Armory - Discussion Thread

[Newar]

I am switching over from bitcoin-qt to armory.  I have a wallet that I want to import the addresses to Armory on an offline computer.  Whats the easiest way to do this and once imported how can I send the btc to a newly generated offline address. 

Last time I tried importing private keys to Armory from Bitcoin Core there was no support for importing compressed keys in Armory. So, right now you can't.

[Rampion]

Alan, what do you think of "Hash Hyena" claims?

He basically says that he is generating trillions of addresses non stop and getting a few collissions that will grow over time. His logic is that the entropy of Armory, Bitcoin Core, Multibut, Electrum etc. (and generally any other wallet that uses a RNG based on software) is flawed and results in a highly reduced keyspace, which will result in collissions with enought computing power and space devoted to private keys bruteforcing.

See: https://bitcointalk.org/index.php?action=profile;u=380718;sa=showPosts

What is your position on that? I remember that Dabs (I think) requested a feature in order to input "true entropy" from a physical source (a decks of cards, dices, etc.) - how's the progress on that? It might be a good thing to mitigate software RNG issues, but I assume it might be difficult because an unssavy user might make entropy worse by not using "true entropy" right.

[coraz]

He basically says that he is generating trillions of addresses non stop and getting a few collissions that will grow over time. His logic is that the entropy of Armory, Bitcoin Core, Multibut, Electrum etc. (and generally any other wallet that uses a RNG based on software) is flawed and results in a highly reduced keyspace, which will result in collissions with enought computing power and space devoted to private keys bruteforcing.

See: https://bitcointalk.org/index.php?action=profile;u=380718;sa=showPosts

This guy is either trolling for the sake of it, or he is going to bundle some malware into his upcoming "bitcoin cracker".

Anyway, if you don't trust your machine's RNG, there is a way to use your own entropy via restore backup feature: https://bitcointalk.org/index.php?topic=673035.0

[etotheipi]

Alan, what do you think of "Hash Hyena" claims?

He basically says that he is generating trillions of addresses non stop and getting a few collissions that will grow over time. His logic is that the entropy of Armory, Bitcoin Core, Multibut, Electrum etc. (and generally any other wallet that uses a RNG based on software) is flawed and results in a highly reduced keyspace, which will result in collissions with enought computing power and space devoted to private keys bruteforcing.

See: https://bitcointalk.org/index.php?action=profile;u=380718;sa=showPosts

What is your position on that? I remember that Dabs (I think) requested a feature in order to input "true entropy" from a physical source (a decks of cards, dices, etc.) - how's the progress on that? It might be a good thing to mitigate software RNG issues, but I assume it might be difficult because an unssavy user might make entropy worse by not using "true entropy" right.

If those apps are using truly poor RNGs, then I suppose it's possible.  But it would have to be pretty bad to be getting collisions within trillions of addresses +/- a few orders of magnitude.  Address strings are 160 bits.  So a collision means that two different public keys produce the same address.  If you want to "collide" with someone else, i.e. get an address string that matches someone else who has coins, it will take you approx 2¹⁶⁰/numberOfUsers.  Even if we assume a trillion users&addresses (2⁴⁰), that's still 120 bits left which is within range of universally secure (128 bits).

Now, if he wants to generate gajillions of addresses and find collisions within any two of them, that's remarkably easier, but also remarkably less useful.  The idea being that he wants any two public keys that have the same hash.  i.e. generate addresses until you have two that match.  Not useful, since you're guaranteed to collide with an address that has no money, but it would still be interesting.

In such cases, if the number scales with approx sqrt(N).  i.e. if you are generating 160-bit values, there's 2¹⁶⁰ possibilities.  You need to generate approximately sqrt(2¹⁶⁰)=2⁸⁰=1,208,925,819,614,629,174,706,176 to have a single collision.  That's a trillion trillion... while that might actually be possible with an extrardinary amount of computing power (in the future), you're still generating collisions that don't benefit you.

Scale all this by the quality of the RNG -- if the RNG is weaker, the state is reduced.  Technically, if the hash160 (ripemd160) is weak, the address space could be smaller too.  But I would have a tough time believing that desktop apps would not be producing enough entropy, as long as they are using a good RNG.

On that note, Armory uses Crypto++ was is considered a cryptographically-secure RNG (X9.17 with OS-provided seeding).  On top of that, Armory pulls in system files, mouse clicks, keypresses, and a desktop screenshot, to add to the Crypto++ RNG entropy pool.  I made sure when selecting these sources that it would guarantee at least 256 bits of entropy to be added to the pool even if Crypto++ was really weak.

