[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/biz/ - Business & Finance


View post   

File: 506 KB, 967x779, IMG_20200221_161348_01.png [View same] [iqdb] [saucenao] [google]
17373467 No.17373467 [Reply] [Original]

https://blog.chain.link/improving-and-decentralizing-chainlinks-feature-release-and-network-upgrade-process/

>> No.17373488

>>17373467
>it's only 40k bro chill that's like nothing

>> No.17373525

>>17373467
>fudders btfo
>lost 40k for a customer due to a typo
How are fudders btfo again?

>> No.17373539

>>17373488
It is, in the grand scheme of things.

>> No.17373540

>>17373488
That literally is nothing to an ecosystem aiming to handle trillions. I'm a mega poorfag student and my 2017 link stack is worth 40k lmao

>> No.17373552
File: 12 KB, 741x443, PREMINED SHITTOKEN.png [View same] [iqdb] [saucenao] [google]
17373552

>> No.17373566

>>17373467

INCOMING DUMP! IT'S FUCKING OGRE GET OUT NOW!

>> No.17373569

>>17373540
what you fucktards don't understand that the nomical amount means fucking nothing

they got fucking lucky as shit not many people were holding sXAU or it would have completely bankrupted synthetix

>> No.17373610

>>17373569
It's fucking synthetix themselves who used their fat oafish fingers to request XAU instead of XAG.
>their main coder is literally Dimitri

>> No.17373653

>>17373610
No retard, quit the magical thinking, it comes directly from feeds.chain.link go look at the xag pair before you open your fucking retarded mouth

>> No.17373661

So basically tldr they were updating some stuff and then someone wrote xau instead of xag and they didn't have multi signatures in place.

>> No.17373692

>>17373661
basically. now I wish I didn't panic sell my stack

>> No.17373715

>>17373467
>The Standard

>> No.17373753

>>17373653
Who was updating the contract smartass?

>> No.17373795

please come back dimitri

>> No.17373846

>>17373753
some chainlink dev who should be unemployed right now

>> No.17373902

>>17373467
>He thought it was a decentralized oracle network but it's actually just Adelyn doing manual data entry

>> No.17373906
File: 49 KB, 809x456, chingalingLing.png [View same] [iqdb] [saucenao] [google]
17373906

>> No.17373938

Embarrassing but a minor cost ultimately. And the best possible cause. Nothing to do with the system.

>> No.17373967

So the Chainlink network itself is fine and dandy right? This was 100% human error?

>> No.17373985

>>17373938
>the best possible cause. Nothing to do with the system.
This.

The oracles themselves are rock solid, it's the infrastructure around it that is still a little iffy.
Which is normal at this stage, and exactly why Sergey won't rush things.

>> No.17374008

>>17373967
Yes.
Read the article.

>> No.17374027

So much copium

>> No.17374054

>>17373846
So should be on the github then, and we even have the timing. Let's see who it was.

>> No.17374086

>>17373715
Kek

>> No.17374095

>>17373753
Told u sergey can't write contracts

>> No.17374138

there goes defi

>> No.17374143

>>17373467
>typo by chainlink dev can cause liquidations and profit on defi

>decentralized
mfw people put $1 billion marketcap on LINK

holy delusion

>> No.17374155

>>17374027
>>17374138
2 y e a r s

>> No.17374162
File: 194 KB, 854x811, 1582241521471.jpg [View same] [iqdb] [saucenao] [google]
17374162

>>17373467
>the king is dead
>long live the king

>> No.17374182

>>17374143
The oracles worked perfectly.
It's the peripherals that are still taking shape.
Not to mention all the security measures that still aren't in place

>> No.17374227

>The human error occurred while seeking to improve the XAG/USD network by releasing additional data reliability features requiring an update to the smart contract’s configuration about the specific job IDs being run by node operators servicing the XAG/USD oracle network, leading them to incorrectly request a gold price (XAU) instead of the silver price (XAG).

Why update the contract in the first place? Just make a new one with the new features, test it, then switch to this version and stop using the old one.

>> No.17374235

>>17374182
oracles are NOT working perfectly if some dev can swap what they point at whenever they wish.

thats literally centralized trash, underpaid chainlink dev could drain ALOT of syn usd over 5 mins if he wanted.

>> No.17374252

Human errors exist in every business, this just gets magnified because huge community of investors and public transparent errors. It's still a major fuck up that rightfully makes you questions the team's ability to handle security. The problem here is that clearly the update process is a centralized point of failure. Early fuck ups are inevitable and you can claim its actually an important learning process but it's cope, truth is they should have planned a solution to this process in the first place and it's terrible PR. Don't think everything is fixed and fine now, they're gonna have to swallow a lot of shit in the dev community because of this and competitors will have serious fud ammo to discredit the project. It's centralization was always its weakness. Having said that it's also important to point out that the reference data feeds aren't the Chainlink network itself, it's maintained by the Chainlink team to incentivize adoption but this means it was just a bad use of the protocol, not the protocol itself that is the issue. Both reassuring yet still real fud at the same time because it really makes them look incompetent if they can't even implement their own software without fucking it up.

