Slashdot Mirror
Researcher Finds Another Security Flaw In Intel Management Firmware (arstechnica.com)
An anonymous reader quotes a report from Ars Technica: Meltdown and Spectre are not the only security problems Intel is facing these days. Today, researchers at F-Secure have revealed another weakness in Intel's management firmware that could allow an attacker with brief physical access to PCs to gain persistent remote access to the system, thanks to weak security in Intel's Active Management Technology (AMT) firmware -- remote "out of band" device management technology installed on 100 million systems over the last decade, according to Intel. [T]he latest vulnerability -- discovered in July of 2017 by F-Secure security consultant Harry Sintonen and revealed by the company today in a blog post -- is more of a feature than a bug. Notebook and desktop PCs with Intel AMT can be compromised in moments by someone with physical access to the computer -- even bypassing BIOS passwords, Trusted Platform Module personal identification numbers, and Bitlocker disk encryption passwords -- by rebooting the computer, entering its BIOS boot menu, and selecting configuration for Intel's Management Engine BIOS Extension (MEBx).
If MEBx hasn't been configured by the user or by their organization's IT department, the attacker can log into the configuration settings using Intel's default password of "admin." The attacker can then change the password, enable remote access, and set the firmware to not give the computer's user an "opt-in" message at boot time. "Now the attacker can gain access to the system remotely," F-Secure's release noted, "as long as they're able to insert themselves onto the same network segment with the victim (enabling wireless access requires a few extra steps)."
If MEBx hasn't been configured by the user or by their organization's IT department, the attacker can log into the configuration settings using Intel's default password of "admin." The attacker can then change the password, enable remote access, and set the firmware to not give the computer's user an "opt-in" message at boot time. "Now the attacker can gain access to the system remotely," F-Secure's release noted, "as long as they're able to insert themselves onto the same network segment with the victim (enabling wireless access requires a few extra steps)."
Totally different things. I imagine they find software and firmware vulnerabilities all the time. Hardware is difficult to patch around, and obviously comes with the noteable performance hit.
I hope heads will roll...but like always...never happens! I could even imagine an increased sales for the new generation of Intel processor with the meltdown flaw fixed. smh
If you have physical access you can do anything...
If MEBx hasn't been configured by the user or by their organization's IT department, the attacker can log into the configuration settings using Intel's default password of "admin." The attacker can then change the password
So, the "flaw" is that the user forgot to set the lock? I am stunned that this is considered a vulnerability/flaw. I mean, when I buy a new gun safe or document safe for my home or office, it comes from the factory with a default combination. I have to set it to one of my choosing. If I choose to not change the default combination, then that is on me.
Now, you might argue that it should be more like keyless entry for an automobile: the manufacturer sets a code a and provides you a device (key fob) for entry. However, if Intel did that, they would be accused of making their products difficult to use or crippling them (because people would certainly lose their AMT key fobs and Intel would either be unable to recover them, or would charge a fee for the service) or taking advantage of the user (because they would certainly lose the key fob). Plus, that would make it an absolute nightmare for central IT, the target audience for this particular feature.
The point is that if you are buying machines that have this capability, then you are buying mid-range to high-end business/professional stuff. AMT is not available on entry-level and most consumer gear. Besides, the people who don't bother setting the MEBx password on their systems (assuming they don't have central management through IT) are probably the same sort of people who buy a wireless AP, turn it on and leave the password set to the default and the admin function accessible over the wireless interface.
Intel has problems, but this one is definitely way down on the list.
Shouldn't AMD benefit from this long term?
I've worked in the IT field for 15 years - in academia, for financial institutions, for Fortune 500 companies, and at small, locally owned businesses.
You would balk if you saw how many of the "top companies in America" don't give 2 shits about security, outside of whatever the latest CNN scare story is. I personally find it amazing how some of these corporations will spend tens of thousands of dollars on fancy security equipment.... that they never bother to actually configure.
You can show your C-levels the lock and hand them the key, but you can't make them set the latch.
An enigma, wrapped in a riddle, shrouded in bacon and cheese
Millions of devices ship with default passwords. It is an issue only if it is not possible to change it, and the need to change is not clearly explained when it was shipped. Ideally it should not be the same password for all devices but something unique to each chip, given to the manufacturer as part of shipped chips.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
Over the last few months, we've come to know that:
1) For years Intel has put a backdoor in hundreds of millions of CPUs, called "Management Engine"
2) That the aforementioned backdoor, besides being evil in itself, is also full of bugs
3) That their CPUs, because of some mentally diseased architectural features, somehow allow javascripts from a browser to read kernel memory, something that would have been inconceivable until a decade ago
4) ... and finally that Intel doesn't even want to refund customers for all of the above, they just issued "mitigations" that don't work, decrease available computing power, and are also bugged themselves!