You're right about the external entropy:  we don't provide an option for users to add their own entropy for the reason you described (it becomes a channel for creating brainwallets, too).  However, we do it ourselves because it's pretty simple with a deck of cards and a python shell with armoryengine loaded.  I've posted elsewhere how to do this, but I don't say it a lot because I don't want it too easy to find out  Primarily because I'm 200% comfortable with Cryptopp + external entropy, so I don't feel there's any practical reason to prefer that beside enjoying "seeing" the entropy yourself.

[btchris]

Alan, what do you think of "Hash Hyena" claims?

He basically says that he is generating trillions of addresses non stop and getting a few collissions that will grow over time. His logic is that the entropy of Armory, Bitcoin Core, Multibut, Electrum etc. (and generally any other wallet that uses a RNG based on software) is flawed and results in a highly reduced keyspace, which will result in collissions with enought computing power and space devoted to private keys bruteforcing.

See: https://bitcointalk.org/index.php?action=profile;u=380718;sa=showPosts
...

...
On that note, Armory uses Crypto++ was is considered a cryptographically-secure RNG (X9.17 with OS-provided seeding).  On top of that, Armory pulls in system files, mouse clicks, keypresses, and a desktop screenshot, to add to the Crypto++ RNG entropy pool.  I made sure when selecting these sources that it would guarantee at least 256 bits of entropy to be added to the pool even if Crypto++ was really weak.
...

It's rather telling that he claims a "paperwallet" (I assume he means bitaddress.org) has a safer RNG for key generation than Armory (which, among others, has undergone "heavy testing" by his team) because "paperwallet" uses mouse input....

[Deth]

Hi! Is it possible to run a "server" instance of Armory which does synchronisation with bitcoin core and whole network, and connect to it remotely via multiple "client-side" instances of Amory? So I can download and sync blockchain once for all other machines, for example? Thanks and sorry if there is an answer already somewhere

[Adrian-x]

feature request:
> once I double click on a wallet, I get the "Wallet Properties" window.
   (It displays all my addresses)

at the top of this window are 3 check boxes;  1) Hide Empty 2) Hide Change 3) Hide Unused.

The one I want to use is missing 4) Hide Used I would like to see empty but unused addresses but hide the empty used addresses. and make that dealt.

thanks,

[goatpig]

Hi! Is it possible to run a "server" instance of Armory which does synchronisation with bitcoin core and whole network, and connect to it remotely via multiple "client-side" instances of Amory? So I can download and sync blockchain once for all other machines, for example? Thanks and sorry if there is an answer already somewhere

That would be the equivalent to a supernode/litenode paradigm. Supernode is on its way, litenode... well we have a road map but no one working on it yet.

[Deth]

Hi! Is it possible to run a "server" instance of Armory which does synchronisation with bitcoin core and whole network, and connect to it remotely via multiple "client-side" instances of Amory? So I can download and sync blockchain once for all other machines, for example? Thanks and sorry if there is an answer already somewhere

That would be the equivalent to a supernode/litenode paradigm. Supernode is on its way, litenode... well we have a road map but no one working on it yet.

Could you be more specific, please? Where can I read about it?

[goatpig]

Hi! Is it possible to run a "server" instance of Armory which does synchronisation with bitcoin core and whole network, and connect to it remotely via multiple "client-side" instances of Amory? So I can download and sync blockchain once for all other machines, for example? Thanks and sorry if there is an answer already somewhere

That would be the equivalent to a supernode/litenode paradigm. Supernode is on its way, litenode... well we have a road map but no one working on it yet.

Could you be more specific, please? Where can I read about it?

There isn't much to read about it, really. You are asking for a server instance of Armory (aka supernode, tracks all addresses in the blockchain) and a client instance (aka litenode, has no block data, and fetches wallet history from a server instance). That's the plan for the future iterations. Supernode is well on its way, as for litenode, we gonna make a robust supernode release before thinking about that one.

[justusranvier]

Hi! Is it possible to run a "server" instance of Armory which does synchronisation with bitcoin core and whole network, and connect to it remotely via multiple "client-side" instances of Amory? So I can download and sync blockchain once for all other machines, for example? Thanks and sorry if there is an answer already somewhere

btcd has that architecture and is supports the Armory wallet format, but their GUI is currently... minimalistic.

[Winterfrost]

The one I want to use is missing 4) Hide Used I would like to see empty but unused addresses but hide the empty used addresses. and make that dealt.

