Bitcoin Forum

https://twitter.com/alistairmilne/status/856405606630133761

https://i.imgur.com/pc5MQMV.png

I'm more worried about other sorts of exploits in code of this quality...

I wonder if there will be a closed or open source update for this one?

Okay you win on the direct title thread, but I prefer my ironic one Popularity of XT/Classic nodes rise in proportion to BU nodes (https://bitcointalk.org/index.php?topic=1884346.0).

This attacks comes just over a month after the last, so I wonder if bugs responsibly disclosed have not been fixed and are now public knowledge?

Or perhaps everybody gets afraid when I decide to fire up my 2 GH/s USB miner against BU.  :P

It is really funny to see that the BU bug once again bit Roger Ver when he tweeted that bitcoin unlimited is production ready,poor Roger Ver ,he is a laughing stock now a days and i do not know how he is going to over come this and it looks like his dream of getting a general consensus for Bitcoin Unlimited is far from over. ;D

That is just the thing. Bitcoin software is being attacked every single day. Every type of hacker is looking for potential holes in the code (all ranging from white hackers to government paid 'master' level hackers). This is the primary reason for which Bitcoin software needs to be "top-notch" quality in terms of security. If BU developers spent less time trying to lobby miners in China, they could have avoided some of these.

Roger Ver has zero credibility when it comes to technicalities. This reminds me of the time he said Mt. Gox was solvent, not long before everything went up in flames.

Comedy relief from social media:

---

Then you have people showing signs of delusion:
https://i.imgur.com/fmeWRRL.png

Yeah we should definitely diversify so if one type of node is full of bugs and crashes all of a sudden the entire network doesn't go under.

Someone said that too many times. now picture bitcoin forked by BU and that happens literally people should trust millions and billions to such forks of Core?

Time when I believed BU was only a change of 2 lines of code which were only about increasing max block size after someone I considered viable said it to me has past.

Funny thing is that, they want you to privately tell them about any bug that you find so that they could go and fix it in their comfortable chairs in green and safe zone like the wolves out there in the world going to wake the watch dog up before they attack the herd lol.

It only show the incompetence of the entire BU gang as they can't squeeze out bugs before releasing their garbage (but most of the people were already aware of that). Anyone supporting BU contributes to his own downfall. In that regard, I consider everyone pointing his gear towards a pool that votes/supports BU to be equally as worse.

Not a bug, it's how emergent consensus works. All the nodes crash and the last node standing decides the consensus rules. Better known as consensus by DoS.

And you are?
Last man standing?
I don't know why bother competing with Core/ original bitcoin?
Why not creating a version of their liking totally and point their hash power towards mining that?
I can guess, because they want the virgin, they want BTC if they were confident enough then starting an altcoin shouldn't be a problem.

Please let BTC be only one version since only one version's rules are being followed and blocks mined under the one version's protocol, just start BTCU already, how many times people should say this? there are only 6 up to 10 people which I suspect 4 of their accounts belong to 1 person here are defending BU.

You can't hijack the brand put it in to your skulls.

can someone explain to me in a non-drama kind of way, why is this becoming a common thing with BU? specially since it is a fork of bitcoin! i mean they literary forked bitcoin/bitcoin on github, so what did they change to lead to bug after bug and crash after crash?

He was being sarcastic.

They have forked an old version of Bitcoin Core (0.12.x) and have written their own code for things such as Emergent Consensus, Xthin, et. al. However, their developers are extremely incompetent and there is basically no quality assurance process.

Of course they have forked from bitcoin core 0.12.x, as they are against the segwit soft fork so don't want the pollution from that code. Unfortunately, it seems some of the alternative development teams are not keeping up with other fixes and improvements.

They have made 920KB of changes to the code and completely muddied the entire codebase. I debated listing the major problems they have introduced, but I really don't know where to start and don't have the strength to list them all. To sum it up, they clearly do not understand how the codebase works at all. They are willy-nilly making changes without understanding the full ramifications of those changes. Their code is very messy and hard to read. They have broken compatibility with Core so cannot accept patches from Core and are stuck on an old outdated version. They moved core functionality out of their own separate modules and hacked it in all over the place. They do not understand how Bitcoin uses asserts to detect situations where it is running in an impossible state and crashes to prevent exploitation, so instead they straight up removed this entire system putting them at huge risk. They do not have any test or QA procedures. Often developers commit code without any third party review. The developers have no experience developing anything like Bitcoin at all and have crazy ideas about how it works (recently a BU dev tried to convince me CAP theory applies to Bitcoin). Thats just some stuff I can remember off the top of my head.

