Microsoft Denies Windows 7 "Showstopper Bug"

Barence writes "Windows chief Steven Sinofsky has taken the unusual step of responding in the comments of a blog posting that claimed Windows 7 was suffering from a potential 'showstopper bug'. Stories had been sweeping the Internet that using the chkdsk.exe utility on a second hard disk would lead to a massive memory leak bringing the operating system to its knees in seconds. Responding to a blog post titled 'Critical Bug in Windows 7 RTM,' Sinofsky wrote: 'While we appreciate the drama of "critical bug" and then the pickup of "showstopper" that I've seen, we might take a step back and realize that this might not have that defcon level.' He signs off with the words: 'deep breath.'"

Comments

RAM optimization

[sopssa]

I wonder how this obviously one-sided summary even got posted -- it just sounds like a calling for bashing from people who dont read the article. Here's another snippet from Steven's response:

We had one beta report on the memory usage, but that was resolved by design since we actually did design it to use more memory. But the design was to use more memory on purpose to speed things up, but never unbounded â" we requset the available memory and operate within that leaving at least 50M of physical memory. Our assumption was that using /r means your disk is such that you would prefer to get the repair done and over with rather than keep working.

And it does make sense for two reasons:
1) Windows has to lock the drive anyways, so its better to get it done fast.
2) You CAN spend RAM. If the whole RAM isn't used, you're just wasting it. In this case chkdsk.exe will use dynamically what there is left, making the process faster. How is this a bad thing?

Rather than a bug or memory leak, this seems like an optimization.

Re:RAM optimization

[FatAlb3rt]

Optimizations aren't supposed to crash the computer.

The original report I read was full of drama, too much IMHO, and the bug could be fixed in the first service update.

Re:RAM optimization

[GreenEnvy22]

Agreed, this is a non issue, or at worst, a very tiny issue. For the very tiny amount of people out there that will run "Chkdsk -r" on a secondary partition, they may see almost all their ram used up while it is scanning the disk. If they have prexisting hardware or software glitches, it might blue screen on them. For the 90% of consumers who would never run chkdsk, and who don't have more then one parition, this is a complete non-issue.

Re:RAM optimization

[Nimey]

Haven't read many kdawson stories?

Re:RAM optimization

[sycotic]

Have a read of this: http://www.bluescreenofdeath.org/?p=94#comment-134

 

UPDATE:
After emailing back and forth with the VP Sinofsky, it was found that the chkdsk /r tool is not at fault here. It was simply a chipset controller issue. Please update you chipset drivers to the current driver from your motherboard manufacturer. I did mine, and this fixed the issue. Yes it still uses alot of physical memory, because your checking for physical damage, and errors on the Harddrive your testing. Iâ(TM)m currently completed the chkdsk scan with no BSODâ(TM)s or computer sluggishness. Feel free to do this and try it for yourselves. Again, there is no Bug.
Thanks all.

Re:RAM optimization

[Anonymous Coward]

2) You CAN spend RAM. If the whole RAM isn't used, you're just wasting it. In this case chkdsk.exe will use dynamically what there is left, making the process faster. How is this a bad thing?

This sounds a lot like the Outlook 2007 discussion on Vista (and some reports on XP). Vista has "advanced memory management" and Outlook "continually asks for RAM, as long as some is available". The result? Outlook allocates ~700M, according to the Task Manager process list, while the Physical Memory free (on a 3G system) reports 6% free. Closing Outlook brings the ram free percentage up to %60. Some MS MVP said just what you said "The RAM is available, so Outlook uses it and the program responds faster, that's a good thing", completely disregarding the fact that the computer is near unresponsive to everything else. A program should never take RAM "because it's available", it should take it "because it's needed". Using over 2G of RAM to open 3 emails is absurd, using 1G for texture and sound data is more reasonable.

Re:RAM optimization

[Your.Master]

Hey, but it's nothing like that. Using all of your RAM to check a disk for damage and repair it in response to a user's specific request is not like having Outlook open in the background.

Re:RAM optimization

[Anonymous Coward]

Excerpt from parent should be added as an update to the summary.

Re:RAM optimization

[Kamokazi]

I think it's time to dust off the 'kdawsonfud' tag!