How can anybody suck so much? Are there any Intel employees on this forum? Can you read this? Why do you suck so much for being human beings? Are you actually human or are you bacteria? What do you think about the fact that all the rest of the world really really hate you? Have you ever considered suicide?
These are programming flaws. Programmers are never held accountable for anything. Every time something like this is reported folks on here make up every excuse why it's not the fault of the programmer.
Watch.
I think the main point is that people don't realize that they have a "lock" that they need to change the combination on. Perhaps with additional education people can "check their sh*t" and see if it needs to be changed. Then the bad actor can just look under their keyboard for the PW, but at least it won't be "admin" anymore.
Bad analogy. The difference here is once the attacker turns on remote monitoring, it occurs silently. There's no indication that it's happened and no way to recover. If you forget the combination to your safe, then 1) it's obvious and 2) you can still retrieve the contents in other ways.
This is not just a case of "stupid user". It's a poor design on Intel's part. Intel handed them a loaded shotgun with a hair trigger pointed directly at their foot.
Democracy is two wolves and a sheep voting on lunch.
can be compromised in moments by someone with physical access to the computer -- even bypassing BIOS passwords, Trusted Platform Module personal identification numbers, and Bitlocker disk encryption passwords -- by rebooting the computer, entering its BIOS boot menu...
How do you bypass the BIOS password if you can't get to the BIOS boot menu, because you don't have the BIOS password? I don't think "brief physical access" covers "opening the case and pulling the CMOS battery".
How is this even a flaw?
It's a case of default state + physical access == ownership.
This is nothing new at all.
whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
Agree this is not a flaw but talking about Intel design in general...we haven't heard any good news about AMT , Intel CPU design and the whole Intel Press Relations in denial.
Equifax
You would balk if you saw how many of the "top companies in America" don't give 2 shits about security, outside of whatever the latest CNN scare story is. I personally find it amazing how some of these corporations will spend tens of thousands of dollars on fancy security equipment.... that they never bother to actually configure.
You can show your C-levels the lock and hand them the key, but you can't make them set the latch.
Absolutely! Except there is going to come a point in time where a concerted effort by a small nation-state sponsored groups will be able to completely destroy corporate giants overnight. When they see empires around them begin to fall they will either start caring or become a casualty of cyberwarfare.
Anons need not reply. Questions end with a question mark.
I don't particularly see this as bad engineering even.
The thing ships disabled by default and with a default local only pwd to enable it OR lock out other access.
It can be disabled in the BIOS (and then the BIOS pwd activated) as well.
The config guide even says setting the password is a non optional step in any multi user/multi access environment, or you can get a sku where it's not even available.
no different than leaving the BIOS unlocked. I could boot a USB device that installs a rogue bootloader on the HDD.
Intel's fscked up plenty lately, this is just flamebait, and it's not even needed... focus on the real issues: AMT remote exec vuln, local buffer overflow in ME kernel (well shit), and of course Meltdown and Spectre.
whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
Stop it's ability to send info. outward via router port filtering ports 16992-16995 + 623-625 Intel AMT/ME uses in a modem/router external to OS/PC.
Intel ME/AMT operates from your motherboard but has NO CONTROL OF YOUR MODEM/ROUTER!
(This stops it cold talking in/out permanently OR being able to remotely 'patch' it to use other ports by Intel OR malicious actors/malware makers etc.!)
Additionally, once you disable the AMT engine's software interface (ez via software like the unistaller for it & DisableAMT.exe + the test in usermode via Intel-SA-00075-GUI.exe to TRIPLE CHECK)? A malware to 'repatch' this = impossible (bios updaters require it in usermode ware, e.g. ASUS).
(I only allow 80, 8080 & 443 in/out here on a SINGLE stand-alone system (no home LAN but TCP/IP connected online in BOTH my modem or router port filters or software firewalls))
HOWEVER - Be CERTAIN your modem/router's internal ware is "solid" too (turn off things like UPnP etc. & CHECK router/modem HAS NO KNOWN BACKDOOR EXPLOITS (tons do unfortunately)) - get it patched ASAP if it's KNOWN exploited & TONS of routers, ARE https://it.slashdot.org/comments.pl?sid=9995967&cid=53488785/
* GOOD ROUTERS/MODEMS HAVE PORT FILTERING OPTIONS (crappy ones don't)!
