Search Posts

Post

Topic

Board Development & Technical Discussion

Re: MacOS X LevelDB Corruption Bounty (5.51 BTC + 200.2 LTC)

whault

on 20/11/2013, 00:40:38 UTC

Some observations. My setup uses two drives, one with the OS and a lower speed one for general storage. I don't use time machine like the poster above, and there's nothing else non-standard about my software.

only the blockchain stored on the internal SSD boot disk gets corrupted, a blockchain stored on the second SATA HDD is never corrupted
corruption seems to happen most often after a system sleep (deep or not), though not always
corruption can happen during the initial sync if it is stopped and then restarted
corruption can happen with FileVault 2 turned on and off
has happened less often since updating to 10.9 only twice so far instead of every few days, though it could just be chance

That's it really. No other behaviour is specific to corruptions for me. Sometimes they happen twice in a day, sometimes not for weeks.

Post

Topic

Board Project Development

Re: [BOUNTY] Help test next major release of Armory! [0.04 BTC/bug]

whault

on 16/11/2013, 07:32:39 UTC

Quote from: etotheipi on November 16, 2013, 07:05:13 AM

Quote from: whault on November 16, 2013, 07:00:14 AM

I don't promote the feature much, because it's minimal security improvement. But it does protect against the simplest of keyloggers. There's this notion that all viruses are omniscient, and targeting your bitcoin wallet. If it's targeting your wallet, the OSD keyboard probably won't help. But if it's just sucking data off your system looking for banking login passwords, etc... you avoided giving away too much info.

But I agree the feature is really quite minimal in terms of its security benefit. (and generally, I try to assume omniscient malware when thinking about security, but not everything has to be super-effective to be part of the interface)

I'm more inclined believe that systems like this do more harm than good, they promote a sense of security in what is a fairly sensitive environment. If the protection is just against casual un-targeted key logging, then I'm not sure of the point, having a password is useless if you don't have the associated Armory wallet binary.

Even more importantly, people who habitually use an onscreen keyboard to type their password (let alone one with a non-standard layout) will ultimately be choosing much weaker passwords as a direct result. This would be detrimental in the situation where malware manages to take the wallet binary but not the password.

Having the feature there implies that it will afford the user security, meaning that people may lapse in believing that will be be an end-all protection from malware, when it's most obviously not.

Post

Topic

Board Project Development

Re: [BOUNTY] Help test next major release of Armory! [0.04 BTC/bug]

whault

on 16/11/2013, 07:00:14 UTC

Quote from: duxZero on November 16, 2013, 05:14:41 AM

Didn't even notice, my bad.

But yeah just a suggestion to offer a somewhat more advanced security lockdown to help 'prevent' malware.

If your machine is compromised, nothing will save you from having the contents of your wallet stolen. Onscreen keyboards, "scrambling" and checking the clipboard are just feel-good measures and offer no tangible security. At best they offer some obscurity, but it's only a matter of time before Armory targeted methods arise.

Post

Topic

Board Project Development

Re: [BOUNTY] Help test next major release of Armory! [0.04 BTC/bug]

whault

on 16/11/2013, 04:42:16 UTC

Quote from: duxZero on November 16, 2013, 02:44:40 AM

Bitcoin 'stealers' have a feature that when a bitcoin address is copied to the clipboard it is replaced with the 'hackers' address.
With saying that i think you should add a warning system to alert the user when this happens with any address's linked to Armory.

If malware has enough access to change the contents of your clipboard, then verifying the contents of it will not in the slightest save your wallet.

Post

Topic

Board Project Development

Re: [BOUNTY] Help test next major release of Armory! [0.04 BTC/bug]

whault

on 15/11/2013, 06:10:05 UTC

Quote from: etotheipi on November 15, 2013, 04:04:11 AM

Quote from: whault on November 15, 2013, 04:01:45 AM

Quote from: etotheipi on November 15, 2013, 03:46:08 AM

OSX errors: frustrating I might not be able to claim 10.9 support yet. I think I have to limit it (and bug bounties) to 10.8.

The print issue predates Mavericks, I can replicate it on 10.8 in both the latest stable and the latest beta.

I completely understand your reluctance to support 10.9, it's foreign territory especially as you're not a native developer.

Oh, that's right... I do remember having to do some special stuff on OSX with the print dialog. But I seem to remember that it was fixed with the OSX-specific code... I guess not?

Seems not. You can open up the PDF preview and print from there if you are extremely fast (race against the window freezing), but if you manage it Armory will crash and lose it's working database.

