Windows SSL Interception Gone Wild
facebook.com
facebook.com
Shouldn't the possiblity have been forseen and addressed beforehand?
Perhaps by...
(1) Anti-virus / anti-malware makers. Does this software not notify the user when strange CA certs are put into a system's root certificate storage? I understand that certain businesses do this for traffic monitoring... so it might be legit... but still, no user notification?
(2) Microsoft. Do their license terms really allow OEMs to install MiTM proxies and screw around with the root certs? Microsoft could do a good thing here by disallowing this sort of malfeasance... or is there some problem I'm not seeing with such an action?
If this were done in, say, OS X (unrealistic, of course), it would be found out and the whole tech world would know about it in a jiffy. John Siracusa would be howling at the Internet moon within a couple of hours...
It was installed by the OEM. Doesn't really help if it only notifies the OEM.
> (2) Microsoft. Do their license terms really allow OEMs to install MiTM proxies and screw around with the root certs? Microsoft could do a good thing here by disallowing this sort of malfeasance... or is there some problem I'm not seeing with such an action?
The general solution to what you're talking about is to prohibit the OEMs from installing anything by default. The problem is the OEMs wouldn't like it and Microsoft has to keep the OEMs happy lest they get any bright ideas about offering their computers with Ubuntu for $50 less than Windows.
Otherwise, some OEMs have tried installing versions of Linux, with negative financial results. A few are still trying. The real problems are selling and supporting them.
Which isn't much of a surprise considering what I have observed so far (in trying to purchase a Linux PC). I can't recall ever having seen an OEM offer Linux for more than a sparse subset of their product line, usually mid-tier or low-tier machines.
>A few are still trying.
Which ones? The situation may have changed since I last paid any attention a few years ago.
>The real problems are selling and supporting them.
The MVP here is to merely accept returns for units that turned out to be particularly troublesome; which they usually do (ie: the Samsung UEFI thing from 2013).
A non-Microsoft UEFI key thing might be nice as well, but that's another story.
If you think there's a market for Linux PCs, you can always set up a company to sell them. You wouldn't be the first to try, but you might be the first to succeed ;-)
Dell used to. I just contacted Dell sales and according to "Hazel" they do not offer any non-Windows OS for consumer products nor will they sell a system sans-OS.
>But the real problem is that one "support incident" eats the profit from about five sales, or more.
Meh, there is a lot of room for argument here. I think the real problem, after MS' many anti-competitive shenanigans is that most people just think MS Windows is synonymous with "computer". Those who really want a Linux PC will just buy the hardware they want and install it themselves.
>If you think there's a market for Linux PCs, you can always set up a company to sell them. You wouldn't be the first to try, but you might be the first to succeed ;-)
Someone someday will probably succeed at that. I'm probably not that someone, and that day may not be today. I do think that there is a small market for it, and there could be a bigger one, maybe if/after Gaben has any success with SteamOS. OTOH, if we ever have a modular laptop standard with a commodity peripheral market then maybe not, as there would be less need. (given that the only OEM pc's I have purchased in the last 10 years were laptops).
They do, it's called "Project Sputnik". It's targeted at developers though, which is a market that clearly makes sense, as AnthonyMouse pointed out.
http://www.dell.com/learn/us/en/555/campaigns/xps-linux-lapt...
The XPS 13 review yesterday was interesting, but I think I need a more beefy machine. Anyone has experience with this precision developer edition on Linux?
For a company specialized in Linux PCs, there is System76.
These days I have a purchasing department that impedes my purchases, so as long as the crap they give me isn't too bad I just let it be.
Not really. I got my info from senior managers at some of the (very large, Taiwanese) companies concerned.
> I think the real problem, after MS' many anti-competitive shenanigans is that most people just think MS Windows is synonymous with "computer".
Microsoft has never been accused or taken to court for any "anti-competitive shenanigans" re the success of Windows, only for ways it tried to exploit that success.
> Those who really want a Linux PC will just buy the hardware they want and install it themselves.
Yes, exactly. And they will install whichever of the 157 versions they prefer. These are among the reasons why it's hard to make a profit selling Linux PCs.
Oh. Why didn't you tell me that you had a real authoritative source? /s
>Microsoft has never been accused or taken to court
Microsoft's historical business practices WRT both Apple and Linux are well documented. Lawyers have made whole careers work generated. There is no need to hash this out for the billion +1th time.
>These are among the reasons why it's h...
I don't know who you think you're arguing with. I haven't asked you for seed money. I've not tried to convince you to go into business selling PC's.
Sorry, reality intruded. I should have known you'd find that a problem. However, you could get a clue from the fact that most companies who have tried to sell Linux have either stopped or gone bust, or do it on a very small number of systems. This is not because they are against making a profit.