Re:RAM optimization

[HermMunster]

I regularly put customer's hard drives into a different computer as a secondary drive and run chkdsk. Your math sort of makes it seem like 4-5% of a market isn't a lot to account for, yet that 4-5% means in terms of the OS market hundreds of millions of users. Should we let you take those support calls?

Re:RAM optimization

[rabbit994]

Actually isn't not unresponsive, Outlook will give up RAM quite happily and it's not opening 3 emails, it's keeping your entire PST/OST loaded into RAM so you don't bitch and complain when selecting different emails is "slow to load". Therefore if you have big .PST/OST expect for it to use alot of RAM.

However, I'm sitting on Windows 7 Ultimate x64 with 6GB of RAM and Outlook is using 200MB total including what's committed for use and what's it happily taking because it can. I have 457MB .OST (Exchange cached file) so wanting to load half of it's not unreasonable. Linux uses similar memory management system and I don't hear alot of complaining about it.

Re:RAM optimization

[Desler]

So the statement "there is no bug" is blatantly false.

The GP you are responding to never claimed there was no bug at all. What is being said is that the bug is in the chipset controller driver or somewhere else, not in chkdsk like this FUD submission is trying to claim. Maybe next time you should learn some reading comprehension.

Re:RAM optimization

[Anonymous Coward]

"isn't not unresponsive"

Congratulations! You have plumbed a new depth in illiteracy. That's a triple negative. Did you really mean responsive or unresponsive?

Re:RAM optimization

[Blakey Rat]

This is utter bullshit:

This sounds a lot like the Outlook 2007 discussion on Vista (and some reports on XP). Vista has "advanced memory management" and Outlook "continually asks for RAM, as long as some is available". The result? Outlook allocates ~700M, according to the Task Manager process list, while the Physical Memory free (on a 3G system) reports 6% free. Closing Outlook brings the ram free percentage up to %60. Some MS MVP said just what you said "The RAM is available, so Outlook uses it and the program responds faster, that's a good thing", completely disregarding the fact that the computer is near unresponsive to everything else. A program should never take RAM "because it's available", it should take it "because it's needed". Using over 2G of RAM to open 3 emails is absurd, using 1G for texture and sound data is more reasonable.

I'm running Outlook 2007 on Vista right now this instant, and it's using 92MB of RAM. Physical memory free = 57%.

Even if Outlook was using all but 6% of free RAM, why would that necessarily make your system "unresponsive to everything else?" 6% of RAM is plenty to keep your machine responsive, assuming it has a gig or more in it.

And this statement:

A program should never take RAM "because it's available", it should take it "because it's needed".

Is doubly-retarded. RAM takes time to fill, yet takes no time to empty. Therefore, all software should fill as much RAM as feasible to make itself more responsive to the user. RAM isn't some physical object you "take away" from something else-- if Outlook allocates RAM than another process needs, the OS just overwrites it as needed.

What about this one?

[DurendalMac]

It seems that if you install Windows 7 on the second hard drive, it will put it's system reserved boot partition on the first drive. This absolutely boggles my mind. Now I need both hard drives just to boot my system? I discovered this when Windows 7 fucked up my Chameleon installation. Then my Hackintosh wouldn't boot into OS X until I reinstalled Chameleon from the iAtkos disc. Then I had to unplug the OS X drive and reinstall Windows 7 so it would stick to it's own goddamned drive and leave the others alone.

Bad, BAD fucking move, Microsoft. Now Windows 7 can easily fuck up unrecognized partitions on other drives during installation. I really hope that gets fixed in the final version.

Re:What about this one?

[Anonymous Coward]

Except during the install it says: "the system reserved partition will be installed on the first boot device."

I remember wondering why it was 100MB myself when I saw that.

Re:What about this one?

[Joe U]

don't you think while using a Hackintosh, trying to dual boot a beta OS, and probably some other crap you didn't mention that you might run into a few problems? And yeah, I am sure your dual boot hackintosh is on the top of the list for a fix.

Hi.

On the top of your browser, there's an address bar, after the http:/// and before the next / does the word 'slashdot.org' appear?

I'm assuming yes, so seriously, what did you expect?