Post

Topic

Board Project Development

Re: [BOUNTY] Help test next major release of Armory! [0.04 BTC/bug]

whault

on 15/11/2013, 04:01:45 UTC

Quote from: etotheipi on November 15, 2013, 03:46:08 AM

OSX errors: frustrating I might not be able to claim 10.9 support yet. I think I have to limit it (and bug bounties) to 10.8.

Post

Topic

Board Project Development

Re: [BOUNTY] Help test next major release of Armory! [0.04 BTC/bug]

whault

on 15/11/2013, 03:23:29 UTC

Quote from: etotheipi on November 15, 2013, 03:12:20 AM

From the terminal, you can just "cd /Applications; open Armory.app"

You probably want to direct users to run this instead:

Code:

/Applications/Armory.app/Contents/MacOS/Armory

The command you have will execute the app but not print any log output back to the terminal.

Quote from: etotheipi on November 15, 2013, 03:12:20 AM

Can you please run from the terminal and confirm that the printed error messages look like this:

There's a lot of output like that, mainly when in the wallet properties view.

Code:

2013-11-15 14:18:28.350 Python[22287:507] modalSession has been exited prematurely - check for a reentrant call to endModalSession:
2013-11-15 14:18:28.830 Python[22287:507] modalSession has been exited prematurely - check for a reentrant call to endModalSession:
2013-11-15 14:18:28.846 Python[22287:507] modalSession has been exited prematurely - check for a reentrant call to endModalSession:
2013-11-15 14:18:28.847 Python[22287:507] -[NSApplication runModalSession:]: Use of freed session detected. Do not call runModalSession: after calling endModalSesion:.
2013-11-15 14:18:38.313 Python[22287:507] modalSession has been exited prematurely - check for a reentrant call to endModalSession:
2013-11-15 14:18:38.328 Python[22287:507] modalSession has been exited prematurely - check for a reentrant call to endModalSession:

For the crash in the print dialogue:

Code:

/Applications/Armory.app/Contents/MacOS/Armory: line 16: 22555 Segmentation fault: 11 $DIRNAME/Python $ARMORYDIR/ArmoryQt.py

The full crash log has been DM'd to you.

EC Calculator:

The labels in this view are a little squashed:

http://i.imgur.com/DKIs5MF.png

Clicking the "address book" icon to select a private key hangs the app.

Code:

(ERROR) armoryengine.py:12767 - Waited 20s for addrbook to be returned. Abort
(ERROR) armoryengine.py:12768 - ID: getTxByHash (86700430)
(ERROR) Traceback (most recent call last):
  File "/Applications/Armory.app/Contents/MacOS/py/usr/lib/armory/qtdialogs.py", line 10549, in execAddrBook
   dlg = DlgAddressBook(parent, parent.main, targWidget, defaultWlt, actionStr, selectExistingOnly, selectMineOnly)
  File "/Applications/Armory.app/Contents/MacOS/py/usr/lib/armory/qtdialogs.py", line 10213, in __init__
   self.setAddrBookTxModel(defaultWltID)
  File "/Applications/Armory.app/Contents/MacOS/py/usr/lib/armory/qtdialogs.py", line 10340, in setAddrBookTxModel
   self.addrBookTxModel = SentToAddrBookModel(wltID, self.main)
  File "/Applications/Armory.app/Contents/MacOS/py/usr/lib/armory/armorymodels.py", line 807, in __init__
   for abe in TheBDM.getAddressBook(self.wlt.cppWallet):
TypeError: 'NoneType' object is not iterable

Traceback (most recent call last):
  File "/Applications/Armory.app/Contents/MacOS/py/usr/lib/armory/qtdialogs.py", line 10549, in execAddrBook
   dlg = DlgAddressBook(parent, parent.main, targWidget, defaultWlt, actionStr, selectExistingOnly, selectMineOnly)
  File "/Applications/Armory.app/Contents/MacOS/py/usr/lib/armory/qtdialogs.py", line 10213, in __init__
   self.setAddrBookTxModel(defaultWltID)
  File "/Applications/Armory.app/Contents/MacOS/py/usr/lib/armory/qtdialogs.py", line 10340, in setAddrBookTxModel
   self.addrBookTxModel = SentToAddrBookModel(wltID, self.main)
  File "/Applications/Armory.app/Contents/MacOS/py/usr/lib/armory/armorymodels.py", line 807, in __init__
   for abe in TheBDM.getAddressBook(self.wlt.cppWallet):
TypeError: 'NoneType' object is not utterable

Sometimes after a crash when doing a full rescan, the app will crash out again with the output:

Code:

-ERROR - 1384487000: (StoredBlockObj.cpp:1063) Cannot get tx copy, because don't have full StoredTx!
-ERROR - 1384487000: (StoredBlockObj.cpp:1063) Cannot get tx copy, because don't have full StoredTx!
/Applications/Armory.app/Contents/MacOS/Armory: line 16: 23075 Segmentation fault: 11 $DIRNAME/Python $ARMORYDIR/ArmoryQt.py

This crash log has been DM'd to you as well.