> Microsoft's historical business practices WRT both Apple and Linux are well documented.
Up to a point. But most of the inexpert comment I see is badly informed and usually wrong. Still, who reads documentation?
> I don't know who you think you're arguing with
My mistake, I didn't know I was arguing....
The problem seems to be that they're always trying to put them on budget machines, which is completely the wrong market. It's chasing the customers who pinch the last penny and you're never going to make any money from them regardless. Meanwhile those customers don't know what an "Ubuntu" is but pick it because it's cheaper, and then you get overrun with support calls when they want to install Turbo Tax.
The place where it makes much more sense is the corporate and professional markets where the customers actually know what they're buying. An IT department which is just going to nuke whatever the OEM installs in favor of their own volume licensed disk image would be happy to save the cost of a [redundant] Windows license for every machine. And professionals like programmers and scientists who actually use Linux would appreciate being able to buy workstation-class hardware with official driver support.
The main attempt to sell Linux to end users was the use of different versions on netbooks, which were mainly bought on price by relatively clueless users.
I talked to one supplier about the obvious cost-of-Linux-support problem at their launch. We won't do support, they said, it will be like an appliance: we'll just reset to factory condition.
You can imagine how that turned out...
I got my new Lenovo Y50, visited my own website with it and decided to see how my https cert looked. I got quite scared when I saw I was being MITMed but I googled it and there were already a ton of forum posts saying it's just stuff bundled with Lenovo. So I uninstalled it.
Complete instructions here: http://www.pcworld.com/article/2886278/how-to-remove-the-dan...
Perhaps because it was persistent and on the TCP stack level the phonehomes never succeeded? The retry logic should be robust enough to try to deliver the fraud list anyway, even if it will only accept that it has been delivered after a secured connection is restored.
[1] http://en.wikipedia.org/wiki/DigiNotar#Issuance_of_fraudulen...
Maybe this is a practice that needs to stop. Malware scanners can scan on the local machine after the browser has decrypted the communication and web filtering, I think, is nothing but a sign of mistrust against the users.
It's really kind of a giant security vulnerability. If an attacker can compromise the machine doing the MITM on all the encrypted connections then they get every password and credit card number for every user in your company for every website.
What if it's the user who wants this filtering?
I run a local proxy that MITMs to filter out ads, tracking scripts, and other undesirable things. It works in all the browsers I use regularly, and any browsers that happen to be embedded in apps, because this way the stuff I want filtered out never even reaches the browser.
Usually Chrome is eager to show security-related notifications but for this there isn't even a yellow notification bar with "OK, got it" option.
Google wants to further promote it's closed Chrome ecosystem, and to do that it needs to gain corporate support, for among other things, its Chromebooks and ChromeOS platform.
And it's obviously more important to appease corporate IT than to protect users security.
Built in Google-spying and now, support for corporate spying too? I wouldn't trust a Chromebook as far as I can throw it.
Google codes Chrome in order to make it more useful for various kinds of customers, such as customers who have virus scanners.
And this becomes "Google putting its own interests ahead of its users"?
Back here in reality, that's called the customer is always right and is a fundamental tenet of business.
Banks... imagine the irony.
Which obviously didn't work here, as Chrome was one of the most affected targets.
Firefox on the other hand, was more or less absent altogether. I know which browser I will trust.
Thus Firefox is the more secure browser.
see those for example: https://forums.lenovo.com/t5/Lenovo-P-Y-and-Z-series/Persona... http://www.thestudentroom.co.uk/showthread.php?t=3013039 https://forums.lenovo.com/t5/Lenovo-P-Y-and-Z-series/Lenovo-...
Besides cleaning your box, you can blame Lenovo, stop buying their products, promote the boycott, etc. All things that regular people can do and serves as an anger/stress/steam release valve.
The NSA news, even though it is/should be a much more important or pressing issue, it's something you "can't do anything about". I mean, ostensibly you can do a lot as a citizen, however most of those actions have long term effects and thus are not as useful as a release valve. It involves commitment and even sacrifice, whereas blaming a corporation (however righ you might be) is much more immediate and serves the purpose of having someone to blame for that and lots of other stuff, i.e. you can then blame the general state of IT security, then how the govt does nothing about it, how privay is nowadays non-existent, think of the children, etc.
I also believe another factor is the way news have found a way to tap into this need for the audience to have a release valve. Something or someone to be angry at and so all your problems can be channeled to that. Where I live I've seen a growing amount of newspapers and news media that just basically do a certain journalism that does not bring anything to the table but things to be raging about.
I guess it's easier to sell stuff when you can easily get people "on your side", and since there's always a lot of people angry at something, it becomes easy to have an audience.
