Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: GnuPG: users

private key protection

 

 

First page Previous page 1 2 Next page Last page  View All GnuPG users RSS feed   Index | Next | Previous | View Threaded


jerome at jeromebaum

Oct 18, 2011, 6:39 AM

Post #26 of 47 (439 views)
Permalink
Re: private key protection [In reply to]

> It doesn't prevent a trojan from signing something other than what you
> intended (if it's your master key on card, even another key or a new
> sub-key) but whether this is a problem depends on your threat model.

I should mention that the current OpenPGP card spec doesn't let the card
know whether it's signing a key or signing data. So there's no way to
prevent this attack other than not keeping your master-key on card.

I prefer keeping the master-key encrypted thrice and printed out in a
vault, surrounded 25x8 by guards authorized to use lethal force.

But seriously, I keep the master-key encrypted/printed and store it in
my safe deposit box. The sub-key goes on the card. Trojan issue is a
much smaller issue then, as the card includes a signature counter. I
also keep a backup of the encryption key in case the card breaks. That's
probably a good idea.

--
PGP: A0E4 B2D4 94E6 20EE 85BA E45B 63E4 2BD8 C58C 753A
PGP: 2C23 EBFF DF1A 840D 2351 F5F5 F25B A03F 2152 36DA

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


rjh at sixdemonbag

Oct 18, 2011, 6:49 AM

Post #27 of 47 (436 views)
Permalink
Re: private key protection [In reply to]

I'm going to keep this as short as possible, because we've already hit
the point at which we're casting far more heat than light.

> Oddly, I don't recall Jerome ever making a statement remotely like
> "If I steal your decrypted key, ...". I only remember him stating
> that he thought, as did I, that the OP meant that he wanted ways to
> prevent people stealing his secret key material when he said: "what
> is the best way to protect your private key from getting stolen?".
> Anthony interpreted it as somebody stealing the keyring, and Jerome
> disagreed on that interpretation. As do I.

GnuPG depends on you having physical control of the hardware for the
duration of your use of the system. If this fails, then there's nothing
GnuPG -- or anything, for that matter! -- can do to keep your secret key
material safe.

If I put my secret key on a system that is later compromised, I can
still be confident in the security of my secret key. If I log into that
machine and use my secret key even once, though, that key needs to be
considered compromised because I've failed to uphold the absolute
prerequisite for GnuPG usage: control of the hardware during my
interaction with it.

Secret key material can only be compromised in two situations: either
(a) someone you don't trust has root on your system while you're using
GnuPG, in which case it's a game-over and the only defense is "well,
don't do that, then!", or (b) someone compromises your PC while you're
not using GnuPG and steals your private key.

(a) is true, but it doesn't lead anywhere useful. That makes it
trivial. Why are we even discussing a triviality?


_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


takethebus at gmx

Oct 18, 2011, 7:06 AM

Post #28 of 47 (433 views)
Permalink
Re: private key protection [In reply to]

Thanks to everyone for the helpful answers. Maybe I'll buy a smartcard, it seems more convinient than rebooting for every email.

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


jerome at jeromebaum

Oct 18, 2011, 7:20 AM

Post #29 of 47 (436 views)
Permalink
Re: private key protection [In reply to]

On 2011-10-18 16:06, takethebus [at] gmx wrote:
> Thanks to everyone for the helpful answers. Maybe I'll buy a
> smartcard, it seems more convinient than rebooting for every email.

What country are you in? For Germany, kernelconcepts sells the OpenPGP
card v2 and cryptoshop sells a very basic USB card reader (no PIN entry)
for a total below 50 €.

(IIRC cryptoshop is based in Austria, but they ship to Germany.)

--
PGP: A0E4 B2D4 94E6 20EE 85BA E45B 63E4 2BD8 C58C 753A
PGP: 2C23 EBFF DF1A 840D 2351 F5F5 F25B A03F 2152 36DA

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


peter at digitalbrains

Oct 18, 2011, 7:22 AM

Post #30 of 47 (434 views)
Permalink
Re: private key protection [In reply to]

> (a) is true, but it doesn't lead anywhere useful. That makes it
> trivial.

