Weekly Microsoft Critical Security Issue 518
An anonymous reader sent in linkage to a zd story discussing the latest Windows Security Patches including an especially nice hole letting Java apps gain total control of your machine and assist you in reclaiming disk space by, say, reformating your drive.
It's about time... (Score:4, Funny)
Re:It's about time... (Score:3, Funny)
Re:It's about time... (Score:3, Funny)
jvm (Score:5, Interesting)
Re:jvm (Score:4, Informative)
It's MICROSOFT'S JAVA IMPLEMENTATION.
The problem is NOT Java.
The problem is (and always has been) Micro$oft's purposely broken version of Java.
Re:jvm (Score:5, Insightful)
Re:jvm (Score:2)
I'd guess the number is very near ZERO.
Re:jvm (Score:3, Insightful)
A bit of history and compiler theory.... (Score:5, Interesting)
Unfortunately, Microsoft chose to take a different approach and introduced new operators into the core byte-code interpreted by the Virtual Machine. As these additional instructions were only valid within Microsoft's version, users were effectively left with no choice but to use the exact VM for which the code was compiled. This decision by Microsoft to modify the base instruction set of the Java language made it impossible to port code from one platform to another, thereby ensuring that users would have to remain on the Windows platform. In fact, Java programs compiled for MS's VM would not even work on the same OS if another vendor's VM was used to run it. This is why some applets wouldn't work with the JVM shipped with Netscape (which was Sun's JVM).
The instruction set supported by a Java VM is determined and maintained by Sun. In order to implement your own VM, you must agree to a license with Sun stating that you will not modify the core instruction set. In adding direct support for OS access (such as formatting a hard drive), Microsoft violated this license agreement. Microsoft also added their own keywords to the core language (delegate and multicast [sun.com]) which further ensured incompatibility.
The Java byte code is a single byte in size and, as a result, the Java VM spec supports up to 256 op codes. Not all of them are used, however. Out of those potential 256 opcodes, only 200 valid operators are specified. Opcode 186 is not used, opcode 201 is used for debugging, and codes 254 and 255 are used for trapping and tracing. The remaining opcodes are reserved for future use. Clearly, if a compiler introduces new opcodes, the other compilers won't know about them and won't be able to run programs built with those opcodes. This is in direct violation of the VM specification and is exactly what Microsoft did. This was the basis for the Sun v. Microsoft lawsuit, for which Microsoft was found in willful violation.
So, it would seem as if Microsoft did intentionally break their own version of Java.
If you still do not understand how Microsoft did this on purpose, I suggest that you take a look at the Java Virtual Machine Specification [sun.com], as well as a nice book on general compiler theory [amazon.com].
RTFA (Score:5, Informative)
The three warnings, all issued on Wednesday, involve the Microsoft Virtual Machine for running Java applets on Windows
So it's Microsoft's VM implementation...
But quickly fixed... (Score:5, Informative)
Re:But quickly fixed... (Score:5, Interesting)
One can be excited when they patch things this quickly. My real concern is to whether we will see tons of patches for forthcoming software. That is, will all of the talk of more 'secure' computing be just talk.
I certainly agree that Win 2k, XP, etc. all seem to have more security bugs than you can shake a stick at. Given the problem, the question is can MS make any sort of headway? Can they actually offer a product that will really be stable and secure? My theory is that we will know a lot more about the answer to these questions in six months. If Win 2003 server has 18Mb of patches in the first 6 months then we will know the answer. Personally, I am hoping the start doing better.
Re:But quickly fixed... (Score:2)
Try subscribing to Redhat's automatic update feature. See how many security updates you get then!
To be fair though, these updates are nearly always for applications, not for the operating system.
Re:But quickly fixed... (Score:2)
Re:But quickly fixed... (Score:2)
Re:But quickly fixed... (Score:2)
Re:But quickly fixed... (Score:5, Interesting)
Thanks to a long list of overlapping issues, this is going to cause my employer (and a vendor that shall remain nameless to protect the guilty) a bit of a headache--and I doubt that we're alone in the world on this one.