>> No.17374263

>>17374235
That's not at the oracle level you retard.
This was at the job id coding level.

The oracles worked flawlessly.

>> No.17374333

>>17374263
Imagine staking is live. You're running a node. You report the CORRECT price but you lose your stake because 50%+1 nodes reported the wrong price. TADA!

>> No.17374349

>>17374263
Best way to put it chainlink works fine but it's smartcontract.com that fucked up

>> No.17374365
File: 46 KB, 1348x300, wheredtheygo.png [View same] [iqdb] [saucenao] [google]
17374365

>>17373467

uhg i dont feel good

>> No.17374401

>>17374333
If your output is the same as your input, you don't lose your stake.
Whether that input is correct or incorrect.

>> No.17374421

>>17374365
The oracles worked perfectly, so this sentence makes no sense.

>> No.17374446

>>17374421

wording argument? the "oracles" worked perfectly and the oracle "network" failed?

>> No.17374448

>>17374365
You can say goodbye to compound
Sergey must be seething right now

>> No.17374486

>>17374446
The oracles worked perfectly, the backend fucked up.
If you get data from oracles and you mishandle it, then the oracles aren't to blame.

>> No.17374509

>>17374448

what is compound? is that the part of the oracle that puts the data on ethereum?

>> No.17374513

>>17374365
Whats the source anon

>> No.17374534

>>17374162
>Sirgay is dead
>Long live NuSirgay
Fixed it for you.

>> No.17374537

>>17374513
Literal who is the source

>> No.17374551

>>17374448
Compound runs on ETH.
You do know about ETH right? All the fuckups it had?

>> No.17374570

>>17374509
It's the biggest defi project, Sergey has been trying to build credibility with the smaller projects so he could bag compound, it's months of work undone because of a typo. Their reputation is fucked.

>> No.17374583

>>17374570
See >>17374551

You're cute when you try though.

>> No.17374591

>>17374252
Didn’t read not selling

>> No.17374618

>>17374401
Citation needed.
What's the point of staking and reputation if you can get away with reporting data well outside of consensus?

>> No.17374672

>>17374618
you report the data the contract asks you to report
if the contract asks the oracle to report the value of gold instead of the value of silver, as in this case, that's what it does. authoring the contract correctly is on the contract's author.

>> No.17374679

>>17374618
>Citation needed.
The SLA is entirely up to the parties to decide.

>What's the point of staking and reputation if you can get away with reporting data well outside of consensus?
Consensus in itself is in no way a conclusive manner to determine whether an oracle tampered with data or not.
This should be obvious.

The whitepaper literally describes how outliers could be identified before averaging even takes place, which is a step before consensus even.

>> No.17374701

>>17373902

Kek

>> No.17374759

>>17374672
>it's not chainlinks fault the smart contract asked for the wrong data durrrrr

https://feeds.chain.link/xau-usd

why do you faggots even open your mouths

>> No.17374788

>>17374679
Anon, most link holders have never read the whitepaper. Even less nolinkers have read it. Their baseless fud doesnt really deserve a well thought out response in my opinion. Thanks for educating /biz/ anyway.

>> No.17374795

>>17374759
He's 100% correct.
It's just that in this case Chainlink was running the backend, which is obviously only meant to be temporary.

>> No.17374806

>>17374252
u stupid fuck. who do you think write or update those smart contracts? your Allah?
stop fudding Mostafa!

>> No.17374813

>>17374486
chainlink network mishandled it, it had GOLD for SILVER

poor defi dapps got $40k lost coz chainlink reported wrong price coz DEV made a centralized TYPO

>> No.17374827

>>17374795
Synthetix smart contract did not change and ws requesting the price of silver, but chainlink claims to have fat fingered it accidentally gave the price of gold instead

cope how you want but don't lie about it

>> No.17374835

>>17374813
>chainlink network mishandled it
Chainlink, yes.
Because Chainlink is still running most of the entire system for the time being.

But the "chainlink network" (i.e. the oracles) handled it perfectly.

>> No.17374868
File: 947 KB, 1920x1080, 1569292827300.png [View same] [iqdb] [saucenao] [google]
17374868

didnt read

never fucking selling.

>> No.17374877

>>17374827
>chainlink claims to have fat fingered it accidentally gave the price of gold instead
Yes, the code upstream from the oracles got bungled. This is "contract writer" error.
I never said otherwise.
The oracles themselves however worked perfectly.

Contract writing (and SLA writing etc.) is meant to be handled by third party services (or the parties themselves) once the security measures are in place.

>> No.17374904

>>17374877
right you're handwaiving it away through weak semantic arguments

this was a project ruining level error they that luckily dodged by fucking a pair with little open interest. Like I said, if this happened on sETH SNX and LINK would both be worth 0 right now.