Seems like you keep asserting Jerome posed (a) as something insightful. I don't
remember someone other than you posing (a) at all.

I really see no point in keeping on telling people they said something different
than what they meant to say. Isn't it way more productive to determine what they
meant, rather than what /you/ (or anyone) read in it?

Are we here to catch eachother on potentially saying something, for instance,
trivial, and going "aha! Got ya!"? Or are we here to discuss crypto and stuff
having to do with crypto?

Meanwhile, you are right about the heat versus light ratio, so I will stop this
side discussion as well, hopefully even if it annoys me some more :).

Peter.

--
I use the GNU Privacy Guard (GnuPG) in combination with Enigmail.
You can send me encrypted mail if you want some privacy.
My key is available at http://wwwhome.cs.utwente.nl/~lebbing/pubkey.txt

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


wk at gnupg

Oct 18, 2011, 7:22 AM

Post #31 of 47 (433 views)
Permalink
Re: private key protection [In reply to]

On Tue, 18 Oct 2011 15:19, rjh [at] sixdemonbag said:

> Arguably we should be using 'certificate' to describe keys, but

We tried that in the Gpg4win manuals. However it turned out that this
term as other problems when used with OpenPGP keys (ah well, keyblocks).

> honestly, that's a losing battle: the community's inertia on the subject
> of 'key' is immense.

Right. There is a public key and there is a private (aka secret) key.
How they are made up is a technical detail.


Shalom-Salam,

Werner

--
Die Gedanken sind frei. Ausnahmen regelt ein Bundesgesetz.


_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


jerome at jeromebaum

Oct 18, 2011, 7:23 AM

Post #32 of 47 (442 views)
Permalink
Re: private key protection [In reply to]

> Well, not quite. Eventually you would get it. The task of security
> systems is to make "eventually" be longer than:
>
> o the payoff is worth; or
> o the time it takes to be discovered; or
> o the time it takes for the secured object to lose its value.
>
> Statistically, that is. You could get it right on the first try, but
> you very probably won't. You are guaranteed to get it right if you
> try every possible value.

Right, that's a good point I think we all considered "trivial" when
maybe we shouldn't have. In your threat model you should determine for
how long your data should be safe (per attacker type) before you go
ahead and make decisions about key protection.

While we're discussing the STEED proposal in the other thread, do you
think it's better to educate your users and risk loosing them or do you
think it's better to provide "sensible" defaults for the "average"
threat model and assume they'll learn everything else over time and
start tweaking?

I suppose the latter model fits the "power user" case well, where they
start using the tool and eventually learn about other features and start
tweaking.

--
PGP: A0E4 B2D4 94E6 20EE 85BA E45B 63E4 2BD8 C58C 753A
PGP: 2C23 EBFF DF1A 840D 2351 F5F5 F25B A03F 2152 36DA

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


wk at gnupg

Oct 18, 2011, 7:25 AM

Post #33 of 47 (433 views)
Permalink
Re: private key protection [In reply to]

On Tue, 18 Oct 2011 15:05, rjh [at] sixdemonbag said:

> No, it's still a single file ("pubring.gpg", for instance, is the public
> keyring). I just can't promise that it's still a raw stream of RFC4880
> octets.

It still is for the public keys.

2.1 changes the format of the secring (well, dropped it entirely and
stores only the needed bits elesewhere).


Salam-Shalom,

Werner

--
Die Gedanken sind frei. Ausnahmen regelt ein Bundesgesetz.


_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


jerome at jeromebaum

Oct 18, 2011, 7:26 AM

Post #34 of 47 (439 views)
Permalink
Re: private key protection [In reply to]

> Right, that's a good point I think we all considered "trivial" when
> maybe we shouldn't have. In your threat model you should determine for
> how long your data should be safe (per attacker type) before you go
> ahead and make decisions about key protection.