We are running a Digital Imaging (digital radiology) sytstem that has a web-based server for allowing physicians to review images and interp from "any PC". The viewer itself is Java based...no client required (ahem...vendor speak. Client is downloaded automatically, perhaps? Anyway...) The elimination of the need to manage/install/maintain a client on thousands of different machines was one of the biggest reasons that management chose this particular system/particular vendor.
Background:
Here's how the IT assessment of the product went...
Yay...Java! This will run on any PC! Well, not Mac or Linux, but since we aren't a Mac or Linux shop, this is acceptable (this should have been our first clue).
Well--make that "any PC running Internet Explorer". Perhaps it's something with a particular DOM. We can live with that. We're running IE on all of our machines, anyway.
OK--make that "any Windows PC running Internet Explorer, using Microsoft's Virtual Machine. Sun's won't work". WTF? I thought this was JAVA. Let me guess...this was written using MS Visual J++, right?? Anyway, according to our management (who is undoubtedly quoting straight from the vendor), "it's a lot faster this way."
Ummm--make that "any Windows PC running Internet Explorer, using one of a few versions of Microsoft's Virtual Machine...the most recent ones will *break* the app". Now, where did *that* come from? But sure enough, if an employee gets overly "helpful" and tries to update their system (we still have some 9x systems on the network, and the boss won't let me firewall the Windows Update site), the application breaks. So whatever the vendor did isn't entirely "legal"...the latest VMs "fix" an undocumented feature that they are depending on...
Final analysis: "This sucks. Either plan on installing their Honest-to-Pete MS-VC++ client on 1,000 PCs or pick another vendor."
So, yes, management went ahead and bought the package - warts, J++ and all - from the vendor for a goodly sum, over the objections of the IS review committee. Yes, we've fought with said vendor for the last few months, to no avail (yet). No, the vendor (until now) claims that there is no reason to update their code to be fully Java-as-in-Sun compliant (or even Java-as-in-current-Microsoft compliant, for that matter), and that we should basically stop whining and get over it. But perhaps, just perhaps, we can now point to this and say "Look. Your cusomers *are* at risk. We *must* upgrade our JVM...we have no choice. If your software won't run on the resulting platform then it's not performing as indicated, which frees us from the contract and any pending payments coming due. Hint Hint."
Well, I'm not holding my breath on the vendor updating their code. I am holding my breath about this cycle of Windows Update problems, however. I imagine that the trouble tickets are already starting to come in to our PC support area. "The Radiology viewer doesn't work," they say. "I can't do my job...fix it now!" they demand. Much work to uninstall the new VM. Much work to re-install an older version so they can "do their job". And much sweating while we hope to dodge the bullet of a malicious Java applet through a combination of virus detection software and dumb luck.
Sometimes, a blind patch via Windows Update isn't the best thing to do, unfortunately.
Am I blaming Microsoft for building unsafe Operating System software? Well, yes, but I'm also a realist--you can't expect perfection. But what I'm really blaming Microsoft for is their knowing and purposeful design and dissimenation of a Java VM and Java development environment that was built to be incompatible with Sun's Java. I'm also blaming the vendor for helping support Microsof
Reformatting my hard drive (Score:5, Funny)
Re:Reformatting my hard drive (Score:2)
if a virus of this sort were possible, and bandwidth bigger it would be interesting to see a rampant virus of Penguins.
Like XBill? (Score:2)
if a virus of this sort were possible, and bandwidth bigger it would be interesting to see a rampant virus of Penguins.
You mean, like in xbill [xbill.org]?
I don't think we can be too critical, actually. (Score:4, Interesting)
Couple of remote roots in Samba, a local ptrace in the kernel and a few OpenSSL probs to get you on the system initially.
Re:I don't think we can be too critical, actually. (Score:2)
Rus
Re:I don't think we can be too critical, actually. (Score:2)
Re:I don't think we can be too critical, actually. (Score:5, Interesting)
It is interesting you say this, and I think this is to blame for a good amount of FUD on both sides.
