Will we hard fork again for another serious bug?

dooglus

legendary

Activity: 2940

Merit: 1333

Quote from: Blazr on August 06, 2015, 09:38:18 AM

On August 15th 2010, somebody found a buffer overflow exploit in the wxBitcoin client (which is now called Bitcoin Core) and used it to create 184,467,440,737 BTC (184 billion) out of thin air (https://bitcointalksearch.org/topic/strange-block-74638-822 ). Just 5 hours later a new version of the Bitcoin client was published with a fix but in addition the blockchain was hard forked and the offending tx removed from it.

If a serious issue like this was ever found in the Bitcoin Core client again, would the offending txs be removed again? if so, then why doesn't this happen when bugs are found in other clients?

Does Bitcoin Core have some kind of privilege in this sense? or was this just a once off thing, or did the small bitcoin community at the time manage to get everyone to agree to have this tx removed?

I think you're looking at it wrong.

The blockchain forked when the transaction creating the 184 billion BTC was mined. "Correct" clients will have rejected that block and started building their own correct fork, while Bitcoin Core went off on an invalid chain.

Fixing the Core code and forcing it back onto the correct chain resolved the fork that was caused by the bug; it didn't create a fork.

It's possible that nobody found a block using anything but code derived from Core, and so the correct branch didn't get to diverge from the incorrect branch, but that's beside the point.

unamis76

legendary

Activity: 1512

Merit: 1012

If a flaw in the software required a hard fork, there will be a hard fork. As said before, I don't think consensus was needed to fix a bug that violates something so basic in Bitcoin. If something similar ever happens, there will be a fork straight away.

seedtrue

legendary

Activity: 963

Merit: 1002

Yes, if there is ever another issue serious enough to create the need to rollback the blockchain and continue on another fork, it can and will be forked. This is a good thing keep in mind because if leaves power to combat malicious attempts.

achow101

staff

Activity: 3458

Merit: 6793

Just writing some code

Quote from: Blazr on August 06, 2015, 09:38:18 AM

On August 15th 2010, somebody found a buffer overflow exploit in the wxBitcoin client (which is now called Bitcoin Core) and used it to create 184,467,440,737 BTC (184 billion) out of thin air (https://bitcointalksearch.org/topic/strange-block-74638-822 ). Just 5 hours later a new version of the Bitcoin client was published with a fix but in addition the blockchain was hard forked and the offending tx removed from it.

If a serious issue like this was ever found in the Bitcoin Core client again, would the offending txs be removed again? if so, then why doesn't this happen when bugs are found in other clients?

Does Bitcoin Core have some kind of privilege in this sense? or was this just a once off thing, or did the small bitcoin community at the time manage to get everyone to agree to have this tx removed?

I don't think Bitcoin Core has this kind of privilege really. The offending txs were removed not because they were made in Bitcoin Core but because it was a massive consensus breaking bug that caused the problem. It will probably happen with other clients too if something comes from those clients that breaks consensus or completely messes with the blockchain like making 184 Billion BTC out of thin air.

Quote from: LiteCoinGuy on August 06, 2015, 10:23:12 AM

but i doubt that such a major bug will happen again.

I wouldn't count on it. Heartbleed went for 2.5 years before being discovered. The shellshock vulnerabilities have been around ever since bash was first created in 1989. Even if we don't know about a vulnerability doesn't mean that they don't exist.

DannyHamilton

legendary

Activity: 3528

Merit: 4945

Quote from: Blazr on August 06, 2015, 09:38:18 AM

If a serious issue like this was ever found in the Bitcoin Core client again, would the offending txs be removed again?

I would assume so.

Quote from: Blazr on August 06, 2015, 09:38:18 AM

if so, then why doesn't this happen when bugs are found in other clients?

It would.

If a full node client were to accept something into the blockchain that the protocol doesn't allow, then those that implement that client program would have a choice. They could either roll back the the last block that the protocol allows and then re-sync with the rest of the network, or they could fork their own new blockchain and convince enough users that their bitcoin is the real bitcoin and that any other client that doesn't implement the protocol the way they did is an invalid fork.

Quote from: Blazr on August 06, 2015, 09:38:18 AM

Does Bitcoin Core have some kind of privilege in this sense?