To clarify, this is what we should tell the OT instead of telling him
stuff like "smart cards are 'better'". Kumtraya!

--
PGP: A0E4 B2D4 94E6 20EE 85BA E45B 63E4 2BD8 C58C 753A
PGP: 2C23 EBFF DF1A 840D 2351 F5F5 F25B A03F 2152 36DA

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


mwood at IUPUI

Oct 18, 2011, 7:59 AM

Post #35 of 47 (436 views)
Permalink
Re: private key protection [In reply to]

On Tue, Oct 18, 2011 at 04:23:42PM +0200, Jerome Baum wrote:
[snip]
> While we're discussing the STEED proposal in the other thread, do you
> think it's better to educate your users and risk loosing them or do you
> think it's better to provide "sensible" defaults for the "average"
> threat model and assume they'll learn everything else over time and
> start tweaking?

I think we would be in error to think about "users" as a single class.

I usually try to educate lightly -- to make all users aware that there
is much more to learn, and to indicate how more learning might be to
their advantage. Then provide sensible defaults, so that those who
choose to go no deeper will get some benefit, and in-depth
documentation for those who do choose to go deeper so that they can
reap the full benefit (or, at least, as much as each is willing to
work for).

I was pleased to see room for different classes of users in the STEED
paper. When I encounter software that tries to be helpful, my own
first thought is: how do I turn that off? But I recognized long ago
that I was never a "typical" user and my own inclinations are no guide
to popularity. :-/

--
Mark H. Wood, Lead System Programmer mwood [at] IUPUI
Asking whether markets are efficient is like asking whether people are smart.


jerome at jeromebaum

Oct 18, 2011, 8:10 AM

Post #36 of 47 (435 views)
Permalink
Re: private key protection [In reply to]

> I was pleased to see room for different classes of users in the STEED
> paper. When I encounter software that tries to be helpful, my own
> first thought is: how do I turn that off? But I recognized long ago
> that I was never a "typical" user and my own inclinations are no guide
> to popularity. :-/

That's a big UI bug with Thunderbird IMO: The automated account setup is
really nice, until you run into a case where it doesn't work. There's no
"expert" button to force a setup. The workaround is to go offline and
then setup the account...

So yes definitely expert buttons, I was talking about those users that
aren't yet experienced with crypto.

I like your idea of giving guidance on where-about they are still
getting good returns on their learning efforts.

--
PGP: A0E4 B2D4 94E6 20EE 85BA E45B 63E4 2BD8 C58C 753A
PGP: 2C23 EBFF DF1A 840D 2351 F5F5 F25B A03F 2152 36DA

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


dcenteno at ydl

Oct 18, 2011, 9:23 AM

Post #37 of 47 (434 views)
Permalink
Re: private key protection [In reply to]

On 10/17/11 5:18 PM, takethebus [at] gmx wrote:
> Hi everybody,
>
> what is the best way to protect
> your private key from getting stolen?

Page 29 (http://www.gnupg.org/gph/en/manual.html#AEN513) of the Gnu
Privacy Handbook (http://www.gnupg.org/gph/en/manual.html)recommends a
strong passphrase to protect the key. Another strategy is to create
sub-keys derived from the private key and use those sub-keys for signing
and encrypting anything. This would also mean that you export the
public key of whichever sub-key you decide to use -- not your private
key. As the use of the public sub-key cannot be used to derive the
private key utilizing the sub-key strategy may be the most sensible
strategy.
>
> I think:
>
> 1. Using gnupg on a windows PC with internet connection is not good, because there are too many trojans out there.
In all fairness, the PC is as weak or strong as it's user. In other
words, if you are not willing to do the "nitty-gritty and sometimes
research as relentless in nature as Indiana Jones - regarding how you
defend your operating system then believe it or not choosing Linux or
the Mac won't save you from your laziness. Sorry, but that's the truth.