First off, anyone thinking there will make an uncrackable system is both naïve and asking for someone to break into their system. No one will make an unbreakable system (and plugged in), it is just that harder systems will take longer to break.
In this same vein, the nature of a piece of software's security can not be measured only in security updates or patches. You are right, OSS has had cracks recently, but the fact that you know about them and that (most) of them are fixed is reassuring. I would venture to say that something that didn't have patches or updates was simply not worth hacking, or not maintained (i.e. MS has not sent out patches/security updates for Win3.1, but does that mean there are no more problems with it?)
On the other hand, This believe must be mitigated by the understanding that more vulnerabilites announced are also not necessarily a good thing, and may reflect in shoddy programming.
This is the double edged sword that we must cut ourselves with. The real "tell" (I believe) is the level of sophistication in (most of) the vulnerabilities. Unfortunately, I know almost nothing, so I leave it up to others to tell me how bad they are. I guess it is a good thing I am not a sysadmin.
Thank Goodness... (Score:2, Funny)
Re:Thank Goodness... (Score:2)
Hmm... (Score:4, Interesting)
The timing seems a little too good to be true...
Re:Hmm... (Score:5, Informative)
Actually the court order is to put Sun's version of the JVM into Windows - exactly to fix this type of stupid problem.
JDK (Score:5, Funny)
Good thing Microsoft JRE is so broken, that all exploits ended up not working!
Write once, debug everwhere.
Re:JDK (Score:2)
That's copyrighted by Symantec -- it was the ad tagline for the debugging component of Visual Cafe.
I don't think they knew how funny it was, but I had it on my cube wall for a time (1998-1999 era)
--
Ok (Score:2, Insightful)
Re:Ok (Score:2)
Not trying to be flaimbait or anything here, but I'd like to see some evidence. Every where I turn I see somebody up on their soapbox shouting about this insecure Windows...
I myself use non-microsoft operatingsystem not because of security, but another reason to add to my list would be great (with proof of course)...
Re:Tech support != "geeks" (Score:3, Informative)
No, they don't occur purely because of failed memory allocations. If a memory allocation fails, the pointer will be uninitialized and the application will segfault due to that. You just don't understand what you are talking about.
Several people there have said they've actually solved segfaults by re-clocki
Hit the monkey and win $$$! (Score:2, Troll)
Oh joy! Cha-ching!! (Score:2, Insightful)
Mo money for me! Everytime this happens I go out and patch up my customers. Cha-ching, cha-ching!
And I always offer and *suggest* that they go with Linux but they are *afraid* of change.
They would rather live in fear and subserviance than live in security freedom...
Go figure..
Re:Oh joy! Cha-ching!! (Score:2, Insightful)
Not that I love M$, but it seems that your bashing Micro$haft unjustly. Linux seems to be pumping out even more fixes and patches than old Billy boy's crappy product.
It seems like for the last month or so I have received at least 2 RedHat erratas a day, and the majority of them are for security reasons.
For my RedHat email server, there have been 98 updates put out by RedHat and the Linux community. Of those 98, 16 were bug fixes, 4 were enhancements, and 78 were for security c
Re:Oh joy! Cha-ching!! (Score:2)
And how many patches were included in the service patches?
Re:Oh joy! Cha-ching!! (Score:2)
Microsoft did well in this instance, and perhaps this is the start of their focus on security.
You'd think though, that with a software company that's as big as Micro
Not quite true... (Score:4, Insightful)
Nevertheless, my last sentence becomes quite irrelevant, as Windows user tend to work as $root.
Re:Not quite true... (Score:2)
In that case you can't fully blame m$, as you wouldn't blame a unix manufacturer if a unix admin was always running root, and a small bug could prove to be disasterous.
And in other news... (Score:3, Insightful)
Would this be considered Microsoft bashing? Nahhhhh....