So what's the point then (from the POV of the media) of bringing "important" (for different values of important) news to the front page when that would require their audience to commit to actions that would last several years (change your country's politics for example) and thus not as easily enticed to "get on your side" (and thus buy your media), if on the other hand you could bring, I guess you could call them "anger-bait" (like click-bait) news, and have everyone talk about it by virtue of functioning as an escape valve where people relieve their stress, fear, anger, etc?
I'm not saying it's a good thing, but I've seen more and more evidence that points in this direction, and I guess that would be my answer as to why one has much more attention than the other.
Edit:
As an analogy, I read somewhere about the recent Charlie Hebdo (sp?) attack and how it got disproportionate attention vs the two thousand killed by ISIS (I believe it was ISIS... or Borok Haram?). Maybe it's a similar thing. You believe you are able to do "more" when it's close to home (Western nation) vs far (somewhere in Africa, far away from me).
If it wasn't for the SIM story, I'd have missed the Five Eyes legal restraints dodge:
https://plus.google.com/104092656004159577193/posts/2ncBEdPV...
Your Wikipedia article link doesn't directly address this. It points to several other documents though:
A 2000 ZDNet article by Duncan Campbell:
http://www.zdnet.com/article/echelon-world-under-watch-an-in...
"Under a secret agreement signed in 1947, called UKUSA, the English-speaking countries agreed to share responsibility for overseeing surveillance in different parts of the world."
That doesn't tell much. But this does:
"On 6 September 1960, two NSA defectors held a press conference and revealed the worldwide scope of NSA's activities:"
"'We know from working at NSA [that] the United States reads the secret communications of more than forty nations, including its own allies... Both enciphered and plain text communications are monitored from almost every nation in the world, including the nations on whose soil the intercept bases are located.'"
It also discusses the Church Commission hearings (1975).
I'm not sure how I'd classify this, but I see general awareness as being vastly greater. And as someone who's been paying attention to this story for a long time (15+ years), it's news to me.
http://www.nytimes.com/library/tech/99/05/cyber/articles/27n...
I guess widespread speculation that avoiding domestic surveillance laws is one of the things done with the system isn't the same as knowing that it is going on, but my point was that the widespread speculation had proceeded Snowden by quite some time.
From your NY Times article (published May 27, 1999):
Until last Sunday, no government or intelligence agency from the member states had openly admitted to the existence of the UKUSA Agreement or Echelon.
The mutual surveillance / legal evasion possibility appears to be suspected but not demonstrated. Again as with much else, what Snowden's done is to specifically document such activity. Which is of and by itself a material distinction.
European Parliament officials have also expressed concern about the use of Echelon to gather economic intelligence for participating nations.
And:
While few dispute the necessity of a system like Echelon to apprehend foreign spies, drug traffickers and terrorists, many are concerned that the system COULD be abused to collect economic and political information.
(All-caps emphasis added -- minimal HN formatting options have their drawbacks.)
So, I'll maintain that the documentation of such abuse is a New Thing.
The SIM heist confirms that few entities have capabilities that almost everyone assumed they have.
Superfish enable anyone to attack significant percent of internet users.
- it's an immediate and very serious threat to a lot of people (every script kiddie with room-temperature IQ level can use it to clear someone's bank account)
- it's a very clear example of how customers are literally being fucked over by businesses, and how a big and trusted company turned out to be represented by flat-out lying assholes (one rarely gets to see a case without any room for doubt)
- it's a case that you can (and should) do something about
But I don't think you need a CA at all since plugins can see the full DOM (whether SSL or not). Like if you "inspect element", view source, or run firebug.
The plugin is already written too: https://addons.mozilla.org/en-US/firefox/addon/windowshopper...
http://www.similarproducts.net/
> SimilarProducts is a monetization platform that uses Superfish technology to help users find and discover products visually. The technology instantly analyzes images on the web and presents identical and similar product offers.
They did this for years, actually. They paid add-on developers to bundle their shopping app with the developer's app. I remember this going on ~2010/2011 at least.
People were not happy about it to say the least.
https://www.crunchbase.com/organization/superfish/investors
I'd love to see people put money where their mouth is and refuse to be funded by those investors... but I'm pretty sure it's not going to happen.
https://twitter.com/ow/status/568935755344580608
Superfish: Go shame yourself. If I was an investor in your company, I'd pull my money now.
(another reason to put Flash behind click-to-play and/or push for HTML5 video)
[I work at FB, but not on sounds or directly on https man-in-the-middle detection.]
To quote a Chrome developer: "Click to play is not actually a security boundary. In particular, it has always been subject to click-jacking."