You have to have your own drive to master whatever technology
(mathematics, coding language, nuance and more) necessary to defend
yourself, your family and your property. If you don't or won't make the
effort -- understand that this is exactly what those who create malware
rely upon. The other crowd who rely on your "lack of will" are the
commercial entities who benefit from those who just want "someone else"
to handle the details and who are willing to pay for whatever appears
"on the shelf".
>
> 2. Using gnupg on a linux PC with internet connection (like privatix, see http://www.mandalka.name/privatix/index.html.en ) is better since there are fewer(?) security holes and trojans out there. How big do you think is the thread?
>
IF you decide you are serious regarding Linux then Debian or Red Hat
remain the two you should rely upon. Everyone else, follows them. Of
course, if you are really brave and really know what you are doing then
Slackware is reliable.

Again don't rely on anyone, especially in Linux, to provide you with a
satisfactory and reliable defense if you have no clue as to how it
works, or how you can repair it should something go wrong or how to
improve it's reliability as hacking and threat environment's increase.


> 3. The best way is to have one PC connected to the internet and another, without an internet connection (missing network drivers and a fully encrypted hard disk for instance), which you use to decrypt and encrypt messages. You use an USB stick to carry messages from the internet PC to the one not connected to the net. If you don't have two PCs, you can use another USB stick with privatix without network drivers on it.
>
> Which software can I use under point 3 to put my messages in order (date, sender, etc.) on a linux system?
>
> Most people use something like point 2, don't they?
>
> Point 3 is the only satisfying to me, since I find it hard to judge the the thread in point 2. Additionally point 3 makes it easier to see when your key might have been stolen: If you see traces that someone broke into your house and searched everything for the hidden privatix USB stick. Only experts might notice a trojan under point 2.
>
> Thanks for answers,
> Jan
>


I think I recall seeing that question (3) on a Computer Science exam.
The truth, unfortunately, is that there is no "best way".
Unfortunately, there is another level of system attack which was used
successfully against HBGary and should be a tale elevated to the level
of Grimm's Fairy Tales until it seeps into the unconscious and conscious
level of each persons awareness. Read this article and I'm sure you'll
get my point:
http://www.theregister.co.uk/2011/03/17/hbgary_anon_hacker_interview/

HBGary believed it's own hype regarding their sophistication and skills;
simply stated as a corporation they failed the same way or close enough
as the individual who believes s/he is a "legend - in their own mind".
The trap very similar to that limited thought is to believe that your
system is safe because it is isolated; in fact the weakness of your
system (regardless what you buy) is really -- you!

This side of the problem can be intuited by understanding how many
people fall the Nigerian or Russian or other scam ploy every day.
In other words, be aware of your own susceptibility to being tricked,
taken, and mislead such as when we are distracted. It is one thing to
be enjoyably tricked at a magic show, quite another emotion is
experienced when your data is stolen and you have no clue how or why
until you realize that it was your fault for trusting so and so.

I have no intention of being overly discouraging as much as underlying
the fundamentals regarding why computer security, encryption methods,
etc. are constantly becoming more complex and involved. There really is
only one reasonable approach: dive in and master the details yourself.
You wouldn't trust a used car salesperson or insurance guy to tell you
everything is fine, right? You've got to know quite a bit to know when
you are being "taken", right? Well, technology is no different. In
some ways, it's harder because a lot of people don't want to work that hard.

If you remember however that both criminals and commercial markets are
depending upon that natural laziness which we each have -- you may have
a chance of developing your own incentive to learn and master what you
must and maybe a little more. That is the best defense.

All the best...
Attachments: signature.asc (0.25 KB)


faramir.cl at gmail

Oct 19, 2011, 1:43 PM

Post #38 of 47 (433 views)
Permalink
Re: private key protection [In reply to]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

El 18-10-2011 10:07, Peter Lebbing escribió:
...
> A capable enough hacker might infect the USB pendrive while it is
> in your internet-connected PC and that way still gain access to the
> non-connected system.

