation to our customers to use current
software on mission-critical machines?  How much are bandages and the time
it takes to put them on costing us against the cost of a new system that
doesn't need as many bandages?  I push to get users and companies off of
operating systems that are obsolete, whether because they have security
exploits, because they are no longer cost effective to maintain, or because
they are no longer supported.  Sometimes I win, sometimes not, but I always
try.
 
  It doesn't matter whether we choose to move to open source or the next
Microsoft product; we shouldn't be using a system at the end of its life if
at all possible.  While WMF may be a high priority at the moment, I would
honestly say that the security design improvements and other general
improvements to the NT family since NT 4.0.SP6a should have been a high
priority before now.

  To your question regarding the NT 4.0 seg fault, Microsoft does still have
what they call "critical updates" for NT 4.0.  They do not feel that the WMF
flaw meets their definition of a critical flaw for NT 4.0, so they may not
put the extra effort forward until such point it meets those criteria.
While it would be nice if NT 4.0 was easily fixed and happened to be patched
along with everything else, it isn't worthwhile to fix NT 4.0 bugs anymore
at the cost of forward progress.  Microsoft threw a lot of weight at this
patch.  Should Microsoft have held the WMF patch release until there was a
fix for NT 4.0 and Windows 98?  Should they work 24/7 on a patch for them
now?  How much weight is reasonable to throw at those older systems?

  Perhaps it would be easier to see if rather than having names like Windows
2000, XP, and 2003, we noted they are all NT version 5.x.  The Microsoft
product timelines might also be helpful:

http://www.microsoft.com/windows/WinHistoryIntro.mspx

  Getting rid of NT 4.0 seems like something drastic because it is arguably
the first widespread and long-lived corporate version of Windows to be
EOLed.  I really wouldn't refer to it as a huge campaign for years followed
by a quick EOL, however.  It's been patched for almost a decade with six
major service releases and countless hotfixes.  There was at least a year of
warning about its EOL, followed by an extra two years of extended support
due to high customer demand and agony over it being the first major EOL
announced.  This isn't under the blanket of ending support; it's under the
blanket of long since ended support.  Almost all software written for NT 4.0
still works on newer versions of the OS.  The reason we don't see patches
anymore is because the new OS versions are becoming so different that
usually the same patch doesn't work for both, or only NT 4.0 requires the
patch.

  I can understand skipping Windows 2K Server if you worried about it as a
beta for Active Directory.  I hope you skipped Windows Me.  I understand
budget issues and upgrade nightmares.

  I believe Windows Vista is NT 6.0, putting NT 4.0 two major kernel
versions and several NT 5.x releases behind.  The question becomes how long
you want your vendor to support each version of the products they release,
and how much effort you want them to put toward the older releases.  I am
glad Vista/Longhorn/Blackcomb was delayed for XPSP2 (NT 5.1's major security
revision) because it marked the beginning of a huge turn-around in security
efforts from Microsoft.  I don't feel things should be similarly delayed for
patches to NT 4.0.

  Sincerely,
    Donald


-----Original Message-----
From: Matthew Schiros [mailto:schiros () gmail com] 
Sent: Saturday, January 07, 2006 11:22 PM
To: don () videon-central com
Cc: info () footvision com; security-basics () securityfocus com
Subject: Re: WMF Exploit Patch Released

Donald,
Perhaps I chose my words poorly.  My point was not that Microsoft was
using this patch as an attempt to push users away from NT 4.0.  I know
that NT 4 was has been EOL'd for some time now, and I'm aware that
there are many viable replacement OS's put out by Microsoft since.  At
the same time though, there's the issue of taking responsibility for
your software when you aren't willing to reveal the source and allow
others to make the fixes that you aren't willing to.  This isn't a
matter of a low-risk bug in a piece of legacy word processing
software, it's a highly dangerous exploit in software that, if its
being used today, is used on what's likely mission critical machines.

While I'd like to say that MS can't have its cake and eat it too, it
can.  It can spend years pushing a product, get everyone to use that
product, and then relatively quickly EOL that product, and get you to
move to something else.  It's a great business model, obviously,
because people keep buying, but at the same time, how many System V or
AIX exploits do you think appear that go unfixed because of the age of
the OS?  What kind of relationship do you have with your customers if
you just refuse to take responsibility for a drastic flaw under the
blanket of ending support?  Nobody is saying that Microsoft should be
obligated, or should even consider, doing anything like doing DirectX
updates or anything, just fix fatal flaws in already existing code. 
If it suddenly turned out that NT 4.0 seg faulted every time it
recieved an incomplete TCP packet (obviously this isn't the case, but
whatever), would you say that Microsoft had no obligation to fix that
problem?

Matt
On 1/7/06, Donald N Kenepp <don () videon-central com> wrote:
Hi Matthew,

  Sadly, it isn't so much Microsoft saying you should upgrade for this
patch, but Microsoft saying you should have upgraded from Windows NT 4.0 a
long time ago.  NT 4.0 has been being retracted from the market since
2001.
It was declared closed for normal support in 2003.  They are now phasing
out
extended support in 2005.  Windows NT 4.0 first showed up back in 1996.
We
have since had 98, Me, W2K, XP, and now Vista is coming.  The server end
has
seen W2K and 2003 with a service pack.  Should an OS be supported for ten
years past its inception?