>> No.17374907

>>17374583
You're being low iq here, Sergey has been trying to court him for months, he even leaked their private emails and all the competition came in the comments to shit on link, it's fair to say he had a negative opinion on link or that his advisors are negative on it, and the recent campaign to showcase link as the savior of defi after the synthetix and bzx hacks were part of Sergeys plan to establish major credibility and woo compound, because if you woo compound you own defi and that's major credibility to move on to the next steps. This is a major setback in the plan to convince compound to adopt, and their reputation will suffer in the entire ecosystem. It's not the end of the world but it's one step forward two steps back. The sad thing is that it was going so well, the bzx thing was a marketing opportunity from heaven and if it weren't for a fucking fat finger they could have locked them down before summer. Not gonna happen now. Simply having a conversation is going to be a tough thing.

>> No.17374944

>>17374513

defi telegram is talkin about oracles right now

>> No.17374958

>>17374904
lmao no I'm not you dimwit.
This was an error in the writing of the contract by Chainlink, who are only temporarily handling contract writing because mainnet hasn't been opened up yet.

>>17374907
Compound runs on the same ETH that has gone through many gargantuan fuckups over its lifespan.
End of story.

>> No.17374974

>>17374907
>or that his advisors are negative on it
we already know krug is and threatened to pull funding if they used chainlink.

>> No.17375029

>>17374958
Eth had fuck ups, that's your argument? Go tell him that, I'm sure that'll convince him to adopt. Flawless logic.

>> No.17375050

>>17375029
Of course it's flawless logic.

>> No.17375053

>>17374958
>14 posts from this user
its crazy how much you care about our financial well-being.

>> No.17375065

>>17375053
Just doing my part, you're welcome.

>> No.17375096

>>17375053
please don't sell sir I bought at $4.80

the oracles are fine! it's just our dev mixed up silver and gold on your critical software is all

>> No.17375099

>>17375053
>using 4chan in its intended manner is evidence that you are trying to trick me!
Can we stop with using “# of posts by this ID” as if its evidence for anything? Theres plenty of thread on /biz/ that i have posted in that many times, and there is plenty i have posted in only once. Thats how conversations work.

>> No.17375103

>>17373902
Kek

>> No.17375280

>>17375096
point is the mistake could have been made by anyone (though not at this point, as the team handles everything), and isn't really pertinent to the service chainlink aims to provide.
the contract shouldn't have gone live with incorrect parameters, a mistake like this absolutely puts the team in a bad light.

>> No.17375353

its over, its fucking over. we could have been rich...

>> No.17375378

>>17375280
that's actually fair
but all the people shit talking bzx and 'bend the knee' fags probably have no idea of how close this came to vaporzing billions in marketcap

>> No.17375724

>>17373467
lmao what a shit show of a blog post. "sorry about your 40k you lost man. we good though?" It's fucking amateur hour over at ChainStink.

>> No.17375777

>>17374263
this is literal "it's not the bullet that killed you, it's the blood loss that did" level cope.

>> No.17375844

>>17375777
Checked and based

>> No.17375858

>>17373906
I don't understand why people suck off Oracle so much on /biz/. They've always been a shit company with shit products that exist because of inertia and choking out the competition by non-technical means.

>> No.17375859

>>17375099
stfu you stupid faggot bitch.

>> No.17375863

>>17373467
Its actually quite funny how the whole puzzle seems to fit together now, and this human error wont stay a single case. Believe me or not I am a 8figure worth oldfag business owner who started to buy Bitcoin in late 2014 and nothing else - not a single altcoin in my whole life. I rarely come here any more let alone post, but I will experience how this whole board will implode in itself, with all of its users emotionally and financially investing their literal life to this. Stay away from cults, or you will learn it the hard way

>> No.17375924

>>17373467
>>17373467
>Under 40k effected
>NOBODY USING IT SO NOTHING OF VALUE IS LOST.

Seriously though mad respect for coming forward with an article.

>> No.17375933

>>17373906
Fake account

>> No.17375940

>>17375858
>Reeee I don't like Oracle's success because it means Link is successful!

>> No.17375957

>>17375858
>shit products that exist because of inertia and choking out the competition by non-technical means.

Sounds just like Chainlink

>> No.17376011

>>17373569
hi jim

>> No.17376018

>>17375863
As long as this is an error at the contract level it matters less. In the future there will be audits of contracts before they are implemented by nodes.

>> No.17376045

>>17376018
>audits on smart contracts
This. Problem solved.

>> No.17376256

>>17376018
>In the future there will be audits of contracts
This.
And reputation systems I reckon (both for contract providers and pre-written contracts).

>> No.17376308
File: 914 KB, 1921x870, duuse2.png [View same] [iqdb] [saucenao] [google]
17376308

Join the actual movement

>> No.17376436

>>17374252
>competitors will have serious fud ammo
There are literally no competitors lmao

>> No.17376514

>>17373467
2+ years and a basic KYC price feed medianizer (the one thing available on mainnet) failed. they're great at networking with potential clients and partners but ultimately the tech needs to get there

>> No.17376627
File: 1.07 MB, 1536x1403, B06D1645-9996-4AAB-88C1-81E42CE408D5.jpg [View same] [iqdb] [saucenao] [google]
17376627

QSP will be huge, dyor

>> No.17376684

>>17376436
Kleros