Ok, but if the online computer uses Windows, and the offline one
uses Linux, then it would be a multiplataform trojan horse... that is
not likely to be a common case.

Best Regards
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBCAAGBQJOnzaAAAoJEMV4f6PvczxAxxcH/RyWHL7x47kCWDFE8uYL1fY9
eS7beCvPQpWvsGKZaQkjFeTVn86o442AkbrZ7Awy03WtzJJvOezQ6km6NrcB2dHa
R0bnLYj41kjvA8s2/AenDk/OvNm3iPgJrHtp6NA+O9sT4QITiQNb0yVQqGQoQwuY
gfDT4Ne1ZpKC6yml3Fl/wfnK6Mm1YXK6o7LEIk7GbDeaeMl8LDAzR9SYmFYnxwps
r6Qk0abh7RtC6DET6DbxamD2VDN9bglrrBqVUMqFUzeYfe0luxGyfSBL9ToDDKc+
YZ54vvrVC4ABgD2oJJTbQ1kB3cYOhaeFlbWXFfvvrrJnNNrH5T09kHahadGFMjc=
=RV1p
-----END PGP SIGNATURE-----

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


peter at digitalbrains

Oct 19, 2011, 1:54 PM

Post #39 of 47 (431 views)
Permalink
Re: private key protection [In reply to]

On 19/10/11 22:43, Faramir wrote:
> Ok, but if the online computer uses Windows, and the offline one
> uses Linux, then it would be a multiplataform trojan horse... that is
> not likely to be a common case.

Define your threat model... are we talking random trojan infection or a focused
attacker trying to gain your key? Because in the latter case, I hardly think
commonality matters.

Peter.

--
I use the GNU Privacy Guard (GnuPG) in combination with Enigmail.
You can send me encrypted mail if you want some privacy.
My key is available at http://wwwhome.cs.utwente.nl/~lebbing/pubkey.txt

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


rjh at sixdemonbag

Oct 19, 2011, 1:55 PM

Post #40 of 47 (428 views)
Permalink
Re: private key protection [In reply to]

On 10/19/2011 4:43 PM, Faramir wrote:
> Ok, but if the online computer uses Windows, and the offline one
> uses Linux, then it would be a multiplataform trojan horse... that is
> not likely to be a common case.

At this point we're throwing conjecture onto conjecture. If the offline
one happened to be a PowerPC architecture running Yellow Dog Linux, then
the first bit of malware would have to target Windows/x86, the second
would have to target Linux/PPC, and that's even *more* unlikely to be a
common case, and oh, don't forget if we're actually... etc., etc.
Attachments: signature.asc (0.18 KB)


faramir.cl at gmail

Oct 19, 2011, 2:09 PM

Post #41 of 47 (431 views)
Permalink
Re: private key protection [In reply to]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

El 19-10-2011 17:54, Peter Lebbing escribió:
> On 19/10/11 22:43, Faramir wrote:
>> Ok, but if the online computer uses Windows, and the offline one
>> uses Linux, then it would be a multiplataform trojan horse...
>> that is not likely to be a common case.
>
> Define your threat model... are we talking random trojan infection
> or a focused attacker trying to gain your key? Because in the
> latter case, I hardly think commonality matters.

You are right, I was thinking about random trojan infection (maybe
not 100% random, since a private key stealing trojan would be focused
on OpenPGP users, rather on average users). But if somebody wants MY
private key, then probably there would be an attack involving picking
my lock, infecting my BIOS, or some other 007-like activity. But in
that case, the victim might be involved in some organization that
should develop policies to deal with that risk.