Especially after this finding: Intel Management Engine pwned by buffer overflow vendor patches for the vulnerability may not be enough http://www.theregister.co.uk/2017/12/06/intel_management_engine_pwned_by_buffer_overflow/ & Marcus Hutchin's "magic bit" patch doesn't help vs. this either.
APK
P.S.=> Good luck - it's the BEST EASIEST & CHEAPEST DEFENSE using what you already have (hopefully, again as not ALL modems have port filtering but most do & certainly GOOD ONES DO) vs. this threat by stopping it being able to communicate in/out period, from OUTSIDE of the INTEL chipset external to it via a router/firewall hardware... apk
The assertion was that AMD has other flaws. Your response doesn't even address that. This is independent of Meltdown (Intel only) and Spectre (all speculative execution chips).
OTOH, there haven't been any recent announcements of new AMD flaws that don't also apply to Intel.
I think we've pushed this "anyone can grow up to be president" thing too far.
Russia invests enormous resources into making their own Elbrus chips which are based off of MIPS architecture. They are certainly lagging a few generations from the fastest process technology, but that's what you got to do if you don't want to have NSA backdoors in weapons systems: https://en.wikipedia.org/wiki/Elbrus_(computer)
Rule 1 of security. Physical access trumps everything else. So you can't claim finding a defect that can be exploited physically is a breach. For that matter, someone could start plugging things into the motherboard. This just a lot of stupid hoopla. Everyone in OpenSource knows the REASON Open Source works is to bypass security through obscurity. Open Source DOES NOT and WILL NEVER (and neither will any security system) foil physical access 100% of the time. As for this - I've never even seen this option in my Bios choices. Just pop sensationalism.
Change log:
2018/01/01 - Added 14 Useful Links. Disable Intel ME 11 via undocumented NSA "High Assurance Platform" mode with me_cleaner, Blackhat Dec 2017 Intel ME presentation, Intel ME CVEs (CVSS Scored 7.2-10.0)
Intel CPU Backdoor Report
The goal of this report is to make the existence of Intel CPU backdoors a common knowledge and provide information on backdoor removal.
What we know about Intel CPU backdoors so far:
TL;DR version
Your Intel CPU and Chipset is running a backdoor as we speak.
The backdoor hardware is inside the CPU/Bridge and the backdoor firmware (Intel Management Engine) is in the chipset flash memory.
30C3 Intel ME live hack:
[Video] 30C3: Persistent, Stealthy, Remote-controlled Dedicated Hardware Malware
@21:43, keystrokes leaked from Intel ME above the OS, wireshark failed to detect packets.
[Quotes] Vortrag:
"the ME provides a perfect environment for undetectable sensitive data leakage on behalf of the attacker".
"We can permanently monitor the keyboard buffer on both operating system targets."
Backdoor removal:
The backdoor firmware can be removed by following this guide using the me_cleaner script.
Removal requires a Raspberry Pi (with GPIO pins) and a SOIC clip.
2017 Dec Update:
Intel ME on recent CPUs may be disabled by enabling the undocumented NSA HAP mode, use me_cleanerme_cleaner with -S option to set the HAP bit, see me_cleaner: HAP AltMeDisable bit.
Decoding Intel backdoors:
The situation is out of control and the Libreboot/Coreboot community is looking for BIOS/Firmware experts to help with the Intel ME decoding effort.
If you are skilled in these areas, download Intel ME firmwares from this collection and have a go at them, beware Intel is using a lot of counter measures to prevent their backdoors from being decoded (explained below).
Useful links (Added 2018 Jan 1):
Disabling Intel ME 11 via undocumented HAP mode (NSA High Assurance Platform mode)
me_cleaner: Set HAP AltMeDisable bit with -S option
Blackhat 2017: How To Hack A Turned Off Computer Or Running Unsigned Code In Intel Management Engine
EFF: Intel's Management Engine is a security hazard, and users need a way to disable it
Sakaki's EFI Install Guide/Disabling the Intel Management Engine
Intel ME bug storm: Hardware vendors race to identify and provide updates for dangerous Intel flaws.
CVE-2017-5689: An unprivileged network attacke
I guess using shit that incompetent admins spent extra money for and then left wide open by never configuring is now Intel's fault? This is the manual single-machine way to provision AMT / vPro. The only way this is a "flaw" is because the monkey in IT either didn't know what they were buying, or didn't bother to actually use what they were sold. By the way, once AMT / vPro are configured in any way (and if the admin of a machine is at all smart, they are doing this through a software provisioner when installing the OS in order to make sure the settings are the same each and every time), this "flaw" is no longer an option without the password, or without a few reboots and access to the BIOS.