Right now, bitcoin core is considered by most to be "the reference implementation". In other words, the exact specifications of the protocol are defined as the being "the protocol as implemented by Bitcoin Core". If someone else was to create their own full node implementation, and was to convince a significant majority of users that their implementation should be considered to be the "reference implementation", then Bitcoin Core would lose that privilege.

medUSA

legendary

Activity: 952

Merit: 1005

--Signature Designs-- http://bit.ly/1Pjbx77

Quote from: Blazr on August 06, 2015, 09:38:18 AM

If a serious issue like this was ever found in the Bitcoin Core client again, would the offending txs be removed again?
if so, then why doesn't this happen when bugs are found in other clients?
Does Bitcoin Core have some kind of privilege in this sense?

Was there a bug found on other clients that threatened the bitcoin protocol? I haven't heard of any.

Quote from: Blazr on August 06, 2015, 09:38:18 AM

Does Bitcoin Core have some kind of privilege in this sense? or was this just a once off thing, or did the small bitcoin community at the time manage to get everyone to agree to have this tx removed?

The bug created more bitcoin than initially designed. I don't think a consensus is needed to fix the bug.

I believe it is highly unlikely another major bug like that would happen again. Bitcoin has been running for 5 years after that incident, any major vulnerability would have broken out or fixed by now.

LiteCoinGuy

legendary

Activity: 1148

Merit: 1014

In Satoshi I Trust

Quote from: Blazr on August 06, 2015, 09:38:18 AM

On August 15th 2010, somebody found a buffer overflow exploit in the wxBitcoin client (which is now called Bitcoin Core) and used it to create 184,467,440,737 BTC (184 billion) out of thin air (https://bitcointalksearch.org/topic/strange-block-74638-822 ). Just 5 hours later a new version of the Bitcoin client was published with a fix but in addition the blockchain was hard forked and the offending tx removed from it.

If a serious issue like this was ever found in the Bitcoin Core client again, would the offending txs be removed again? if so, then why doesn't this happen when bugs are found in other clients?

Does Bitcoin Core have some kind of privilege in this sense? or was this just a once off thing, or did the small bitcoin community at the time manage to get everyone to agree to have this tx removed?

if this kind of bug occurs we would try to go back and start from block X with the new software. but i doubt that such a major bug will happen again.

AtheistAKASaneBrain

hero member

Activity: 770

Merit: 509

Quote from: Blazr on August 06, 2015, 09:38:18 AM

On August 15th 2010, somebody found a buffer overflow exploit in the wxBitcoin client (which is now called Bitcoin Core) and used it to create 184,467,440,737 BTC (184 billion) out of thin air (https://bitcointalksearch.org/topic/strange-block-74638-822 ). Just 5 hours later a new version of the Bitcoin client was published with a fix but in addition the blockchain was hard forked and the offending tx removed from it.

If a serious issue like this was ever found in the Bitcoin Core client again, would the offending txs be removed again? if so, then why doesn't this happen when bugs are found in other clients?

Does Bitcoin Core have some kind of privilege in this sense? or was this just a once off thing, or did the small bitcoin community at the time manage to get everyone to agree to have this tx removed?

The chances of a hardfork on softwares get more and more slim as time goes back and the backbone of the software gets more solid.
I think hardforks will be due blocksize increase if there isn't a way around it.

Blazr

hero member

Activity: 882

Merit: 1006

On August 15th 2010, somebody found a buffer overflow exploit in the wxBitcoin client (which is now called Bitcoin Core) and used it to create 184,467,440,737 BTC (184 billion) out of thin air (https://bitcointalksearch.org/topic/strange-block-74638-822 ). Just 5 hours later a new version of the Bitcoin client was published with a fix but in addition the blockchain was hard forked and the offending tx removed from it.

If a serious issue like this was ever found in the Bitcoin Core client again, would the offending txs be removed again? if so, then why doesn't this happen when bugs are found in other clients?

Does Bitcoin Core have some kind of privilege in this sense? or was this just a once off thing, or did the small bitcoin community at the time manage to get everyone to agree to have this tx removed?

Topic: Will we hard fork again for another serious bug? (Read 810 times)