Yes, I shut down the daemon using CTRL+C, backed up the wallet and tried to kick it back on when I received the failed message. I guess this could be where the problem occurred.
Corrupt Masternode Walet after RC3 soft fork
- Thread starter blackempress
- Start date
Propulsion
The buck stops here.
- Feb 26, 2014
- 1,008
- 468
- 183
- Dash Address
- XerHCGryyfZttUc6mnuRY3FNJzU1Jm9u5L
How were you able to shut it down with CTRL+C? As far as I'm aware, if you start darkcoind with the daemon flag, to shut it down properly, you would need to issue the command "Darkcoind stop"
Also when I first loaded the new soft fork I did this:
./darkcoind masternode list
And got an empty hash
{
}
./darkcoind masternode list
And got an empty hash
{
}
As long time supporter and being software dev, why on earth did you not join testnet, but ran a test on mainnet?
Not trying to pick a fight, truly! But you're not making sense. No testnet, dont agree with the absolutely logical and understandable closed-source, for a "massive loss" you dont even have private keys dumped, you didnt encrypt the wallet, c'mon!
You cannot stop the daemon with CTRL+C. Its "darkcoind stop".
You are the first and only to experience such phenomena.
Could please stop antagonizing an obvious community desperately trying to help you?
Not trying to pick a fight, truly! But you're not making sense. No testnet, dont agree with the absolutely logical and understandable closed-source, for a "massive loss" you dont even have private keys dumped, you didnt encrypt the wallet, c'mon!
You cannot stop the daemon with CTRL+C. Its "darkcoind stop".
You are the first and only to experience such phenomena.
Could please stop antagonizing an obvious community desperately trying to help you?
My point was only that the way you present yourself matters. Given the tone and hysteria of the post (not to mention it was in the wrong forum) it had little chance of getting a pleasant response. And yes, in any community, there are large number in it for the money. To use this as a drawback is rather, like you always say, absurd.
Last edited by a moderator:
No matter what you say next, you either running stable version and not RC, or your darkcoind.conf has masternode=0 instread of masternode=1, both a clear indication almost nothing you say makes sense.
Propulsion
The buck stops here.
- Feb 26, 2014
- 1,008
- 468
- 183
- Dash Address
- XerHCGryyfZttUc6mnuRY3FNJzU1Jm9u5L
I'm almost positive the corruption occurred from not stopping the daemon properly.
Just because I didn't run nodes on the test net does not mean I was not participating in other ways. Also first to report such a phenomena does not mean I'm the only one, the wallet could have been corrupting in other situations but no one reported it because they didn't lose coins.
I'm not saying you are incorrect but I have never had this issue arise any other time.
So the private key would have only been in my ram and never written to the disk?
I will stop to "disrupt" your thread, but before that a small reminder from the documentation for all users (advanced like the OP or not) :
"Backup wallet.dat. Frequently backup after, this is BETA software."
PS : Don't try to copy wallet.dat while the client is running
"Backup wallet.dat. Frequently backup after, this is BETA software."
PS : Don't try to copy wallet.dat while the client is running
Last edited by a moderator:
Propulsion
The buck stops here.
- Feb 26, 2014
- 1,008
- 468
- 183
- Dash Address
- XerHCGryyfZttUc6mnuRY3FNJzU1Jm9u5L
I'm not well enough versed in the daemon to answer that. eduffield or another developer could answer that.
If you press <CTRL><C> on linux you send a request to kill the process. This can be intercepted by the process (darkcoin) to cleanly shutdown itself or ignored (you get a error message back)
<Explanation>
superuser*com*questions*262942*whats-different-between-ctrlz-and-ctrlc-in-unix-command-line
I don't know the behaviour of the wallet. You can check if darkcoin isn't terminated/it's running with
ps aux | grep dark
If its running you can bring it back and cleanly shutdown/and save/create hopefully a working wallet.dat.
I didn't used my linux knowlede for years and it is little bit aged. You can get the console back with fg (put program in foreground) and bg (sends program to backgound). Please see above ecxplanation, hopefiully it helps you.
Best Wishes
hrodeberht
Last edited by a moderator:
The darkcoin instance is definitely shut down by now, I have dd'd my mem in the hopes I might be able to find it in there but I can't think of anything else to do.
