Re: [gentoo-user] TPM feature - do I need it?
On Sunday 30 Nov 2014 03:21:16 Rich Freeman wrote: On Sat, Nov 29, 2014 at 6:44 PM, Mick michaelkintz...@gmail.com wrote: Thanks Rich, Also, what happens if the TPM chip, or the whole MoBo blows up? Will I ever be able to access my data using another PC? Only if you encrypted it. A TPM chip doesn't do much more than except store and retrieve data, and digitally sign things. It just tends to be used in a way that greatly limits the ability of arbitrary processes to access the data stored on the chip. With Linux you're basically completely in control. You choose to encrypt your drive and store the key in the TPM, and you instruct the TPM to only hand it over under the conditions you specify, such as a particular bootloader, kernel, and initramfs (or something like that - I've never implemented it myself). If somebody tries to boot your system with some other kernel/bootloader/initramfs then the TPM will not have the valid signature chain and it will refuse to divulge your full-disk encryption key. I imagine that you could generate the key outside the TPM and keep a copy of it somewhere and load it into the TPM, so that if you mess up you can just mount it manually. OK, but as I understand it although I can set up a passhphrase for the private key stored by the current oligopoly of manufacturers in a TPM, I can't extract it from the TPM. Would this mean that I will have no means of decrypting my drive, if I lose the TPM hardware module (e.g. due to hardware failure, fire, theft, etc.)? Access to my data will then become conditional on my having access to this unique TPM piece of silicon and its manufacturer's installed key, besides any private key passwd that I would have set up. Have I got this wrong, or is it that the TPM private key is merely the CA root certificate's key and I won't need this, unless I am creating/revoking user keys? Is there a way of using the user key separately and offline (on different hardware) without verification by the CA root certificate? Hmm ... I wonder if dm-crypt, LUKS and friends are a better way to achieve data protection for Linux users, without using some manufacturer's suspect certification credentials. I guess as long as I don't *have* to use Trusted Computing™, I won't care too much if it is on the MoBo. -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] TPM feature - do I need it?
On Sun, Nov 30, 2014 at 4:41 AM, Mick michaelkintz...@gmail.com wrote: OK, but as I understand it although I can set up a passhphrase for the private key stored by the current oligopoly of manufacturers in a TPM, I can't extract it from the TPM. Would this mean that I will have no means of decrypting my drive, if I lose the TPM hardware module (e.g. due to hardware failure, fire, theft, etc.)? Access to my data will then become conditional on my having access to this unique TPM piece of silicon and its manufacturer's installed key, besides any private key passwd that I would have set up. I'm not an expert on TPM, but I suspect you could generate the key outside the TPM, save a copy somewhere safe, and then load the key into the TPM and secure it. Hmm ... I wonder if dm-crypt, LUKS and friends are a better way to achieve data protection for Linux users, without using some manufacturer's suspect certification credentials. Encrypting a hard drive does not require trusting any key/certificate issued by a manufacturer. The advantage of using a TPM is that you can secure the drive without requiring a boot-time key to unlock the drive. This would be particularly useful if the system runs daemons that you might want to have run without anybody logging in. Also, using LUKS/etc requires a manually-entered key which will always be limited in entropy (mitigated using multiple rounds typically). The TPM is most useful in corporate environments though, since it doesn't require the PC owner to trust the PC user to not lose the key/etc. The whole point of the TPM is to reduce the risk of somebody who has physical possession of a machine compromising its security. In the corporate environment, the system owner often doesn't have physical control over the machine, and doesn't want to tell the user the encryption key which they can then proceed to write on a post-it stuck to the laptop. Those concerns apply just as much to a linux desktop as a windows one. Sadly, only ChromeOS really seems to take advantage of these capabilities in the linux desktop world. The whole Trusted Computing thing tends to turn off a lot of the linux community, much like UEFI is associated with thinks like unrootable devices. However, when used PROPERLY they're great tools. It is wrong for a device manufacturer to prevent a device owner from rooting a device they own. On the other hand, it is a great feature for a device owner to be able to prevent somebody who steals the device from being able to root it, or to block malware. The key is to keep control in the hands of the legitimate owner. -- Rich
Re: [gentoo-user] TPM feature - do I need it?
On 29/11/14 19:53, Mick wrote: I'm looking to buy a new PC and while looking at FM2+ MoBos I saw ASUS offers one with a TPM feature. It also sells it as a separate component it seems: http://us.estore.asus.com/index.php?l=product_detailp=5793 I recall reading in this list about it, but I am not sure if it offers any benefits to me as a user, or just adds a layer of complexity without any substantial benefit. Your views and experience with this TPM thingy? one thing that is very useful is using tpm to feed random number generator $ time dd if=/dev/random of=/dev/null bs=1 count=100 100+0 records in 100+0 records out 100 bytes (100 B) copied, 26.7494 s, 0.0 kB/s real0m26.751s user0m0.000s sys0m0.001s after starting trousers and rngd is much much much faster for real(er) random $ time dd if=/dev/random of=/dev/null bs=1 count=100 100+0 records in 100+0 records out 100 bytes (100 B) copied, 0.000275625 s, 363 kB/s real0m0.001s user0m0.002s sys0m0.000s it's also a safer place to drop keys into for example for use with grub trustedgrub and basically does secureboot without the need for uefi you can also use it to encrypt/decrypt *if* you trust it is not backdoored but does mean you can use it for LUKS instead of say a GPG crypted pass file - or you can have the tpm crypt your password into gibberish and then that gibberish phrase is the real password for LUKS
Re: [gentoo-user] TPM feature - do I need it?
