in fact ewallets will be the vast majority of bitcoin transfers in the future
No, but mobiles with NFC secure elements will be. Carried just like cash, but even more secure. And yes, in a few years even grandma is going to be using what's today called a "smartphone". (I won't answer protests from anyone not having at least researched what the [NFC] secure element spec is ;) There's no reason, even for online wallets, to require my private keys. |
|
|
|
In my case I have two identical clients where one cannot send while the other can. I've even copied the blockchain between them so I know it's not corrupt.
So, just for fun, I redid the exact same transaction I couldn't do with one client with the other one instead. Again, the only real difference is that one is port mapped and has >100 connections, the other is behind NAT with 8. They use the same block chain files, they're up to date and all that. Both clients seem to write 1MB/s constantly to disk (can someone else on Mac please check this? edit: And are you using Filevault) but the well-connected one runs off an SSD which means I've never noticed it before. NAT client: "0/unconfirmed", "0/offline?" - sends are never announced. Tried multiple times to send 2 BTC (old coins), no fee. Mapped client: Sent 1 BTC (old coins), no fee. The announce was instantly visible on bitcoincharts/bitcoin and after a few minutes it could be seen in blockexplorer and had 1 confirm. I guess one explanation lies in how many peers they have, but the NATed client is never even able to make the announce to the network even though it gets block updates just fine. If that's the general state for NAT clients then that needs to be rethought. Else I'm leaning towards an OS X specific problem in .24 that might or might not be related to the weird disk writes. |
|
|
|
I am referring to the case of the original poster, not your followup posts which may or may not be the same issue.
I'm indeed talking about "0/unconfirmed" (even "0/offline?") and never announced, with an 8 peer connected client fully up to date (and getting updates) block chain. The thread starter had a month old transaction. Even with zero fees that would've gone through. It's likely it, like in my case, was never announced. |
|
|
|
I would think that my password should be strong enough ..
I have not touched on my MT Gox account elsewhere in the home and from my office, and both lines are secured with codes.
The vector would be that your email adress, as everyone elses, became known from the MtGox hacking incident. Somehow someone decided it was worth trying to hack your emailaccount - making it possible to do a pw reset on MtGox. Why someone targetted you in particular, and managed to guess your email password, is another question. The most likely cause would be that you signed up somewhere else Bitcoin related and re-used the same pw as you use with your email provider. (If your email account has been hacked you of course need to make sure to clean your account from any hidden forwards and then basically consider EVERYTHING you've ever signed up for as "broken" since the attacker could've used it to gain entry to a lot more than just MtGox) Of course, all the above is based on MagicalTux both being honest about what the logs say and that if the logs say pw reset through email that there's no way to fool the system into doing pw resets some other way. |
|
|
|
Maybe this problem will motivate you to send fees. Like i always do.
If you sent without fees, then you would have used an altered client that doesn't comply with the fee rules
Wait, what? None of the two replies above have anything to do with the issues discussed in this thread. (TX not getting sent at all is not due to fees and fee-less .24 is a configuration option and not something "altered") In my case I have two identical clients where one cannot send while the other can. I've even copied the blockchain between them so I know it's not corrupt. |
|
|
|
A post by Theymos on July 1st, in another MyBitcoin thread: MyBitcoin is still accepting payments with only 1 confirmation. This is insane for a bank. Any miner capable of mining two blocks in a row can steal money from MyBitcoin pretty easily. I'm surprised no one has attempted it yet. - http://bitcointalk.org/index.php?topic=22221.msg309173#msg309173Looking at the reorg log, there may be some truth in Tom's statements. However, I'm not too sure what the proper odds are for how often a reorg should statistically happen, nor do I have the other versions of the blocks listed (my client has been left off for some time). Maybe we could have two threads. One where people are randomly screaming "GOXED" and another where we discuss the above which to me seems really interesting. I assume the reorg-link shows the finally accepted blocks? If so, comparing them to the dropped ones should show large transactions where someone transferred money away from mbc. |
|
|
|
Didn't realize there were so many Gox Apologist in action!
How is my knowledge on northern European English skills being "Gox Apologist"? We can go into more detail if you want to learn more about how hearing perfectly spoken English, but seldom using it in writing, result in people who use american expressions but with grammar and/or spelling mistakes - but I suggest you start another thread for that study. PS: "hagget" is a Danish version of "hacked", an English word having been Danishified. Even a simple Google search would've told you that. |
|
|
|
The phony broken English is totally fake.
It looks perfectly on par for someone from Denmark (and other European countries where people grow up with subtitled american TV series). |
|
|
|
how many connections is it showing and are you using the latest client? what you did should have worked.
Both him and me are using the latest client on OS X - and at least mine has 8 connections and is getting block chain updates. I still believe there's a bug that shows up on some system configurations. I'd be interested to hear if others on Mac are seeing the 1MB/s constant writing to disk as well. It could be a tx-thread gone into a loop. ("Some", since my other client has no issues performing transactions) edit: Possibly related thread about the Mac client and hard drives on the newbie forum: https://bitcointalk.org/index.php?topic=33237.msg415894#msg415894 |
|
|
|
Simply put, this password should not have been guessed or brute-forced on a live system over the Internet.
edit: Speculation superseded by MT's post after this. Agreed. If brute forcing this password was the attack vector then someone has access to the MtGox hashes incl. salt or is able to perform an enormous amounts of live tries towards the API.(J. has stated the pw only ever existed at MtGox, wasn't reused etc etc) |
|
|
|
I'm still interested in exactly what the Bitcoin client is doing writing 1MB/s to disk with no actual file getting any larger though.