But of course I won't mention that windows Update is free and it worked immediately after I got notice of the patch.
Re:And in other news... (Score:2)
Really? I've had no problem whatsoever. up2date seemes to grab them fine for me. Then I can ship them to whichever machine I want to.
Must be your network...
Re:And in other news... (Score:2)
Re:And in other news... (Score:2)
Re:And in other news... (Score:2)
Re:And in other news... (Score:2)
Sure there are _lots_ of vulnerabilities in the Linux distros but look at what they include. A typical Linux distro contains everything a typical desktop and/or server user could want. Most distros contain at least 3 mail servers. How many mail servers ship with Windows? How many serious apps are considered part of Windows? Yet in this case you are comparing one application to an entire distribution! I guarantee that if you plug as many apps into the Windows Update serv
Re:And in other news... (Score:2)
Re:And in other news... (Score:2, Informative)
Look troll, blah blah blah.
I'm sick of this crap. There is no 100% secure OS for the x86 platform. None.
Re:And in other news... (Score:2)
Now to expand. Originally, Microsoft recoginzed that Java did several things that they liked and created extensions to allow people to write Windows applications using Java. Sun cried foul, so these extensions are mostly disabled. (I think you could access any COM object through MS Java, and that was one of the extensions.) I don't remember the exact details, but for a while, MS was very excited about the
Finally! (Score:2, Funny)
Finally someone wrote something to get rid of all that spyware thats installed itself on my system! Thank you MS!
Help me out here (Score:3, Insightful)
Just curious. I mean, if the intent is to inform.
Re:Help me out here (Score:2)
Re:Help me out here (Score:2)
However, despite having to scramble for bugs o
Re:Help me out here (Score:2)
You certainly have a point there. I've become very selective in what type of update I apply to my boxes, because of the potential I see for something going belly up when 15 "cumulative patches" are applied at the same time. IMO, Microsoft's solution to this problem is closer to being a problem itself. The "OK, we'll disclose all the bugs and here are all the patches for them" approach doesn't really work. For me, at least.
Re:Help me out here (Score:2)
KDE 3.0.5b and KDE 3.1.1a [kde.org] were announced today to address a Ghostscript-related vulnerability.
Please compare Oranges to Oranges (Score:2, Insightful)
You have to compare the SEVERITY and NATURE of the bugs. Sure, there are bugs with whatever OS, but as to this level of Severity and of this Nature, you're just wrong, there are not that many with Linux, Apple or Solaris or whatever. Windows takes the cake.
If you think this is all overblown hogwash,
Clueless (Score:5, Insightful)
I'm sure the above is a troll, but I'll answer anyways. When you install windows, you get, well, windows. And internet explorer, and freecell. That's about it.
When you install linux from RedHat (or Mandrake or...) you get the OS, severl browsers and mail clients, 2+ office suites, 4+ text editors, java, perl, c, python, 25+ games, 3+ window manages, etc (not that you have to install all that - but they're available in the install).
I'd say Redhat is doing great to only have 2x the security bulletins as microsoft considering they supply 4x or 5x the software on their cd's.
Plus, it's been documented many times before that bugfixes are available much quicker in the OS world than the MS world.
I'm increasingly convinced that Linux is dying off. The lies and distortions we are seeing on slashbot have become more and more desperate over the past two years.
Name one "lie" regarding linux that you've seen on slashdot that's demonstratable not true (articles only, not posts). Remember, nobody is going to agree with all the opinions expressed on this site.
Re:Clueless (Score:2, Insightful)
That's about it? Well, I'm sure that your second sentence is a troll. The OS, IE, and freecell? Really, man. When's the last time you installed Windows?
Re:Clueless (Score:3, Interesting)
Lie: Posted on Slashdot recently that Office 2003 does not export formatting information in its XML documents.
Truth: It does. I've been using it for weeks and while the XML is ugly (can you blame it?) it does contain all needed formatting information. Slashdot never even bothered to post a retraction.