Pollute what? Segwit has been running on testnet for months and even on some live networks (altcoins). Number of crashes caused by Segwit since it was 'production ready': 0. The BTU team has polluted the code with their own amateur coding.

'Pollution' is obviously subjective as whether you agree with the segwit soft fork direction not. If it activates, alternative implementations would have to implement it to remain on the true p2p network.

If EC activates, alternative implementations would have to implement it to remain on the true p2p network. In other words, water is water only when I want it to be wet? What is your point?

Reading further into comments it's not even clear this is an exploit being attacked at all and could well be a spontaneous 'coordinated' bug due to network conditions causing a massive memory leak on the BU client. One person said he didn't have problems on his PC with 256GB ram  ::)

Yeah I was thinking that too. Reading between the lines of what the devs are saying it could be related to the large mempool. It may not be an exploit at all.

The person with 256GB RAM was a dev too. Takes the old saying "works on my machine" to a whole new level.

Quite possible. I did read an issue in their repository where mining nodes were increasing in memory usage overtime.

EDIT: Yep, looks like OOM killer did its job.

But is anybody really thinking that BU is anything else but an "I don't want Segwit" thing, in other words "I want to keep bitcoin exactly as it is, but I want to make you think that I want also a "solution" for crashing the lucrative fee market" ?

For some people it's:

I want Bitcoin to split in two, so I can 'double' my coins!
I hate the censorship theymos is doing on his privately owned forums, therefore I am going to express my anger of this by supporting BU
I think the devs have full control over Bitcoin and I trust the miners more than I trust the devs, so I want miners to have the power to make economic policy.

 :D That one is funny  :)

What does EC stand for? Exchange Circles?

It is proven time and again that they are really not competent enough to handle all these lines of codes,but what i do not understand is that Roger Ver is known to spend money to keep people in his side so that BU has some followers and he is know to be a big whale,but why is he not able to spend those money to get competent developers and testers before boasting about it, after all these failed attemps who is going to believe him.

Extra Crashy

Some people can't be bought? I joke that I'm bitcoin operated since I code for bitcoin, but I'd never code on BU just because he threw money at me (note I'm a mining/pool software developer, not core code.)

out of this entire topic there seems to be only one post that is seeing the big picture

its about some people thinking core should be king. or thinking th debate is only about BU vs core and who gts to be king

there should be no king. just a diverse decentralised peer network
meaning
lots of "brands" all uniting with a consensus of rules they all follow

that way we dont have a issue with core should they have a bug (EG 2013 leveldb would not have been such a drama event) because there would be other brands keeping core the network alive while core sort out their implementation

that way we dont have a issue with BU should they have a bug (EG assert drama event) because there would be other brands keeping the network alive while BU sort out their implementation

trying to get everyone using just one implementation is where attacks externally can cause mega issues, and dictators internally can cause mega issues

Ehh obviously there's people genuinely pushing for BU, Roger Ver himself im convinced doesn't have any machiavelian intentions when it comes to wanting to leave bitcoin as it is. He genuinely things BU is better than Core which is insane, as we can see the code keeps crashing continuously because BU is a failure developed my amateurs. No one with a functional brain will trust BU over Core with their money.

There are other node implementations, but BU is out of consensus.

It is already diversified with actually safe-to-run code:
https://i.imgur.com/IX0z4tm.png

---

Among a few other thing such as Knots.

all nodes you highlighted are all blockstream managed..
yep knots=blockstream too
=not true diversity, just pretend diversity

your still not seeing the big picture

And despite so many core implementations out there, no instances of crashes taking down large numbers of any version of them. The stability record of core is exemplary.

I edited the source and changed the background color of the GUI to orange, so it matches my wallpaper, and added my name to the list of contributors. I now run my own "alternative implementation" :P

Hopefully this change that I made will prevent my client from getting affected by some bug in Core.

thats the titanic, big bank "too big to fail" mindset... especially ignoring past fails to pretend it will never fail

2013 leveldb transition
now imagine if in 2013 there were implementations wrote in Go and other implementations that had other databases that would not have been hit by the berkely locks

the implementations running non-berkely db's would have been fine and that includes the one running Go too and only the small amount using the old Berkeley with the lock problem would have been held up

The network is already diversified:

https://i.redditmedia.com/IpX3OQOQ0DCmv172h33ja_L_CGDBn1aRgE8D_bEzLe4.png?w=1024&s=047ed3c3e7ab399f29f09d37da94db1d

No need to diversify on shitty software that steals 99% of code, adds 1% and this 1% is always the code that makes the nodes be prone to all kinds of exploits.