Click-to-play prevents Firefox from running any plugin code without explicit user action. I am 99% certain this is also the case for Chromium-based browsers. Source: I am a Firefox developer and I have worked on the click-to-play code, e.g. http://bugzil.la/899347
Wrong: https://code.google.com/p/chromium/issues/detail?id=174963
If you go through the Chrome bug tracker, you can find several instances where Chrome engineers point out that Click-to-Play is not meant to be a security feature, and that the "Block all" setting is what is actually secure. There are several bugs which demonstrate ways around Click-to-Play which are closed as "WontFix". A quick search yields the following quotes from Chrome engineers:
"Yes, this is why click-to-play is designed as a convenience and not a security feature. If you want plugins blocked in a way that cannot be click-jacked, use "Block all," which requires a protected browser interaction (context menu, page action, etc)." [0]
"The "Click to play" setting is not a security measure. If you want to securely block plugins you must use the "Block all" option, which is a bit less convenient than "Click to play," but provides a click-jack resistant, browser mediated interface." [1]
"I'm kicking this out of the security queue because it isn't a security mechanism ... The secure method of blocking plugins is to select "Block all" and right-click to run. Whereas the "Click to play" feature is for convenience and performance." [2]
"It's not a security feature..." [3]
[0]: https://code.google.com/p/chromium/issues/detail?id=176724
[1]: https://code.google.com/p/chromium/issues/detail?id=225636
[2]: https://code.google.com/p/chromium/issues/detail?id=160707
[3]: https://code.google.com/p/chromium/issues/detail?id=414232
I'm sure there are other instances where they talk about it more, these are just the first results I found.
This can be done without any proxy or certificate installation.
And hijacking Google search on Firefox, http://blog.linuxmint.com/?p=142
This is the second article I've read that states this - Superfish does no such thing.
https://web.archive.org/web/20150220003144/http://www.komodi...
installed as a LSP:
http://en.wikipedia.org/wiki/Layered_Service_Provider
"modify the windows networking stack" is not an absurd description of that.
https://stackoverflow.com/questions/16269624/the-truth-behin...
Never mind that Facebook sees all the computer user's Facebook traffic, and cross-indexes it with every other bit of data gleaned from their vast graph and uses it for profit.
What of sites that unilaterally change rules retroactively? Or fail to provide reasonable alternatives?
Facebook does all of the above.
To an extent that I don't trust it, and don't use it.
But there are plenty of other services which wave the "but you consented!" flag. Google comes to mind, and I've had my set of issues with them as well.
I mean, of course elements that are embedded in Web pages may be traced by the party who sends them (and pages very often include off-site content).
My fear is that these companies will use this Superfish debacle to attack and restrict the ability for users to download legitimate software which leverages these technologies. As users and developers, we want to retain this ability.
Adware sucks, and there are dozens of anti-virus companies who should be all over anyone who tries to pull this crap. The problem here is not with MITM, SSL packet inspection or modification. The problem here is that Lenovo allowed themselves to be turned into a distribution channel for a poorly implemented, spammy piece of adware for a few extra pennies.
I presume 'nugget is talking about the HTML rewriting aspect of the software. Injecting additional/unwanted tracking code == bad, user-requested re-writing of content == good.
Very annoying if you wish to use your own CA or add another and it is also dangerous in that it masks any cert installation by malware.
Your own site, work, or vendor / client sites could be added.
Or you could want to remove a Comodo (or Honest Achmed's Used Cars and Certificates).
http://www.livehacking.com/2011/04/25/honest-achmeds-used-ca...
https://bugzilla.mozilla.org/show_bug.cgi?id=647959
Just because your OS / browser vendor "trusts" a cert doesn't mean you should.
In other words, users should always have the right to control who they (indirectly) trust. That's what the comment above is referring to - it will be even worse if Superfish is used as an excuse to take away this right.
Also if you want to use http://www.cacert.org/ you need to add their cert.
Userscripts and userstyles are very popular, and I see no particularly large backlash against them.
They already have, with HTTP/2. Encryption is mandated for HTTP/2 so something like Privoxy (or even just a caching proxy) has to use a Superfish-like method to bypass the encryption. The only alternative is to modify the browser, which they are also locking down with unchangeable ChromeOS and limiting plugins to only officially sanctioned ones.
...and you won't really even be able to just not use HTTP/2 because the web will be much slower as pipelining is not even implemented in Chrome, and Firefox will no doubt drop it soon. Websites optimized for HTTP/2 could take minutes to load without pipelining.
The real irony is that neither Google nor Mozilla determined what software caused pipelining problems, so guess what, it was Superfish and its like. Instead they made a new protocol that requires Superfish-like MITM interception, to work around problems caused by Superfish-like MITM malware.