I think it is updating the database of peer IP addresses in a stupid way. I never got up to 500 peers but I've noticed that the more connections you have the more frequent the updates. Hmm. The code must be extremely bad if it can reach 1MB/s doing that. I've actually given up on getting this client installation to perform any send transactions. With the new fixed wallet I again tried sending coins and just like last time they stayed at "0/unconfirmed". I even got to see "0/offline?" - this on a client with 8 peer connections and receiving updates to the block chain without issues. (I don't consider the title to be misleading - if we ever want to get non-techies to feel comfortable using Bitcoin with the official client) |
|
|
|
Moral of the story, make frequent backups people!
Yeah I stopped waiting as well now. I have a new automatic & encrypted backup made whenever the wallet file changes so it was easy to roll back (complete version history available). I'm still interested in exactly what the Bitcoin client is doing writing 1MB/s to disk with no actual file getting any larger though. Ah I've missed that, although I do use the excellent pywallet script for wallet manipulation  |
|
|
|
still said "0/unconfirmed" for that transaction, even though I had an updated blockchain.
The address is correct. But, blockexplorer does not show any incoming activity to the address I attempted to send coins to. The coins, however, have been deducted from my balance.
running OS X Snow Leopard. I'm currently running v0.3.24
I'm seeing the same thing with the latest send on one of my computers (tx not announced). On my other (same OS, same version etc) I don't have any problems at all. On the problematic client Bitcoin is constantly writing 1MB/s data to the disk (edit: And not to any file I can see growing in size!). Thinking it was due to a corrupted blockchain I cleaned it up and copied the blockchain from my other client - but it didn't make a difference. Slightly weird. I'm leaning towards an OS X specific bug in .24 that doesn't manifest itself on all configurations. The only difference between my two clients is that one is portmapped (>100 connections) and the other isn't (8). |
|
|
|
I hesitate to mention the obvious.... 1) Most miners live in the northern hemisphere. 2) That summer is HOT!
Really? /Europe |
|
|
|
Doesn't work, I'm getting this Traceback (most recent call last):
File "pywallet.py", line 7, in <module>
from bsddb.db import *
File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/bsddb/__init__.py", line 64, in <module>
ImportError: No module named _bsddb
Do I need a new version of python? And if so, where do I get it?? Looks like you're on a Mac. bsddb is broken in Apple's default installation, the solution is to install another Python version via Macports. |
|
|
|
Any other mac users having this problem and found a solution? It's almost impossible for me to have the mac bitcoin client running as it slows down everything else massively! I'm thinking now of installing a virtual machine to get the windows client up…
On my MBP Activity Monitor shows a constant write of 1-2MB/s to the disk with the Bitcoin app open. That's way way more than needed to perform blockchain activities. |
|
|
|
Correct, so even if your mythical ASIC miners are willing to sell at $2 a bitcoin, doesn't mean anyone else will be
Mythical? You do know that there's at least one person using ASICs since several months back - just as he began using GPUs before everyone else? Anyway, I don't know why you believe Bitcoin wouldn't follow free market rules but the answer is that it would indeed cause a downwards price pressure. Did the price crash when ppl started using GPUs?
Back then people weren't as price obsessed (may - sept 2010) http://bitcointalk.org/?topic=1009.0edit: That is, you cannot see anything in the charts to state either way. http://bitcoinx.com/charts/ |
|
|
|
#1) Many more bitcoins will be in the hands of non-ASIC miners.
Completely irrelevant. The market price is what someone - anyone - sells at. We already know not all mined coins are being sold. #2) No other resources dynamically adjust their difficulty to a fixed rate. If you get a magic gold finding machine, the gold doesn't automatically become harder to find after you use it a few times.
The above is only relevant when everyone is using the same technology. This is the second time that won't hold true. #3) All of the price is dependent on miners nonsense only holds true if other bitcoins were being destroyed after use so the only way you could get bitcoins was to buy from a current miner and if the difficulty of mining bitcoins didn't change with technology. It is impossible to have a major disruption in bitcoin supply. Disruptive technology may quickly drive previously profitable mining operations out of business, but it won't do jack to the BTC price which is determined by buyers and sellers. Some of those sellers are miners, many aren't.
See #1 and #2. |
|
|
|
I agree with you that GPU mining will eventually be made unprofitable as ASIC mining (and botnets) takes over, but that has to do with difficulty adjustments, not price adjustments
Yes, and in the meantime those who switch to ASICs can undercut everyone mining with GPUs and still make a huge profit. That creates a downward pressure on the price of bitcoins. It's the same as in any other resource extracting market, and always happens when there's a disruptive technology shift. I think it will be at least 6 months before you can buy an ASIC miner at all, and it'll be a year minimum before they're cost effective
https://bitcointalk.org/index.php?topic=14910.msg219049#msg219049I don't understand what you mean with cost effective. Of course an ASIC rig will be more cost effective from the first second, the actual ROI is something else and depends as much on guesswork about the future as anything else. You fail to take into account that defxor said it, and therefore it must be true.
Right. Imagining shapes in charts is of course much better  |
|
|
|
By your analysis, bitcoins should have been super expensive when they were being mined by CPUs.
No. You might want to study Bitcoin's concept of difficulty to understand why. You might want to read more than the first sentence of the guy's post before responding. The rest was irrelevant due to the point I singled out. Mining with CPUs wasn't a problem until GPUs came online - and since they were so much more effective that caused a disruptive shift where all CPU based mining stopped. The same will happen in the shift from GPU to ASIC. During that transition, anyone mining with the more effective equipment can sell their coins at great profit much cheaper than everyone else - who will still have to try to offload them though to pay for electricity. Anyone not factoring in the shift away from GPU mining right now aren't doing their homework. |
|
|
|
|
Show Posts