Different versions of reliable Core software to guarantee in the rare case the latest version has a bug the entire network doesn't go down is already in place as you can see. Im sorry that people only trust Core with their money, must be a hard pill to swallow for antiCore shills.

My thoughts exactly. The network is pretty diversified when it comes to versioning.

---

While franky1 continues to shill for the disaster that is BU, we are back to this again:

https://pbs.twimg.com/media/C66gpcRWgAAFlMt.jpg:large

Well, i'm not even suprised seeing this bug ore whatever happend to BU. This whole BU project is one major joke, BU software is bugged, nodes are bugged... screw this shit man freaking amateurs.

Why cant Jihan and Roger just admid BU is dead? The only thing they got left is hashing power, both will continue talking FUD on Twitter and pushing for a HF  :-\

What kinds of rules or protocols can be changed and at the same time other versions accept them?
Let me simplify my question here, imagine all Core nodes crash, now what can BU nodes do that Core nodes couldn't? they can not change anything that Core nodes wouldn't accept and if BU nodes try to change anything then miners would not accept and if they do then they would be forking away from the Core code/ version.

By all means let BU nodes stay active so everyone could see which version is superior and is more stable.

If Core nodes crash then BU nodes which are standing still have to follow the Core rules or else they will be forking away from the network which were ruled by Core code.
When hospitals have no grid power their reserve generators kicks in and provides them with electricity but if Core nodes fail then BU nodes will have to provide the same electricity Core was providing the network and they can't instead start providing WIFI signals :D I hope you get what I mean.

lol.. that's ... a good one.. having to turn btctalk.. off - as I can't stop laughing and I'm sitting at my desk at work..

This pic is telling uss exactly what BU supporters think about this situation. Like always, they love to blame and trash everything and everyone exept there 'mighty' bug unlimited :o

Edit: they even blame the media and social media for this attack..LOL

Spot on.  Even if the devs are corrupt, they're not naturally/inherently that way.  Miners have a profit motive, so voting from miners is solely based around who has the most money to throw around for Bitcoin.

Core has been consistent enough not to cause a giant crash in the number of nodes being ran.  Anyone committed enough to run a full node will have an understanding that alternative implementations exist, and yet most of them don't choose BU or other nodes, simply because Core comparatively know what they're doing. 

BU's status as a kind of miner uprising is terrible.  It's basically just miners hoping to increase the transaction capacity directly for profit who are terrified of offchain transactions to make microtransactions convenient or other solutions to actual problems in the protocol.

if core was humans.. it would be called several generations of imbreds

lol i think you need to learn diversity. you know totally different groups/families with different genetic makeup (code/language) all co-existing

and as for lauda

your the one that mentions BU in nearly every topic. you keep failing to pigeon hole me into one brand. because i am not someone that wants just one brand..
while i try to talk about diverse decentralised bitcoin peer network that way it minimises risks

EG lets say there was a bug in version 0.3 of core.. and 0.4 and 0.5 and 0.6 and 0.7...
but doesnt trigger until 0.8
... oh wait that happened..
and that was not just a 5 minute bit of comedy/drama..

now imagine if there was btcd (written in go) and other implementations that didnt have the berkeley db...
we would not have had much of an issue in 2013 as what actually happened.

please dont have the blockstream defender hat on and scream KNOTS because that too is blockstream.. im talking real diversity..

even to the point of where the devs allowed users to set their own settings at runtime to not be spoon fed from the same group of devs.
= real independence and diversity.

seriously, take just a 10 minute break from the keyboard and remove your blockstream adoration cap.. and think critically about bitcoin 120 years
not blockstreams next 3 years.

devs come and go. and kissing one butt cheek will leave u lonely when they retire, get bored, or move onto something else.
think beyond blockstream and think about bitcoin

I even like the thinking of having a diverse decentralized peer network but some people are really trying to monopolies things and that is when the problem arises,look at the course of actions done by BU,they created a paper monster but the monster keeps on failing on the real test,let us critically think about the future of bitcoin,would you stand by people who are driven by monetary gains or those who knows about maintaining a certain protocol.

It's so fake how everytime BU crashes, all the nodes go up again pretty much at the same time. Someone is obviously running a ton of BU nodes and he turns it on all at the same time.

Also no, we don't need a lot of different clients to "diversify the network", the different versions running at the same time are enough. Satoshi was against anything that wasn't the Satoshi client for a reason.

franky1 is a known anti-Core, anti-Blockstream, pro-big blocks shill.

imagine if..
all the network had was
core v0.13.x core v.014.x core v.15.x next year and there was a bug in all 3

think about it

Someone is feeding hackers all of the bugs that were fixed in Bitcoin Core since version 0.11.2 (or whatever version Unlimited forked from). I'm guessing they slam the BU nodes with a "new" exploit monthly, to keep pace with the news cycle and have "BU HAXXED OMFG" in the headlines once per month.