On Sunday 30 Nov 2014 19:05:52 thegeezer wrote: *if* you trust it is not backdoored Well, yes, in the post Snowden era I do not trust it. At all. -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] TPM feature - do I need it?
On Sun, Nov 30, 2014 at 5:19 PM, Mick michaelkintz...@gmail.com wrote: On Sunday 30 Nov 2014 19:05:52 thegeezer wrote: *if* you trust it is not backdoored Well, yes, in the post Snowden era I do not trust it. At all. Keep in mind that you have to consider your threat model. I think it is fairly likely that the NSA has backdoors into most TPM chips (I'm not sure I'd even limit that to US-made chips). On the other hand, I also consider it likely that the NSA has zero-days for every browser around, and likely privilege escalation attacks against linux, and maybe even attacks against most of the hypervisors out there. If the NSA REALLY wanted your data, then they're going to get it. Of course, some attacks are a larger concern than others. Actually accessing a TPM back-door, or utilizing a zero-day, likely requires actively modifying your internet traffic, which carries a risk of detection. They won't hesitate to do that if they are on the trail of some terrorist, but they probably won't do that as part of some kind of widespread surveillance net since there would be a pretty high likelihood of detection if they did that to everybody, and then they lose their zero-day. On the other hand, if TPM-seeded random numbers aren't really random then they might be able to passively decode your SSL traffic which is something that would be almost impossible to detect, and that could be done as part of an effort to read everybody's TCP streams. Honestly, I don't really worry about the NSA. If they want to read my traffic they're going to do it, and the only way to stop them would be to wrap so much tinfoil around my head that I basically couldn't interact with anybody online. I'm more concerned with things like identity theft, cryptolocker, physical theft of laptops, and so on. A TPM provides a significant level of protection against some of these attacks. LUKS does as well, but at a cost of convenience and a risk of downtime if you're running it on a server and it ends up rebooting when you aren't around. Why would you care about full-disk encryption on a server? Well, ever have a hard disk die on you? Can you guarantee that when a hard drive dies that you'll always have the ability to wipe it before returning it for a warranty swap? With full disk crypto you're safe even if you can't wipe it. -- Rich
[gentoo-user] TPM feature - do I need it?
I'm looking to buy a new PC and while looking at FM2+ MoBos I saw ASUS offers one with a TPM feature. It also sells it as a separate component it seems: http://us.estore.asus.com/index.php?l=product_detailp=5793 I recall reading in this list about it, but I am not sure if it offers any benefits to me as a user, or just adds a layer of complexity without any substantial benefit. Your views and experience with this TPM thingy? -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] TPM feature - do I need it?
On Sat, Nov 29, 2014 at 2:53 PM, Mick michaelkintz...@gmail.com wrote: I'm looking to buy a new PC and while looking at FM2+ MoBos I saw ASUS offers one with a TPM feature. It also sells it as a separate component it seems: I can't get that page to load, but I can't imagine that you could find a motherboard that DIDN'T have a TPM that has been made anytime in the last decade. It doesn't tend to get a lot of use in the Linux world, though the Chromebook would be a BIG exception there. In the corporate windows world it gets very heavy use for full-disk encryption, and I think Win7 supports this out of the box (though big companies tend to use 3rd party software). Main uses for TPM include remote attestation, full-disk encryption (without the need to type a boot password), and secure credential storage only accessible via a trusted code path. The Linux kernel has support for TPM, but if you want to use many of the trusted boot features you need a bootloader that supports TPM. The main downside with TPM with something like Gentoo is that if you aren't careful you can make your keys inaccessible. I'd keep a copy of the keys somewhere safe if you plan to use it for something like full-disk encryption (and/or do regular backups). Otherwise if you incorrectly update grub you might find your drive completely inaccessible (if you're using a trusted boot path then you need to update the TPM when you update your boot path or the chip will no longer trust your grub/kernel/etc). The upside is that if you do it right you retain full control over the encryption and your system will be VERY hard to break into (without inside access - it is quite possible folks like the NSA have a backdoor, but you'll be very safe from more ordinary threats). -- Rich
Re: [gentoo-user] TPM feature - do I need it?