Re:Clueless (Score:3, Interesting)
Debian is widely regraded as the most secure linux distribution.
Really windows can't compare with that. Not even close.
Re:Help me out here (Score:2)
Which tells us nothing. Is it because Red Hat ships their OS with more software than Microsoft? Is it because more people are looking for bugs? Is it because Microsoft has more bugs unreported? The quantity of bugs reports is a meaningless statistic, regardless of what any zealot might say. OTOH, the speed of bug fixes is far more useful. In that regard, Microsoft has improved quite a bit in recent years
More Anti-MS FUD! (Score:4, Funny)
Re:More Anti-MS FUD! (Score:2)
1)Write a bit of "malicious code"
2)Post it on a webpage
3)Make it as his homepage on
4)Wait
5)Profit???
Slashdot is widely read (how mnay millions was that) and is largely visited using IE (70% ?). Even if a small percentage is unpatched that is a lot.
hard disk could be formatted? ok (Score:2, Funny)
Dilemma. (Score:5, Funny)
* Let baddies in at their will.
* Run Windows Update, expose my machine to Msoft, sign away my soul through the patch EULA.
Help!
Download the patch without Windows Update (Score:2, Informative)
If you don't want to run Windows Update, or don't want to use Internet Explorer 5+ in order to use Windows Update, here is a list [microsoft.com] of recent security related patches that you can download individually.
Of course, you should realize that you have already signed your soul over to Microsoft by having Windows on your machine. You might as well close your eyes and agree to the EULA for Windows Update.
This just in... (Score:5, Funny)
"Lies all Lies! The infidel Linux computers are not secure. The coilation will fall in the wake of the mighty secure Microsoft operating system!"
More at 11.
Re:This just in... (Score:2)
I love that guy! He reminds me of a skin-and-bones girl with anorexia asking "....Am I fat...?"
Re:I dunno (Score:2)
RedHat's job is to issue advisories and patches when a hole is found. Furthermore, RedHat's up2date tool makes installing the patches _very_ simple (no reboots, BSODs or DLL errors).
Not overly suprising (Score:5, Insightful)
I think this is where the philosophical differences of Open Source Software really make a big difference. Even though OSS still has bugs, the live testing cycle is un-paralleled. However I think the biggest difference boils down to this: there is no one saying we have to have this product out the door by XX date. Rather it becomes stable when it's ready, but you can use the development version if you need or want.
As the lines of code in software grows and the complexity increases, I think we will see a greater number of more sever bugs in closed source systems. Ultimately I believe this will be one of the critical factors leading to OSS's long term success.
Applets, not apps. (Score:4, Informative)
I don't understand... (Score:3, Insightful)
I can understand the need for an array of software unavailable on any other platform (though, what percentage of that software is actually GOOD software?), and the platform standardization issues, maybe even "ease" of use, but honestly, the security and ridiculousness of the MS platform, ideology, and disregard of standards make me sick.
What is the continuing allure? Do you really not mind running machines that are completely insecure? And how can they not fix their own NT 4.0 code? That's absurd. They pitch this solution for years, and bail when the cost to fix their crap gets too high.
I'm not trolling, I'm baffled. Someone tell me why this continues?
Re:I don't understand... (Score:2)
Re:I don't understand... (Score:3, Interesting)
Well, let me explain my situation:
1) I have to use MS Windows at work. We use programs that only have windows versions, such as PLC programming applications. Plus, our customers and suppliers all have MS Word and MS Excel, and say as much as you want about OpenOffice.org being compatibl
Not enough hacking (Score:3, Insightful)
I work for a company that has a good bit of Microsoft, some Sun and some linux deployed. Now, without getting into any religious wars over who's more secure, I'll simply say that the Microsoft servers have been compromised on more than one occassion. The Microsoft servers also got hit very hard by Code Red
Re:I don't understand... (Score:2)
Ah, but the same can be said of many, many, many Windows exploits.
It's not like the majority of Windows bugs appear out of nowhere and are patched in the next week/month/year/whenever MS feels like it. They sit there, unnoticed, for however long, and then they're trumpeted about and patched.
