Topic: Blockchain split of 4 July 2015 - page 13. (Read 45589 times)

Like I said, I am in the process of starting my own electrum server that I know I can personally trust. As of now, I am in the process of downloading the blockchain for my full node, and some level of research needs to be done before I can create my electrum server that relies on my full node (I had found a thread advising how to create an electrum server, however it seems that it either contains some mistake or I am doing something wrong - I am very much a newbie when it comes to linux/ubuntu, so it could easily be either one).

As you can see, the price wasn't affected by it at all.


On the other hand, it also shows that the system still has errors that can happen.
Don't get me wrong,i am glad that the "issue" was resolved, but i find that it shouldn't have happened in the first place.

cheers

One way to think of this is that a group of Chinese miners just gave over $50K away to other miners, to reward those other miners for checking the block versions before building new blocks.

Terribly nice of them, really.  Though I'm sure they'd prefer to've collected the prize themselves, it's nice of them to "support" those who are doing a better job than they are.

So, for a few hours, if you were mining and actually checking blocks for validity, you were getting about double your usual rate of return.

"SPV mining" is one of those silly ideas like "virtual memory" or "virtual sex" - the real thing is always better.

Can someone advice me if this split would affect the price of bitcoin anyhow? If yes how?

Any word on whether Blockchain.info wallets are relying on Core, or something else? Silence from their Twitter for the last couple days. I was hoping to see something on their homepage. I have a small hot wallet there.... only use it to hold a couple coins at any time.

Downloading the blockchain as we speak.... (sigh)

It affects the users who are using Bitcoin Core 0.9.4 or earlier versions, Lightweight (SPV) wallet, Web wallet users. No matter for the receivers or senders, they are both affected!

is it over now 

Some SPV servers were relying on 0.10.x so they were unaffected by the issue, and others were not and were effected. My personal experience with electrum on this incident is that I had a transaction that was sent to me that was showing as unconfirmed for a long time, then all of a sudden it had in excess of 6 confirmations. I am not sure what exact conclusions that I can draw from this except that the particular electrum server that I was connected to at the time was not relying on core 0.10.x as of prior to this indecent.

I also believe that by default, the electrum server that you connect to is random, so it is hard to make the determination if your server is running a "good" version of core or not.

My personal solution is to start a full node, and once the blockchain is downloaded, to run an electrum server that relies on the most recent version of Bitcoin core

but its affect only the transactions incoming from a user with bitcoin core? or between webwallets too? thank u

I imagine we don't know nothing about their servers status. Unless wallets providers will tell us directly that the crisis is averted. That is the problem with the online wallets or like in this case simplified validation wallets, if you don't have copy of blockchain on your machine, things like that could happen and you can't do anything about it. Power in the hands of wallet provider.

How do we know if electrum and mycelium servers have been updated?

looks like a ponzi made to look like multi level marketing pretending to facilitate person to person money transfer.  im skeptical.  also your post is off topic here and may be spammy but i will reply if it stops you from losing any money.

Are blockchain.info web wallets running Bitcoin Core 0.9.5 and later? Just want to know if I should wait for 30 confirmations.


No, this issue concerns web wallets too.

36% is too centralised, seems like its a problem even when not apparent that BTC has become too much about bigger is better.  Will it converge with traditional banking on the same basis, maybe its meant to be
Its a mistake to me, if you want to represent capitalism then the capital value and/or means of production should be within the population ideally

I think... this is just for Bitcoin core, as written on the News above

Is this going to be a major problem ? are blockchain.info wallets safe in the moment? thanks a lot

this may have already been posted these links (op feel free to remove this post if you wish)


https://bitcoin.org/en/alert/2015-07-04-spv-mining


https://en.bitcoin.it/wiki/Comparison_of_mining_pools#SPV_Mining_.2F_Old_Bitcoin_Core



hope it helps ...again if a repeat of previous info feel free to move/delete or whatever

As far as I understand, the incident had some consequences:

* several miners lost the rewards that they mined (at least 6 blocks, maybe more).
* there was a signficant risk of double-spends, since many ciients were watching the bad chain.
* at least one transaction (the one that triggered the fork) was undone after >6 confirms.
* a warning was broadcast telling everybody to wait for 30 confirmations (5 hours).

The incident was not as bad as it could have been because Greg Maxwell and other ore devs were carefully watching the blockchain as the BIP66  triggered the switch from version v2 to version v3 of the protocol, and they immediately spotted the fork.  They soon contacted the mining pools that were mining on the wrong chain.

The fork was started by miner BTCNuggets, that was still running v2 and issued a block N with a transaction that was invalid under version v3.  Miner F2Pool, that was already running v3, saw that block and successfully mined another block N+1 on top of it, not realizing that N was invalid.  Other v3 miners like AntPool added N+2, N+3, etc. on top of F2Pool's block N+1.  Meanwhile some v3 miners did not see the block N issued by BTCNuggets, or noticed that it was invalid under v3 rules and rejected it; so they created their own block N and kept mining that branch.  Until the miners were alerted, the bad branch was longer, so it was accepted by all v2 clients and maybe by v3 clients that did not check all blocks.

Blame is being thrown at the miners, especially F2Pool for mining block N+1 on top of a block N that was produced by a v2 miner, without checking the validity of that block.  The blame may be partly deserved if F2Pool got the block directly from BRCNuggets or from any intermediate relay node still running v2.

However, methinks that the blame should fall on the developers.  After the BIP66 change went into effect, every v3 player (client, node, or miner) should have dropped all links to other v2 players.  Letting v2 and v3 players communicate was begging for this accident to happen.

If F2pool got that block through an intermediate v3 node, the blame should fall on that node.

Miners should not be required to validate the blocks that they receive.  It is their right to decide whether to validate them, or to trust that the sender did so.  Note that the miner that started mining block N+3 on top of N+2 would have had to check also blocks N+1 and N to discover that N+2 was invalid.

Was the incident serious? Suppose that a trash basket caught fire in a gunpowder factory, but it was quickly put out.  was that incident serious?

My wallet is with blockchain.Info. if I am expecting an incoming transaction does this still apply?

Are you talking about certain types of transactions from unknown sources? 

I just checked the blockchain (because again waiting dozens of minutes for confirmation) and this is related, right?