Best Regards
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBCAAGBQJOnzyeAAoJEMV4f6PvczxA7eEH/j3wjkHNcwPNd2hSz1NXmIl0
KCMkE2H2BEqS19AhpDMmYdB4EVddDsDJg1rLa7W+he5o/4g6WPueLoeeh+Rqbj0T
IZCNN6KlVWgZ2P9JLt9cc5H9TVU1f3O1HtJUThwQJfsFygDBrk/HqpTvsJcXqU51
yAd2aw2gudI8FtJAz5hawRMABzIKObH3wJGbpQfVR1ih91zsjisPCJXt+4grwg2b
lxTS2tR8RnuZJPkmmBZTyAKNkapdGnJ2BiXPKYY8rqtPzM035hqDlsiVAHvea0ie
UYtOkTVXGVgW0xQlXY/0j4HKBm/xuNltUiZPja8EIGV2KMvoV16iYmCVa5CpURc=
=MJed
-----END PGP SIGNATURE-----

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


rjh at sixdemonbag

Oct 19, 2011, 3:04 PM

Post #42 of 47 (433 views)
Permalink
Re: private key protection [In reply to]

On 10/19/2011 4:54 PM, Peter Lebbing wrote:
> Because in the latter case, I hardly think commonality matters.

As an example:

Three years ago I was thrown into a week-long sink-or-swim course on
malware analysis, taught by an instructor who was a principal scientist
at a company that's a big name in that field. (Due to the subject
matter of this story, I am not allowed to give names: they don't want to
be publicly associated with this story. You'd recognize the company
name if you heard it, though.) The first thing we did was crack our
cases to verify that our machines had no network cards. While we were
doing this, the instructor entertained us with a funny story about why
we were doing this.

A couple of years before that course, a new piece of malware was
reported to the company. In turn it was sent to the malware analysis
lab, where the instructor was the guy tasked with looking at it. He was
running a Windows VM within a Linux environment on a computer that was
physically disconnected from the internet and had the wifi card turned
off. He fired up IDA Pro (a popular debugger) and began studying this
boring, broken piece of malware. Within a couple of minutes the
sysadmins noticed something wrong and killed all network access in the
building. All signs pointed to the instructor's machine being the
source of the problem.

The malware was the work of an evil genius. As input to a PC, it was a
bunch of nonsense that crashed hard before it could do anything. As
input to IDA Pro, it was a carefully crafted input that hijacked IDA
Pro. It then discovered it was running inside a virtual machine, used
an exploit to get out into the Linux environment, brought up the wifi
connection and associated with the first network it could. Wacky
hijinks ensued.

You can find some more on this subject in "The IDA Pro Book," by Chris
Eagle. NIST also has a brief writeup on it:

http://web.nvd.nist.gov/view/vuln/detail?vulnId=CVE-2005-0115

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


Lists.gnupg at mephisto

Jun 25, 2012, 8:08 AM

Post #43 of 47 (240 views)
Permalink
Re: private key protection [In reply to]

On Tue, Oct 18, 2011 at 09:15:14AM -0400 Also sprach Mark H. Wood:
> On Tue, Oct 18, 2011 at 02:10:07PM +0200, Jerome Baum wrote:
> > >> I'm going to lean very far out the window and assume he meant the actual
> > >> private key, not the private key-ring/-file/...
> > >
> > > I'm not sure I understand the distinction you're making there.
> >
> > One is protected with a passphrase (i.e. it's encrypted), the other is
> > in the clear.
> >
> > If I manage to steal your private keyring, then yes the very strong
> > passphrase should grind my attempts to steal your key to a halt.
>
> Well, not quite. Eventually you would get it.

Eventually being... the age of the Earth? Provided one's private key
is protected by a suitably "good" passphrase, then the problem of
cracking the symmetric encryption used to protect the private key is
comparable to the problem of cracking an encrypted message's session
key.

That is to say, if an attacker has the resources to break the
encryption used to protect a private key, in a practical span of time,
that implies that they can apply the same techniques to reading your
encrypted messages without the private key, which makes stealing it
less than essential.