  Will there be a WMF patch for Windows 95 as well?  One way to look at
things is that Microsoft is an evil empire sticking it to the man.  One
might also say they are the average business with new products.

  Regardless of motive, it honestly costs more to maintain NT 4.0 at this
point than to upgrade to a newer OS.  Red Hat 4.0 also came out in 1996.
The amount of patching, manual configuration, and manual administration
involved in a product that has seen its day come and go is much more
expensive than migration.  There is also a fair amount of default
security,
productivity, and usability gains in the newer versions of these products.

  You can still run programs dating back to Windows 95 and NT 4.0 and even
DOS on Windows XP.  That's a lot of overhead Microsoft built in to ease
transitions.  Skipping one OS version for cost reasons can certainly make
sense, but if you are making things last and your workstations and servers
have a five year lifecycle, so should their operating systems.

  Just for some perspective on 1996:

Dell opened internet sales.

Netgear was founded.

Google was first developed.

Sony entered the PC market.

Microsoft introduced Windows NT 4.0 and Windows CE 1.0.

Sun introduced the Ultra workstation family and licensed Java.

Seagate released the original 10k Cheetah drives at 6GB.

Intel released the 200MHZ P6.  The 266MHz PII didn't come until 1997.

  I do wish you the best of luck in patching NT 4.0 systems if you are
truly
stuck with them, but my recommendation to anyone still on NT is to use
this
as one more reason to present the idea of a new OS to management this
year.

  Sincerely,
    Donald

-----Original Message-----
From: Matthew Schiros [mailto:schiros () gmail com]
Sent: Friday, January 06, 2006 12:47 PM
To: info () footvision com
Cc: security-basics () securityfocus com
Subject: Re: WMF Exploit Patch Released

According to Microsoft, WinNT4 and Win2k SP3 users are out of luck.
Their reccomended "solution" is to upgrade your software to a
supported version.  Obviously, all this means is that they have no
solution at all, but this is hardly the first time that MS has stuck
it to WinNT4 users as part of an attempt to get them all moved over to
2k SP4.  As for the viability of disabling the DLL's in question,
while I haven't had any problems as a result of doing that on the 2k
boxes in the office, I haven't had the opportunity to test its impact
on NT systems.  That seems to be the only way of removing the exploit
from your machines though, and I'd be interested in knowing the
results of your attempts.


On 1/6/06, info () footvision com <info () footvision com> wrote:


Hello Everyone,

Unfortunately there are company who are still running NT4 and I was
wondering which alternative do they have

to face this security breach from the fact that Microsoft do not provide
any
patch for NT4 .

Do they have to disable GDI32.DLL and WGDI32.DLL as suggested previously
for
SHIMGVW.DLL?

Regards.

Ernest Matos

IT Security


-----Original Message-----

From: Matthew Schiros [mailto:schiros () gmail com]

Sent: Thursday, January 05, 2006 10:51 PM

To: security-basics () securityfocus com;
bugtraq () securityfocus com

Subject: WMF Exploit Patch Released



Microsoft has released a patch for the WMF exploit a couple of days

early, apparently due to a faster-than-expected testing process, and,

at least I hope, some consumer pressure. It can be downloaded via

Windows Update, or as a standalone install at:

http://www.microsoft.com/technet/security/Bulletin/MS06-001.mspx

As a note, it appears that all of the attempts to circumvent the

problem via disabling SHIMGVW.DLL were irrelevant, and that those who

discovered that GDI32.DLL and WGDI32.DLL were the culprits were

correct.

Happy crawling.

Matt Schiros

Web Developer

Academic Superstore

www.academicsuperstore.com



---------------------------------------------------------------------------

EARN A MASTER OF SCIENCE IN INFORMATION ASSURANCE - ONLINE

The Norwich University program offers unparalleled Infosec management

education and the case study affords you unmatched consulting
experience.

Tailor your education to your own professional goals with degree

customizations including Emergency Management, Business Continuity
Planning,

Computer Emergency Response Teams, and Digital Investigations.

http://www.msia.norwich.edu/secfocus



----------------------------------------------------------------------------


---------------------------------------------------------------------------
EARN A MASTER OF SCIENCE IN INFORMATION ASSURANCE - ONLINE
The Norwich University program offers unparalleled Infosec management
education and the case study affords you unmatched consulting experience.
Tailor your education to your own professional goals with degree
customizations including Emergency Management, Business Continuity
Planning,

Computer Emergency Response Teams, and Digital Investigations.

http://www.msia.norwich.edu/secfocus

----------------------------------------------------------------------------





---------------------------------------------------------------------------
EARN A MASTER OF SCIENCE IN INFORMATION ASSURANCE - ONLINE
The Norwich University program offers unparalleled Infosec management 
education and the case study affords you unmatched consulting experience. 
Tailor your education to your own professional goals with degree 
customizations including Emergency Management, Business Continuity Planning, 
Computer Emergency Response Teams, and Digital Investigations. 

http://www.msia.norwich.edu/secfocus
----------------------------------------------------------------------------


  By Date           By Thread  

Current thread:
[ Nmap | Sec Tools | Mailing Lists | Site News | About/Contact | Advertising | Privacy ]
AlienVault