On Saturday 29 Nov 2014 20:23:51 Rich Freeman wrote: On Sat, Nov 29, 2014 at 2:53 PM, Mick michaelkintz...@gmail.com wrote: I'm looking to buy a new PC and while looking at FM2+ MoBos I saw ASUS offers one with a TPM feature. It also sells it as a separate component it seems: I can't get that page to load, but I can't imagine that you could find a motherboard that DIDN'T have a TPM that has been made anytime in the last decade. It doesn't tend to get a lot of use in the Linux world, though the Chromebook would be a BIG exception there. In the corporate windows world it gets very heavy use for full-disk encryption, and I think Win7 supports this out of the box (though big companies tend to use 3rd party software). Main uses for TPM include remote attestation, full-disk encryption (without the need to type a boot password), and secure credential storage only accessible via a trusted code path. The Linux kernel has support for TPM, but if you want to use many of the trusted boot features you need a bootloader that supports TPM. The main downside with TPM with something like Gentoo is that if you aren't careful you can make your keys inaccessible. I'd keep a copy of the keys somewhere safe if you plan to use it for something like full-disk encryption (and/or do regular backups). Otherwise if you incorrectly update grub you might find your drive completely inaccessible (if you're using a trusted boot path then you need to update the TPM when you update your boot path or the chip will no longer trust your grub/kernel/etc). The upside is that if you do it right you retain full control over the encryption and your system will be VERY hard to break into (without inside access - it is quite possible folks like the NSA have a backdoor, but you'll be very safe from more ordinary threats). Thanks Rich, it seems not all modern MoBos have it. This doesn't: http://www.asus.com/uk/Motherboards/A88XMA/specifications/ While this does: http://www.asus.com/uk/Motherboards/A88XGAMER/specifications/ Besides the complexity of it all and the risk of errors, it's the remote attestation part that worries me a bit. I mean this is not MSWindows, so the only entity I would expect to attest what I'm running on my machine is me. Well, fair enough, portage checks the hashes of the downloaded source files, but I would not want anyone to remotely check anything on my PC. If I enable this TPM thing, do I automatically open ports at pre/post-boot time giving access to my machine? Or is remote attestation something I have a say over? Also, what happens if the TPM chip, or the whole MoBo blows up? Will I ever be able to access my data using another PC? -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] TPM feature - do I need it?
On Sat, Nov 29, 2014 at 6:44 PM, Mick michaelkintz...@gmail.com wrote: Thanks Rich, it seems not all modern MoBos have it. This doesn't: Interesting, I had really thought they were ubiquitous. If I enable this TPM thing, do I automatically open ports at pre/post-boot time giving access to my machine? Or is remote attestation something I have a say over? It has to be implemented in software. The only thing the chip does is let software call it and ask it to sign data. The chip can't talk to the network on its own. Remote attestation doesn't actually give anybody access to your machine. Now, it does let you prove to somebody that they actually do have access to your machine, if you installed software that gave them access. Suppose I gave you a piece of software and asked you to install it on your machine so that I could evesdrop on everything you do. If you refuse to do it I'll fire you from your job. Without remote attestation you could install that software in a VM or emulator somewhere and tell me you complied, and it would be impossible for me to tell that you did this if your VM was well-implemented. On the other hand, if my software implemented remote attestation then the best you could do is block it, in which case I brand you as non-compliant and fire you. With remote attestation I could get a list of the boot path from the firmware to my software and see if there is anything in it that could tamper with my software. If I dictate that you have to run TrustedCo Linux v5 then I could see that you're running an unpatched version of it and know that my process is directly talking to the kernel, which I know doesn't let you tamper with it, and that kernel is directly running on the hardware having been loaded by a trusted firmware/bootloader. Also, what happens if the TPM chip, or the whole MoBo blows up? Will I ever be able to access my data using another PC? Only if you encrypted it. A TPM chip doesn't do much more than except store and retrieve data, and digitally sign things. It just tends to be used in a way that greatly limits the ability of arbitrary processes to access the data stored on the chip. With Linux you're basically completely in control. You choose to encrypt your drive and store the key in the TPM, and you instruct the TPM to only hand it over under the conditions you specify, such as a particular bootloader, kernel, and initramfs (or something like that - I've never implemented it myself). If somebody tries to boot your system with some other kernel/bootloader/initramfs then the TPM will not have the valid signature chain and it will refuse to divulge your full-disk encryption key. I imagine that you could generate the key outside the TPM and keep a copy of it somewhere and load it into the TPM, so that if you mess up you can just mount it manually. It is a bit like UEFI - if you use it properly it becomes a tool which allows the device owner to secure his own device against other intruders. The problems only come when people want to treat your device as if it is their device. With Linux you control almost all of it, though I would GREATLY prefer it if TPM chips did not come with vendor-supplied certificates (that is, if there was no way for anybody to determine if the master key inside was externally loaded by the end user or not). If they didn't come pre-loaded with certificates then they would be just as usable for securing your own machine (since you could add your own certificate/etc), but they would be useless for DRM (since no key was ever in the chip at a time when it was in the possession of somebody trusted by the media companies). I also don't have much more than a general interest in this area, so I'm sure there are some fine details above which are inaccurate. The gist of it should be right though. -- Rich