I've actually been wanting this feature too. I would find it immensely useful and it fits with not reusing addresses paradigm.

[Rogue Star]

The one I want to use is missing 4) Hide Used I would like to see empty but unused addresses but hide the empty used addresses. and make that dealt.

I've actually been wanting this feature too. I would find it immensely useful and it fits with not reusing addresses paradigm.

Thirded, I was looking for just this feature just yesterday and was somewhat surprised it wasn't there for the reason mentioned above.

[Kluge]

What is the progress on the Android client Trace suggested (or reinforced)? Could someone outline its exact purpose and planned features?

(might be worth posting answer here, instead)

[Adrian-x]

[/b]te=1413439215]

The one I want to use is missing 4) Hide Used I would like to see empty but unused addresses but hide the empty used addresses. and make that dealt.

I've actually been wanting this feature too. I would find it immensely useful and it fits with not reusing addresses paradigm.

Thirded, I was looking for just this feature just yesterday and was somewhat surprised it wasn't there for the reason mentioned above.

It's probably missing because of how we use language.
The check boxes there have a direct opposite while the one I would like is more complicated to communicate.
 On second thought it should be: hide used address with a 0 balance.

[Rampion]

Has the latest version (0.92.3) been tested on Yosemite (OSX 10.10)?

[sbrzol]

Last week i tried to use bitcoin qt ( i can see my balance and all transactions ),  but my passphrase did not work there, later i realized (after opened the bicoin.conf file : rpcuser=generated_by_armory) that i installed armory

i did not use Armory more than 4 months , i opened it yesterday but there is no any wallet
where can i find my wallet?   how can i restore it?  
i know my armory passphrase (same as bitcoin qt), but there is no .wallet  file (Appdata/roaming/armory) on my computer to open and recover

or is it possible to recover my wallet  from wallat.dat (bitcoin qt ) and armory passphrase?  

or perhaps i never made any wallet with armory , in the log file i cannot see such (creating) item but i can see  

always 0 :  ArmoryQt.py:2359 - Number of wallets read in: 0

2014-06-26 13:23 (INFO) -- ArmoryUtils.pyc:524 - Executing popen: [u'icacls', u'C:\\Users\\xyz\\AppData\\Roaming\\Bitcoin\\bitcoin.conf', u'/inheritance:r', u'/grant:r', u'xzy:F']  

(it is possible that earlier i used 32bit qt, an now 64 bit, and the database location bitcoinqt and armory changed many times , no enough space on C)

or why does not work my passphrase in bitcoin-qt, if i never created any wallet with armory?  

so how can i get back the access to my wallet ?

[pitiflin]

Hi there,

I'm running the lasts versions of bitcoin-qt and Armory

I want to run Armory through TOR but how can I do it? I know this matter has ben already explained before, but I can't find the pages.

I see that the new version includes an option for TOR.

And second questions, is it secure?

I've had bad experiences with blockchain + TOR and coinbase + TOR losing small amounts with both wallets.

I've read that visiting clearnet sites with TOR isn't a good idea, that's why I ask if Armory + TOR is secure.

I store 90% of my BTC in a cold wallet in Armory, and I don't want to lose them...

Any help would be much appreciated, as I said, I store all the funds in Armory and I'm quite paranoid about security.

Best regards

[picobit]

Last week i tried to use bitcoin qt ( i can see my balance and all transactions ),  but my passphrase did not work there, later i realized (after opened the bicoin.conf file : rpcuser=generated_by_armory) that i installed armory

Armory does not touch the password for the Bitcoin-QT wallet (doing so would be extremely bad behaviour!).  It *does* touch the bitcoin-qt file, but only to set up communication between the two clients.  I think it does set up a password for that communication, but it is totally separate from your bitcoin wallet password.

It indeed looks like from the log file that you never created an Armory wallet.  Your bitcoins must then be in the Bitcoin-QT wallet.  The most likely explanation is that you used another passphase for Bitcoin-QT than the one you are trying now.  Do you have an unencrypted backup or a paper backup of the Bitcoin Qt keys?  Without such a backup, there is no other way to recover than remembering the correct passphase.

[doug_armory]

Has the latest version (0.92.3) been tested on Yosemite (OSX 10.10)?

I haven't tried the version Alan posted. It should work, though. The code has been specifically patched to handle Yosemite. I haven't seen any Yosemite-specific issues come up during testing. That being said, if anybody comes across any issues, please file a ticket.

Thanks!