I'am not allowed to post links and the filter is really good.
replace the first * with . and the fololowing with /
replace the first * with . and the fololowing with /
A good idea would to make a backup of the swap partition/file, too. Hopefully one of the developers/experts is able to recover the private keys from mem/swap (if they are stored there).
I grabbed the swap too, I went through the physical memory and didn't find anything useful. Hopefully I will have better luck in the swap. Perhaps if evan gets back to me he can least give me an idea if it would be possible to recover.
It appears that the problem wasn't darkoin, but with the improper method of shutdown using CTRL-C. I know from personal experience is switching between GUI desktops and command line terminals that I sometimes accidently hit CTRL-C in the terminal when I meant to copy text from the terminal...and that it does NOT shut down darkcoind.
I have an idea, but am not sure how to implement it. Is there a way to get darkcoind to output the private key of an address in your own wallet? If this were the case, it would be a trivial task to output and save such private key (someplace safe) BEFORE sending DRK to the address. Then, even in the event of catastrophic failure, you would be able to recover your DRK.
Any thoughts?
I have an idea, but am not sure how to implement it. Is there a way to get darkcoind to output the private key of an address in your own wallet? If this were the case, it would be a trivial task to output and save such private key (someplace safe) BEFORE sending DRK to the address. Then, even in the event of catastrophic failure, you would be able to recover your DRK.
Any thoughts?
Last edited by a moderator:
I tried loading an older wallet file I was using for software I'm building with the new bitcoind. After closing it out and attempting to open it, it is now it is also corrupted in the same way. All zeroes and is the same file size as the one I sent 1k DRK too. Something very weird is going on, I have a working backup of that wallet file so something was lost, but it is very strange occurrence.
Propulsion
The buck stops here.
- Feb 26, 2014
- 1,008
- 468
- 183
- Dash Address
- XerHCGryyfZttUc6mnuRY3FNJzU1Jm9u5L
Based on this alone, I think it is now safe to say that the issue isn't from the Darkcoin daemon.
Have you checked "binary wallet compatibility" ?
https://github.com/darkcoinproject/darkcoin/blob/master/doc/build-unix.md
libdb break compatibility.
Please report back when you have an idea what might be happening. Posts like "Basically due to this new update I got fucked out of 9700 dollars." seem to have caused a great deal of alarm, and if it turns out that what happened isn't the fault of the update at all it would be nice to undo as much of the damage as possible as soon as possible.
I'm not sure what caused it yet, I'm still talking to Evan about trying to find the core dumps. I'm still going through the logs and trying to figure out what happened, and why the new address didn't write anything to the wallet file and why my old wallet file had 0's written over it.
I did the same things I do with the clients I compile from source and never had these issues and this was the first time I used the precompiled client so I don't know if that was involved with the crash that left my wallet file zeroed out.
Of course I should have checked my backups before transfering coins and I should have transferred a smaller amount first but those don't get to the root of what caused this. This machine has definitely been used for other crypto clients without issue.
I will update the thread as I get more information from my logs.
I did the same things I do with the clients I compile from source and never had these issues and this was the first time I used the precompiled client so I don't know if that was involved with the crash that left my wallet file zeroed out.
Of course I should have checked my backups before transfering coins and I should have transferred a smaller amount first but those don't get to the root of what caused this. This machine has definitely been used for other crypto clients without issue.
I will update the thread as I get more information from my logs.
Yes, very important: Never ever mix clients built against different libdb versions (e.g. libdb-4.8 and libdb.5.1) - the db-format is incompatible, will break your wallet.dat and leave it corrupt.
I was just browsing some random historical threads and found this archived: https://www.reddit.com/r/DRKCoin/comments/2955u5/tried_to_setup_masternode_and_lost_1000_drk_coins/ The address in question seems to be (given by OP a few posts from the end): https://explorer.dashninja.pl/address/Xcp5aAE56XpB19DkUjTvyptvGKGrZPv4cs and it looks like amazingly the 1,000 DASH was recovered 4 days latter, it would be interesting to hear how as the threads don't seem to relate that, just curious, sorry for the necro.