Post

Topic

Board Project Development

Re: [BOUNTY] Help test next major release of Armory! [0.04 BTC/bug]

whault

on 15/11/2013, 03:10:02 UTC

OSX 10.9, general feeling is that the app is not ready for prime-time. There's a number of crashes that I've encountered, this is the only one I can reliably trigger though.

Print wallet:

Opening "wallet properties" > "backup this wallet" > "create paper backup" will open a standard OSX print window, after about two seconds the window loses focus and will ignore all clicks. Clicking a button before the window loses focus will sometimes cause Armory to crash.

http://i.imgur.com/sVgSxwu.gif

Post

Topic

Board Computer hardware

Re: [Sell][in stock] HashBuster ASIC Miners

whault

on 30/10/2013, 23:02:32 UTC

Quote from: wttbs on October 30, 2013, 08:37:50 AM

I think you should conatct hashbuster, if it is a hardware default I am sure they will replace the broken device.

Happy ending. They got back to me and are refunding the defective device, even though there's nothing really forcing them to. My remaining HashBuster is perfectly functional, pumping out 2.8GH/s onto Eligius.

Post

Topic

Board Computer hardware

Re: [Sell][in stock] HashBuster ASIC Miners

whault

on 30/10/2013, 07:19:07 UTC

Quote from: wttbs on October 30, 2013, 07:15:21 AM

Which operating system do you use and which version? Windows ? Ubuntu ? Debian ? other Linux dist.

The one that isn't burnt up is attached to a Debian box.

Post

Topic

Board Group buys

Re: [20 Left] New ASICMiner Block Erupters - .11 BTC or less. Fast cheap shipping

whault

on 28/09/2013, 08:43:17 UTC

International shipping?

Post

Topic

Board Auctions

Re: Auction - 15 Drillbit 8 Boards from Batch 1

whault

on 28/09/2013, 03:52:20 UTC

Quote from: whault on September 28, 2013, 02:32:38 AM

1 @ 3.1 BTC

Retracted.

Post

Topic

Board Auctions

Re: Auction - 15 Drillbit 8 Boards from Batch 1

whault

on 28/09/2013, 02:32:38 UTC

1 @ 3.1 BTC

Post

Topic

Board Beginners & Help

Re: btcguild transfer fees

whault

on 30/07/2013, 12:06:13 UTC

Quote

Pay Per Last N Shares groups shares submitted by all users into a "shift." A shift is currently defined as ~40 million shares. Once a shift is completed, it is considered an "Open Shift". After 10 new shifts have completed, the oldest open shift is closed.

While a shift is open, any blocks the pool finds are paid out to all open shifts (10% of the block per shift). This means that you will continue receiving payments on completed (open) shifts even if you stop mining.

With PPLNS, each open shift is paid the following:
((Block Value + Transaction Fees) / 10) - Pool Fee (3%)

This amount is then split evenly among all ~40 million shares submitted during the shift.

PPLNS involves variance: If the pool is lucky, you can earn a lot more than normal. If it's unlucky, you will earn a lot less. In the long run (a few months or more), you can expect to earn approximately 5% more using PPLNS compared to PPS.

They have a minimum payout of 0.01BTC on top of that.

Post

Topic

Board Beginners & Help

Re: Is the Bitcoin ecosystem sane?

whault

on 30/07/2013, 11:56:53 UTC

Quote from: Lionel on July 28, 2013, 12:15:55 AM

First, the main purpose of Bitcoin is to safely hide your money from the eye of the governement who overwhelmes you with taxes and halves your income.
Actually you may legally run a business making lots of BTC without declaring it to the governement, because for them you are not getting actual money so you don't have to declare it.

I wouldn't say that everybody has that goal. I have no particular qualm paying tax, and indeed happily paid tax on my mined coins. Worth it if only for the confusion on my tax accountants face when I tried to explain it all.

Post

Topic

Board Beginners & Help

Re: Open the Blockchain in a Database

whault

on 30/07/2013, 11:46:53 UTC

The blocks are stored in something called LevelDB, which is already a very capable database format. While you wouldn't have any fun opening it up in Excel (it's 10GB afteral), you can use a python application named Abe to throw it all into a Postgres/MySQL database if you really insist. I would personally advise against it though.

https://github.com/jtobey/bitcoin-abe