Re:What about this one?

[Hurricane78]

How is that new? It is that way since Windows XP, and likely even the whole NT and old Windows line before that!

I know because I have this setup with Gentoo and XP on this computer, and I think I can remember it from Windows 98 too.

Re:What about this one?

[Judinous]

Windows has never played nice with other operating systems one the same machine. The first rule of multiple-booting has always been "install Windows first".

Re:What about this one?

[xtravagan]

You have obviously not installed many OS yourself, and if you really believe what you are writing you should probably stop installing those you already are installing. You can control exactly where and how you want any partitions to be, so even with windows 7. It has a certain default, which is to install a 100MB, let's call it, rescue partition.

Just pre partition the disk the way you want it and you won't have that extra partition. So perhaps the bad move is on your for not knowing what you are doing and still posting as if you did.

Re:What about this one?

[shutdown -p now]

Windows has never played nice with other operating systems one the same machine. The first rule of multiple-booting has always been "install Windows first".

Well, at least it no longer overwrites GRUB when installing (or at least Win7 RC didn't do that) - while XP always did.

Re:What about this one?

[lawnboy5-O]

Its NOT a Bug.... its a FEATURE! It's from M$ correct? That's the way its supposed to be...

Re:What about this one?

[DurendalMac]

No dipshit, Hackintoshing has very little to do with it. As far as Windows 7 is concerned, it was simply another drive. That's all. The point of the matter is that it fucked up a partition that it didn't properly recognize. The same thing could happen to Linux installations as well. It's an ugly oversight that is NOT specific to Hackintoshes, so pull your head out of your ass.

Re:What about this one?

[DurendalMac]

My point is that the user shouldn't have to bloody worry about it. Why should I have to prepartition my drive just to keep Windows from messing with other drives? It should stick to the installation drive by default, not require extra steps to keep it from messing with other drives in the system. Plopping the 100MB system reserved partition on another drive by default means I need BOTH drives to boot, which is stupid. But yes, my bad for assuming Microsoft would do things in a logical fashion.

Re:What about this one?

[just_another_sean]

You're missing the point. Even if you pre-partition the second drive Windows still installs it's boot loader on the first. This is not just true of 7, it's been doing this since NT 4.

Brushing aside your "you should just know how to it" bs ( I thought stuff "just works" in Windows, it's teh easy!) it goes beyond understanding the partitioning. It's about behaving in a counter intuitive way that requires discovery on the user's part. I can
naturally assume that I'll be better off partitioning my own drive. It takes a real WTF moment to realize you have to rip out one of your drives before you install Windows if you don't want the unexpected behavior of your master boot record being on a different drive then the OS. Another poster said "install Windows first, that's the rule". Fine, I get that but it's still f'ng stupid.

Re:What about this one?

[dna_(c)(tm)(r)]

Anyways you could always copy the files and boot sector from the small partition to the Windows 7 one and raze the small one, then you just need to edit the BCD registry using EasyBCD or bootedit.exe to point to the correct partition on boot. But yeah those are both WINDOWS tools... but bootedit.exe should be available from Windows 7 Setup on the DVD if you mess up and can't boot into Windows (press SHIFT+F10), and fixboot.exe can install the boot sector onto any partition.

And then they say Linux is difficult?

Re:What about this one?

[Zancarius]

As far as Windows 7 is concerned, it was simply another drive.

You raise really good points. There is one unfortunate thing in the behavior of Windows' installation process when it comes to drives it doesn't understand. But first, an example:

Let's assume that you're installing Windows on a system with two hard disks. On the first disk exists Linux or BSD. You plan on using the bootloader to boot Windows (off the second disk). The second disk is blank. When you attempt to install Windows to the second disk, it will alert you that it needs to make changes (i.e. wipe the bootloader) on the first disk. It's possible a situation like that might result in unexpected changes, but it's not difficult to resolve--simply load a live CD and replace the bootloader. (At least, this prompt would occur with Windows XP--I have no idea with Windows 7 because of a habit I've acquired. Keep reading.)