Oh wait, is being able to press DEL, F1, F12, etc. and access the BIOS because a password wasn't set a "flaw" now too? Because that's the same fucking thing.
Don't want nefarious people configuring your vPro for you? Either don't buy a machine with vPro built in (yes, vPro is different from AMT - AMT is the base tech, vPro is a feature set running on top that allows the remote control, remote boot options, remote BIOS access, etc.), or fucking configure the thing you paid an extra $30 or so for.
What a sensationalist clickbait article that has been completely documented by Intel and the vast majority of their OEMs for like 6 years.
Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
Here's the good news about AMT and vPro: You can spend $30/system to have vPro on it, you can mass-configure with software Intel provides for free (but you need to buy a signed certificate specifically for AMT provisioning that matches your DHCP's DNS suffix for it to work), and then you can remotely reboot provisioned hardware from ISO images to reimage hardware from anywhere in the world, if you have granted access. It just needs to be on your network, and this includes notebooks that are wireless only.
I implemented this at a Fortune 500 that has about 50,000 Windows PCs spread across the US like trading stamps, and contracted with a 3rd party for on-site support. They charged a minimum 2 hours for a service call. vPro paid for itself in the first year just from reimage tickets alone - never mind the cost savings of not having to license a remote viewer for the help desk that is worth a damn and has the required security features for this company (HIPPA).
If you opt-in on something that:
- allows remote access in hardware
- leave it unconfigured
- allow some bad actor physical access
then you get what you deserve. Change any one of those three, and you're fine.
This "flaw" is in IT staff, not the product.
Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
Did you even use the link he provided? Did you read anything past "AMD has their own" before clicking reply?
He linked to an article about AMD's own in-hardware sideband management solution that is essentially the same fucking thing as Intel ME / AMT.
Who's the fucking shill now? Fanboyism + incompetence + ignorance = Slashdot hilarity.
Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
And by "nobody" you mean "all Fortune 500 companies, and hundreds of millions of home users"
Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
Apparently it's now a "flaw" with Kwikset, Schlage, Yale locks if I don't turn the lever on the inside to the 'lock' position. These lock manufacturers must do something about this immediately!
Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
Now why would you go and destroy the crux of his incredibly flawed and untrue statement with things like facts and experience?
What kind of argumentative strategy is that?!
Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
I find it hard to get worried about an exploit which requires physical access to the machine because if a hacker has that it pretty much means it doesn't matter what you've done and whether or not an exploit exists they're going to be able to get access to your data. Once a hacker has physical access to your machine it is pretty much game over.
I only please one person per day. Today is not your day. Tomorrow isn't looking good either. - Scott Adams
I don't know why you are in defensive posting. Never talked about flaws. Just the whole AMT is a fiasco because it has no use outside work environment yet imposed on consumer products WHICH most will never update the AMT firmware because they don't know how. If you followed Intel in the past two months you will know what I'm talking about. I'm not talking about BIOS or default user/pw value there.
I still wish you made a Linux version, or had capability with WINE
I've tried it in WINE, IIRC it doesn't make it through deduplication.
Not to sound like an ass, or trivialize your work, but it does seem like a fairly simple task from a programming standpoint IE one that WINE should be able to do (download and merge lists, save to a file) even though you may have to save it in the WINE drive, and install the file manually in Linux.
Cheers
I mean, when I buy a new gun safe or document safe for my home or office, it comes from the factory with a default combination. I have to set it to one of my choosing.
And what about when you buy a new electric frying pan? Do you remember to change the default factory passcode for one of your own? Or do you not even realize that your new frying pan has a passcode that needs to be set?
The latter is the situation with a huge number of PC buyers.
Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
So, the "flaw" is that the user forgot to set the lock?
No, the flaw is that there is an extra subsystem living within the CPU which is enabled by default, whether you want it or not, listening on all your network ports and waiting for someone to come along with the default password and take over the system.
If the system had to be enabled manually by someone with physical access (and the BIOS password, if one is set) then it would be reasonable to expect the administrator to change the access codes. The same applies if the remote management capability were the primary reason to choose this model over other designs, which would be closer to your document-safe example. However, being enabled by default out of the factory, including on systems which have no use for it, makes it an open backdoor. (Whether Intel or system integrators are responsible, it is clear that CPUs with this "feature" are being used in situations where remote management capabilities are more of a liability than an asset, and not something the end-user would have requested—unlike a combination on a safe.)
"The state is that great fiction by which everyone tries to live at the expense of everyone else." - Bastiat