Live data from Hacker News

Windows 10 in-place upgrades are a severe security risk

blog.win-fu.com

111–120 of 124 posts

Re: Windows 10 in-place upgrades are a severe security risk

#112
post #36

I don't know whether this works in newer versions of Windows, but it was extremely simple to elevate your priveleges on almost any Windows 7 machine. I've done this dozens of times. I haven't used Windows for years now, so the details are a bit fuzzy, but it essentially worked like this: Start the machine. During boot(when you see the orb splashscreen), turn off power or hold down the power button for a few seconds.…

Oh, so essentially the same bug that existed since windows 98? Where you could, on the login screen, click the little question mark, which would open windows help, then you could click on "open file", navigate to C:\windows and just double click on explorer.exe, which would log you in without a password?

Windows 98 wasn't a true multi-user operating system anyway, security was a simulation. Only the NT line was multiuser at the time (and later XP through 10).

The Windows 98 issue was a bug. The example given involving Windows 7 and renaming executables is NOT a bug. If you give someone unrestricted access to the hardware, they have unrestricted access to the hardware. Working as intended.

You want someone not to be able to mess with a Windows installation? Activate Bitlocker.

That's why this Windows 10 issue IS a bug. Because it bypasses Bitlocker and allows a normal user to escalate to local admin. The Windows 7 issue is NOT a bug because it allows no such escalation (since no security was ever stopping local HDD access anyway).

Re: Windows 10 in-place upgrades are a severe security risk

#113
post #102

Earlier quoted context omitted.

Most professionals?

Why not laptops with docking stations? That's what I see more and more recently.

That's the same thing. There is no OS I'm aware of that has separate versions for "laptops" and "desktops"; a laptop IS a "desktop" these days, as opposed to a "mobile device" running a mobile OS (Android/iOS).

Re: Windows 10 in-place upgrades are a severe security risk

#115

Earlier quoted context omitted.

Not if you're still on Win7, like most corporations still mostly are: https://github.com/carmaa/inception/blob/master/README.md

This requires FireWire or thunderbolt, which is relatively uncommon on Windows machines.

Au contraire, mini-Firewire has been quite common on business laptops, which are the most common use case for BitLocker.

Re: Windows 10 in-place upgrades are a severe security risk

#116

Earlier quoted context omitted.

Why not laptops with docking stations? That's what I see more and more recently.

That's the same thing. There is no OS I'm aware of that has separate versions for "laptops" and "desktops"; a laptop IS a "desktop" these days, as opposed to a "mobile device" running a mobile OS (Android/iOS).

13 year old kids around here all call desktops laptops anyway

Re: Windows 10 in-place upgrades are a severe security risk

#117
post #42
post #29

Earlier quoted context omitted.

That's exactly how responsible disclosure works. You wait until after the patch, then you do the blog post. In that order. Publishing early just damages your relationship with the company, the community, and makes it more well known that you _don't_ have good intentions.

In this case the next time the vulnerability will be available is with the release of the next upgrade, expected around March.

Insiders see this style of Upgrade on a regular basis (with each new major Insider Build). Microsoft just made a big blog post about a new system for this style of Upgrade (the "Universal Patch Platform") and has asked Insiders to keep an eye out on it. A White Hat attempting responsible disclosure could at least check on Insider Builds and attempt to provide feedback on the new platform through official channels.

Re: Windows 10 in-place upgrades are a severe security risk

#118

Earlier quoted context omitted.

As compared to Windows in which an app can't do that? Windows provides UI isolation between privilege levels: https://msdn.microsoft.com/en-us/library/bb625963.aspx As compared to Windows or Mac where apps are sandboxed by default? As the sibling commenter mentions. Mac App Store applications are required to be sandboxed: https://developer.apple.com/library/content/documentation/Se... The same applies for Windows Sto…

Sorry, but how many people actually use the Mac and Windows app stores?

I think that you are underestimating the use of the Mac App store. Many applications are only available in the app store, just to give some examples of popular Mac Apps: Pixelmator, Affinity Designer, OneDrive, Pages, Numbers, Keynote, iMovie, Garage Band, Tweetbot, and Airmail. Then there are many Apps that can be purchased both through the Mac App Store or from the vendor, such as Omni{Graffle,Focus,Outliner}, Fantastical, 1Password, Photoshop Elements.

The majority of apps that I install come from the App Store.

Moreover, some vendors also sandbox non-app store apps. E.g. Chrome tabs, Photos, or Safari tabs.

Re: Windows 10 in-place upgrades are a severe security risk

#119
post #58
post #2

Sounds like a case of 'already behind the airtight hatch'. If you have administrative privileges to install an OS upgrade then you have administrative privileges to disable filesystem encryption. On the other hand, if MS pushes the update to the PC and it self-launches or can be initiated by a non-administrator, then it seems like there is a real security problem here.

It's not earth shattering - somebody could steal a encrypted laptop that's already running and keep it running until an upgrade comes and then bypass bitlocker. Aren't there other ways of bypassing it with a running machine?

> Aren't there other ways of bypassing it with a running machine?

The key element here is that you don't need to be a local admin, just a regular user who has restart permissions.

Re: Windows 10 in-place upgrades are a severe security risk

#120
post #91

Earlier quoted context omitted.

The parent comment is (slightly obtusely) talking about "UWP Apps", which are installed from the store and have mobile-style "sandboxing". Normal Win32 applications can of course use SetWindowsHookEx() and the DirectDraw screen functions to take over your screen.

> The parent comment is (slightly obtusely) talking about "UWP Apps", which are installed from the store and have mobile-style "sandboxing". And if he is, that's an absolutely academical points as 99% of all Windows-users runs zero applications from the store. Nothing like an imaginary straw-man argument on the internet, eh? :)

It's cute that you're accusing someone of an "imaginary straw-man" as you then proceed to pull crap like "99% of all Windows-users do [...]" out of your ass. Source? Or is it okay for you to make stuff up as long as nobody else does? ;)
Post reply on HN