However, I've never actually had windows make any unexpected alterations to anything other than the disk I was installing to. Perhaps it's partially thanks to a healthy dose of paranoia; whenever I install Windows to a dedicated disk--really, whenever I install any OS, I have a habit of unplugging all the drives I don't want to touch. As you alluded to, since the OP clearly didn't take such precautions, he sort of got what was coming to him.

Maybe my measures are a little excessive, but when I'm dealing with the prospect of having to reinstall several OSes just because of a stupid late-night mistake, a typo, or maybe a software bug, I'd rather take the time to make sure it can't happen. Not that this method isn't fraught with complications--it's possible to unplug the wrong drive. But, that's why you check it first to make sure it is the one you want to wipe!

So yes, you're exactly right. The OP really should have taken greater precautions with his data. It would've saved him a reinstall.

Re:What about this one?

[thetoadwarrior]

Well, at least it no longer overwrites GRUB when installing (or at least Win7 RC didn't do that) - while XP always did.

That's a bug and will be fixed in Win 7's first service pack.

Re:What about this one?

[Hatta]

Sure they do. Linux is the only operating system on my home computer. Someday a game may come out that I think is worthwhile, I'm open to installing Windows 7 then. If I can't do that without damaging my Linux install, they don't get my money.

Re:7 Bashing

[godrik]

# dd if=/dev/zero of=/dev/windowspartition

my pleasure.

Nonissue

[Seth Kriticos]

If it is really such a serious bug, than it will be fixed with the first installation and following windows update. (or OEM patches).

No sane person runs a vanilla installation of windows.

Actually, in the first months when win 7 gets released, a lot of even more serious bugs will surface (because of the wide exposure). They also will be fixed and integrated in the update service. It's known that the first months of release is always the release test and fix cycle.

This is just how things go.

Disclaimer: I don't like windows, this is just an objective view.

Payback is a...

[gklinger]

Dear Microsoft,

It sucks when people spread FUD, doesn't it?

Re:Proper facts please

[Desler]

Because kdawson is slashdot's resident FUD artist. If he actually posted factual stories that didn't contain overblown anti-Microsoft FUD he'd be fired.

Re:Proper facts please

[DaveV1.0]

Nah, he wouldn't be fired.

He would just choke on his own vomit.

The real issue: Some people are running scared

[benjymouse]

First a few facts

1. chkdsk.exe is a disk checking and file-system repair tool. Most users will never know about it.
2. The chkdsk functionality kan be invoked through Windows Explorer as well. Some users will find this tool if they deliberately are looking for it.
3. chkdsk.exe with the /r oprion (and *only* with the /r option) has been designed to allocate most of the available physical memory, but always leave at least 50M free. This is not a memory leak. It was a deliberate decision because using more memory will dramatically speed up the surface verification/repair process. Note, it will allocate from available memory, i.e. already allocated memory will not be forced out into paged/virtual memory. If this was a leak the allocation would go on and on, cause more and more swapping until the system trashed itself to death. But it's not. The system remains responsive and the memory is freed when chkdsk ends.
4. The crash condition appears to be an unrelated issue with chipset controller drivers. Propably this issue becomes more pronounced during periods with intensive disk usage and/or low memory conditions. It is not caused by chkdsk, it is a driver/controller issue which has been reported to be fixed by updating drivers to the latest version.

No, the real issue is that Microsoft appears to be slated for a massive success with Windows 7. At this point some Microsoft detractors will leap upon any issue in an attempt to spoil the party. In this category you find Randal C. Kennedy of InfoWorld who leapt on to this issue with blatant disregard for any facts. Even if the original blogger and mr. Kennedy were so stupid as to believe this issue was a memory leak and that it caused the crash, by their own account it would only manifest itself under very specific circumstances:

• chkdsk.exe must be invoked with the /R option to perform a surface scan/repair (this is the most radical option).
• chkdsk.exe must be invoked for a non-system partition (chkdsk must dismount the drive/partition - using /r on the system drive requires chkdsk to run during boot instead).

So, even if this was a bug, only users with

• 2 or more drives/partitions,
• one non-system exhibiting suspicious behavior to warrant a "surface scan".
• users able to find and launch the tool

No, this whole bruhaha has a distinct smell of desperation about it. And kdawson is - as usual - all to happy to assist.