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


wk at gnupg

Jun 25, 2012, 8:44 AM

Post #44 of 47 (241 views)
Permalink
Re: private key protection [In reply to]

On Mon, 25 Jun 2012 17:08, Lists.gnupg [at] mephisto said:

> cracking the symmetric encryption used to protect the private key is
> comparable to the problem of cracking an encrypted message's session
> key.

No, it is not. The entropy in a session key matches the size of the
session key. The key used to protect the private key is commonly much
weaker. A passphrase providing an adequate amount of entropy is not
useful because a user won't be able to remember it correctly. Further,
a brute force attempt on the protected private keys needs to be done
only once, whereas it has to be done for each encrypted message, if you
want to target the session key.


Shalom-Salam,

Werner

--
Die Gedanken sind frei. Ausnahmen regelt ein Bundesgesetz.


_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


rjh at sixdemonbag

Jun 25, 2012, 8:55 AM

Post #45 of 47 (245 views)
Permalink
Re: private key protection [In reply to]

On 06/25/2012 11:08 AM, Kevin Kammer wrote:
> Eventually being... the age of the Earth?

(I do not disagree with Kevin: this is an emphatic agreement.)

There is a minimum energy associated with flipping a bit -- something so
small that a single proton has the energy to flip about a trillion bits.

Let's say you have a remarkably efficient OS that can test a given key
while only flipping 10,000 bits. Multiply that times the number of
attempts you'd have to make to brute-force a 128-bit key and you get a
really big number, so big that it no longer makes sense to describe it
in terms of nuclear warheads. The best, most visceral way of saying it
is, "You must have 340 kilos of antimatter to run your computer."

If you happen to have 340 kilos of antimatter lying around, then yes,
brute-forcing is certainly possible. I deeply hope you don't. I like
Earth: all my stuff is here.

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


rjh at sixdemonbag

Jun 25, 2012, 9:00 AM

Post #46 of 47 (241 views)
Permalink
Re: private key protection [In reply to]

On 06/25/2012 11:44 AM, Werner Koch wrote:
>> cracking the symmetric encryption used to protect the private key is
>> comparable to the problem of cracking an encrypted message's session
>> key.
>
> No, it is not. The entropy in a session key matches the size of the
> session key. The key used to protect the private key is commonly much
> weaker. A passphrase providing an adequate amount of entropy is not
> useful because a user won't be able to remember it correctly.

Speaking purely for myself, my passphrase is 16 bytes from /dev/urandom
dropped into base64. It took me a weekend to memorize it, but the peace
of mind has been well worth it.

It is possible, though, that I'm demented. :)

_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users


mick.crane at gmail

Jun 25, 2012, 11:37 AM

Post #47 of 47 (240 views)
Permalink
Re: private key protection [In reply to]

On Mon, June 25, 2012 5:00 pm, Robert J. Hansen wrote:
> On 06/25/2012 11:44 AM, Werner Koch wrote:
>>> cracking the symmetric encryption used to protect the private key is
>>> comparable to the problem of cracking an encrypted message's session
>>> key.
>>
>> No, it is not. The entropy in a session key matches the size of the
>> session key. The key used to protect the private key is commonly much
>> weaker. A passphrase providing an adequate amount of entropy is not
>> useful because a user won't be able to remember it correctly.
>
> Speaking purely for myself, my passphrase is 16 bytes from /dev/urandom
> dropped into base64. It took me a weekend to memorize it, but the peace
> of mind has been well worth it.
>
> It is possible, though, that I'm demented. :)
reading this it occurs it me that keyboards are cheap so it would be
reasonable to swap all the keys about on a keyboard and then use some
easily memorably combination of real words to save on so much memorizing.


mick

--
keyID: 0x4BFEBB31



_______________________________________________
Gnupg-users mailing list
Gnupg-users [at] gnupg
http://lists.gnupg.org/mailman/listinfo/gnupg-users

First page Previous page 1 2 Next page Last page  View All GnuPG users RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.