So while you say, "The samba root exploit ... went a decade without being patched," that's only because it was only a week ago [slashdot.org] that it was discovered.
The difference is in how it was dealt with. Microsoft like t
Re:I don't understand... (Score:5, Insightful)
That's quite a statement.. ;)
I'm not saying I could do it... But there are some very clever people out there. One thing I've learned is never to underestimate others.
So, stick a file on your machine with your online bank details, and your credit card numbers, and post your IP address :)
Man, they're makin' it easy! (Score:3, Funny)
Well, that takes care of the wicked-long step 1 in uninstalling windows and installing linux! [linuxworld.com]
That is, of course, if this vulnerability affects the version I'm running - Windows Herpes Edition.
Let the Slashdot Madlibs Begin. . . (Score:5, Funny)
Let me save many of us some time:
"Well here we go again. A gaping security hole in Microsoft [ Operating System ]. This never would have happened if Bill Gates weren't just trying to make more money so he could buy more [ plural noun ] to fill up his mansion in [ place ]
This is just one more reason why [ circuit court ] should [ verb ] that [ expletive ] company once and for all.
[ Unix-based operating system ] only had this problem [ number ] in it's entire history, and there was a patch posted in under [ number ] minutes!
[ Text-based word processor ] rulez! Micr- [ Insulting variation on 'soft' ] is the [ Traditional evil diety ]!"
-----
i'm okay (Score:5, Funny)
Amazing (Score:2)
When will microsoft(and others) understand that browsers are http clients and not meant to be used as means of running arbitrary code on a client machine, however secure it might be . The least you can do is to tell the client that co
Yes but ... (Score:3, Insightful)
<reality check>
Until someone actually writes a massivily spreading virus/worm that jumps from Windows PC to Windows PC doing precisely that (formatting hard drives) - people are just going to patch it and not even think about changing OS.
Hell, most people probably won't even patch it. What doesn't affect them, they don't care about.
</reality check>
So Happy It's Thursday (Score:5, Funny)
Official
So
Happy
It's
Thursday for announcing MS holes.
This is great! (Score:2, Funny)
RH's Advisory of the Day? (Score:2)
Just an observation.
Hey... (Score:2)
Pre-emption (Score:2, Funny)
That's not a problem at all... (Score:2, Funny)
this is why i feel bad for all the poor L-Users (Score:2)
is it just me, or does surfing the web on a win box feel like living in some bad neighborhood just trying to avoid the next drive by shooting???
Sun is better (Score:2)
Why use this MS crap anyway?
If you have to deal with Windows, at least get a good java implementation.
Gee, that's funny (Score:2)
Again, compare Oranges to Oranges (Score:2, Insightful)
You're delluding yourself and you're not employing a correct analysis and comparison of the problems.
Re: (Score:2, Funny)
Why the sarcasm? (Score:2)
At least Microsoft is on a schedule. I never know when to expect a critical security patch for [favourite Linux distro name here]!
talking of airplanes (Score:2)
Let me tell you something so that you feel safer: rest assured that the safety-critical systems of airplanes don't run Microsoft Windows (neither do they run Linux).
Wasn't there a model of US warship that stalled because of Windows problems?
Why is anyone using MS' Java VM? (Score:3, Informative)
devil known (Score:3, Funny)
The Virtual Machine (VM) flaw is the most serious, meriting a "critical" rating from Microsoft.
This jumped off the page at me. Could someone explain the value of Microsoft's merits of their own flaws?
Re:How about "Weekly whingeand moan about MS" inst (Score:2)
Re:Formatting bad?? (Score:2)
Re:Repeat after me... (Score:2, Interesting)
Well, thats kind of a silly statement. I could say closed source has the potential to be more secure than open source and still be correct.
Heck in this case, MSFT found their own hole and patched it. If it was an OS JRE with this flaw, then chances are equally good it would be found by a "blackhat" first.