Not that big of a deal. BU is not the greatest code because it doesn't have enough devs. Core is a good team doing bad work for some financial saboteurs... Choose your posion.

Everything is Core && Core supporters fault, even the amateur coding of the BU team. :D

That is not going to happen unless you force the network into a hard fork. You are actually shilling for a hard fork, therefore you are being hypocritical again.

This doesn't look to be the case, it would appear all the crashes so far have been due to a single very buggy feature(thin/xthin blocks) that was written by BU devs, it was never present in Core.

Don't fall into the trap of assuming all kinds of malice going on just because they're doing something wrong in the first place. A lot of bitcoin nodes are almost certainly being run as services on linux machines which are designed to restart in the event of a crash. This last crash was an out of memory error so restarting the node probably will be all that's needed until the memory gets out of control again. I'm pretty sure that none of them even are using a bugfix yet for this latest issue. They just have their nodes set to restart in the event of a crash... or actually what it looks like to me is they're set to crash in the event of a restart  :D

Wow 256GB Ram much wow so impressed it runs.
Geez use it on a clunker that is how you test if stuff really works.

This is what I was assuming to be the case, it was confusing me constantly when people would try to argue that the crashes aren't due to something going on with the network or coding itself and rather some other reason, often more external.

I won't say that the Core devs are good guys. I won't say that the BU can't code, but I don't think that all these crashes are coming from something outside the network. I might be wrong, but I'll assume I have a decent idea of the issue for now.

not so long ago.. I saw a tweet from Roger Ver..  something about Core dev being coder and not economist/merchants..

wich one would you trust between very good coders or very good economists/merchants ?

I'm all in with Core coders (very very good)..

Just take a serious look at where "economic leader" bringed us to in history, specially recent history (10 years)..  HODL to your BTC and buy some more ;)

I go for good coders, without them you cant run a succesfull business.

Oh, and Roger is talking Bullshit with his BU buds...

https://i.imgur.com/1uFFK2V.png

You are so mean, now my belly hurts from laughing  :D :D :D :D :D :D :D

No, oh no, this is getting more mental every day. Segwit is supposed to have messy code, say the BU supporters, and they may be right...but look who's talking  ::)

256GB RAM, and maybe 10TB hard disks every year, but they probably think: BITMAIN can easily afford it, and they will be the only ones left mining anyways with Asicboost+BU

And mind you, Roger already said, passive nodes are useless, so ONE BIG NODE in CHINA with 256000 GB RAM and 10000000 TB harddrives, Bitcoin is safe until 2140.

*hitting head at desk*

That shows that they do not respect the users. They cannot not admit that they made a mistake or they made an inferior implementation and a buggy version of Bitcoin. They are now a joke, I cannot find a reason why some of the Legendary members here still defend them.

Could it be that they are paid to do it? Who will ever know? You have to hate someone or something with a lot of passion to support something that fails 4 times in a time span of a few months. Some of these people might be invested in some Alt coin or even a developer on their team, so they will naturally want Bitcoin to fail.

Some of these people are employed by financial institutions in the Fiat financial system that are being disrupted by Bitcoin. There are a lot of hidden agendas behind this BU support you are seeing. ^hmmmmm^

would not surprise me if chinese farm are pushing for this alone, just to have more coins to dump, they are known to be greedy and care only about their profit in fiat

because it's beyond me how they can still support bu after all this bug and exploit that are discovered each week..


so he tried to corrupt you too? all this under the table drama, make me think there is more than meets the eye here

No, they honorable people from what I have noticed and they are only against Core because they are for the reformation of how things are done today. There is no harm in that. The real harm though is when we hard fork and find out that it will cause bigger problems.


Really? I believe you are reaching.

Not me personally, no.

I was away for a few day and not surprised to see that BU crashed again. However, what was causing the problem? Is the same bug also found in core?? I guess this is a good learning opportunity for everyone (BU or Core) as that the same problem will not happenand can be avoided/prevented in future.

It's likely a memory leak. Oom-killer is a native Linux app that kills apps that are using too much memory before they overflow the system memory. Some are claiming it's in the xthin code. I have yet to see a concrete breakdown. Good that they are fixing their buggy code now when less people are running it...

Somebody posted a great question on one of these threads:

Would you really trust future of your Bitcoins to people who cannot keep their network operational for 2 weeks without interruption?

SEGWIT RULES!

Not a  Ver fanboy, but you must admit that his retort was pretty good: