Topic: ROR (Read 1592 times)

Well since the events happened recently my memory should be considered to be the truth. I didn't know about the other address because I've used it once and 2 months ago.
That did not happen. I sent the transaction. The day after I copied over the Bitcoin folder and removed the HDD. Later that day I opened up Bitcoin Core and set the path to the new folder.

We are comparing my "best guess" against your "best memory" here, so I really don't know which is more likely to be right. However...

The only thing that I can think of that would cause this effect would be if you copied the wallet to disk 2, and THEN sent the transaction while the wallet was still using disk 1.  Then after sending the transaction, the wallet was pointed towards disk 2, and disk 1 was unplugged.

The behavior seen in this thread is EXACTLY what I would expect under those conditions.

Maybe there are other files involved that trigger which addresses are shown or so. Maybe in your old place the transaction and wallet was seeable, then you moved, had other configuration files, and the address including transaction was not shown anymore. Then it was recreated.

Only a guess. I don't know the inner workings of the client so good.

Exactly. I generated the address a day later. That's what I concluded during the investigation and with the helps of others (especially Danny).
No. The wallet.dat was just moved from disk 1 to disk 2 and then the software was automatically (on startup it asked for a location) pointed towards disk 2 and the first one was unplugged.

Sorry, i mixed that then.

So the problem is you received change on an address that only was created later. Though the address had to exist before in your wallet, otherwise you wouldn't had get the change to it.

I think since you copied the wallet.dat that something was happen on the way. I doubt it's a general problem.

I've sent 0.12 to a person and the change which was 0.1799 went to an address that I didn't have at that time. The address that received the change wasn't generated until a day later.
I don't really care about that as this wallet is of low significance. As I've previously stated I didn't even realize that one of the addresses was mine until Danny asked (a bit embarrassing but still).

I wouldn't have made a thread if I didn't suspect an address collision (which it is not).

If you sent the bitcoins to another persons wallet and the coins on the wallet are the change then maybe you are right that bitcoin qt didn't realize that it already had this address as a change address. Can this be? Maybe a question to development subforum.

So you sent 0.179 to the other and the spare 0.12 went to a change address that today is taken as a normal address by your wallet.

Besides... you should reconsider using change addresses as long as you don't want your wallet addresses being easily find out.

Well there isn't much to write in details. I've done the following:
1) I've copied over my whole Bitcoin folder (it was on D:\Bitcoin , not my system drive) to a new external drive
2) I've removed the D:\ drive and renamed the external one to D:\
3) Synced up the wallet
4) Sent the 0.12 to someone last night
5) Opened it up today to sync and generate a new address
6) Made a new thread.

Yes. I didn't check the transaction that I had sent last night on blockchain.info. The address first appeared after clicking "+ new" ~1.3 hours ago

I didn't change anything via any config file.

Ok, so now that we've confirmed that this is not an address collision, the question remains:  What actually happened.

If you want to take the time to write up exactly what you've done with that wallet in the past 36 hours or so in detail, it might be possible to analyze your actions and make an educated guess which step led to these results.

The most likely cause would be recovering a wallet.dat backup.  If you did not recover a wallet.dat backup and you are not using this same wallet.dat elsewhere, then it seems likely that there is either a small bug in the the wallet's use of the address pool, or there is a small bug in the upgrade process (if you upgraded the wallet after the first transaction and before the second).

Just to make sure I understand you: So its a change address from a TX you did in the past (yesterday) and was now shown just recently (~1 hour ago) when you clicked "+ new" in the receiving address window?

What does "getwalletinfo" from the console return for "keypoolsize"? Bitcoin core pregenerates 100 addresses and tries to keep it that way, unless you changed this via the config file.

Okay I am a bit confused on one part, last night did you send .17 btc to an address on purpose or was this a bug from upgrading?  or is the confusion from you upgraded and it is all fine but the blockchain is showing it as happening last night instead of an hour ago?

Sounds like it might be a bug in Bitcoin Core in which it didn't realize it already used an address as a change address.

I guess emotions made me make a thread before actually doing enough research myself. I could have spared others some time, I'm sorry.

Essentially the remaining balance last night went to: https://blockchain.info/address/13EQZzdrBdHciQSg8bGfNr2YnTV9H1Bjmy
however, I have generated that address about 1 hour and 10 minutes ago.

So if you sent the coin last night, what address did you send it to? Since the current address was just generated 1 hour ago? What happened to last nights transaction, is it on the blockchain?

I've realized that a bit too late.

Well I've posted without doing enough research myself, unfortunately. That address is mine, and I've just realized this. I'm the one that sent the 0.12 BTC last night, however the times do not match because of time zones (probably) and that got me confused.
However, this still doesn't explain why the 0.179 went to an address that I've generated 1 hour ago? Could there be a bug in Bitcoin Core so that it had stopped showing the address temporarily or while upgrading from 0.9.x to 0.10.2?

This rules out an address collision.

By the way, in the future when you want to assistance with something like this, I suggest that you create a "self-moderated" thread.  That way you can keep the conversation on track and delete all the nonsense posts from people that aren't paying attention to what you are saying.

What version of electrum Core are you using and possibly they have a bad generation method going on that needs to be updated and fixed asap? Maybe a bug slipped by, I just can't see a random collision especially this early in bitcoins life and the recent activity.

You said the latest but that does include beta builds?

Looks like Bitcoin.org has Latest version: 0.10.2
and github has 0.11

Which one are you using?

Can you please run:


And see if your wallet has the private key for that address?

DO NOT SHARE the private key with ANYBODY, just let us know if the wallet finds one or not.

Just had to ask, before the experts come in and start analyzing your wallet program.

It is just too weird that it (the suspect address) was just used yesterday. That's why I think it is not a collision.

Edit: added (the suspect address)

Well one can never be sure with his own memory but:
1) I do not recall using the wallet yesterday aside from sending a payment to someone
2) It doesn't show up in the transactions
3) I don't recognize the address that sent the Bitcoins to this one either; 


Also I was able to successfully import the address into Electrum and the balance has showed up.

This might be an extremely stupid question, but:
are you sure you did not generate this address yesterday and send that transaction?

Oh.. I didn't check this. In this case I have no idea what could be the reason...

But if it were an address collision... it would most definitely be a once-in-a-lifetime event.

ya.ya.yo!

Okay, I've verified that there is no copying mistake. I've verified multiple times and I'm going to try importing that address into another system using electrum.


I did not check that.

This is not strange at all.

blockexplorer is not up to date.  If you look at their main page, the last block that they have processed is block #358999.  This transaction was confirmed in block #363102.

Actually about 3 months ago I used vanitygen to generate an address 1LaudaexFYZano87acDNGscmqDbM3ZTLBD which I've used to sign a message for this account. So this is a pretty new wallet (time wise), and I've been using it pretty much every day.
Nobody is using the same wallet.dat, nor has anybody access to my computer or the files. Besides the wallet is encrypted.

I'm using Windows 7 Ultimate 64 bit with the latest updates to date. Also for the record, I was able to extract the private key from that address (my balance did not increase even though the address has 0.179 BTC).

This is very strange. So basically we have 2 explorers that are showing the address with a balance and 1 that shows it without.

It exists in blockr.io though.
http://btc.blockr.io/address/info/13EQZzdrBdHciQSg8bGfNr2YnTV9H1Bjmy

An address collision would be extremely unlikely...

The most reasonable assumption would be a bug on blockchain.info. This assumption is confirmed by using blockexplorer: https://blockexplorer.com/address/13EQZzdrBdHciQSg8bGfNr2YnTV9H1Bjmy

ya.ya.yo!

This seems very unlikely, however, there are a few scenarios where you could create a situation that would have this result.

Some questions:

Is this a brand new Bitcoin Core wallet that you just installed, or did you recover this wallet.dat from somewhere?  If you had recovered this wallet.dat from somewhere, then this address might have been used from the same wallet.dat AFTER the backup was created.  As such, when the backup was recovered, it is possible that the transaction was forgotten about.

Are you (or anyone else) using the same wallet.dat in multiple Bitcoin Core wallets?  If you created a copy of this wallet.dat and ran a second instance of the Bitcoin Core wallet on some other computer (or some other session on the same computer), then the other instance of the wallet may have pulled this address from its pool already, and now this instance is jsut getting around to pulling the same address from the address pool in the wallet.dat.

What operating system are you using, and what version of that operating system?  It is possible that the random number generator of the operating system that you are using is faulty (or compromised) and that it is not generating actual random numbers.  As such, if someone else is using the same faulty (or compromised) operating system, then they could generate the same address.

Are you certain that you copied and pasted the correct address?  Perhaps the copy failed, and you pasted a previous address that was still in your computer's clipboard.  If the address actually received bitcoins yesterday, then they should be added to your balance now.  Did the balance on the wallet increase?

Can you run:


in the console of your Bitcoin Core wallet?  Does that address (and transaction ouput) show up there?

I've created a new address using the latest Bitcoin Core client. A few seconds after I generated it, I went to blockchain.info and pasted it there.
The address that I linked is the one that I have generated today (minutes ago).

you are saying that you created a bitcoin address from a wallet like electrum and then you find out that there are bitcoins on it?!

--- removed on request ---