Balancer hack: second update

General
1

Gnosis is and always has been a deeply values-driven chain.

These values stem from the simple and yet powerful idea of a system that serves ordinary people. We are here to build an alternative to the current structures, that are designed to extract the maximum value from ordinary people, in order to continue to make banks richer.

To succeed in this mission and put our values into action, we need “low-risk DeFi”.

Ideas like decentralization, censorship-resistance, and self-sovereignty all flow from that core idea and we have built them into the very foundation of Gnosis. It is impossible for any actor, including a unified Gnosis leadership, to act unilaterally. However there is still some discretion available for key parties to co-ordinate to buy time for the DAO to act.

This week we exercised that discretion to reduce risk, limit harm from a malicious actor, and buy time for the DAO to determine the path forwards.

Key actions included:

  • Restricting all major bridge outflow limits to prevent funds being taken off chain.
  • Executing a soft fork to block stolen funds while resuming normal operations.
  • Removing bridge outflow limits and effectively reopening the bridges.

Now that we have breathing space we will share our learnings and put the next decision - whether or not to execute a hard fork - to the DAO for vote.

We will also propose a set of principles to guide us in any future challenges. This way we can align together on a framework of how various parts of the chain infrastructure (bridges, validators, stable coin issuers etc) can and should react under specific scenarios.These will also be shared for open debate.

Lastly, we want to take a moment to recognise the exceptional dedication of the teams at Balancer, StakeWise, Monerium, Nethermind, Erigon, as well as all of our validators, and the core devs at Gnosis Chain who have worked tirelessly since the incident to protect user funds and give the DAO this time.

Onwards and upwards!

18 Likes
Balancer hack: hard fork
Balancer hack: hard fork
2

Now that the soft fork has been executed successfully and the immediate threat is contained, credit is due. The teams involved pulled off a high-pressure coordinated response and stopped the hackers from running off with the money. That matters. Thank you to everyone that pulled long hours to achieve this. Thank you to everyone who worked hard to make sure the stolen money does not fund some criminal enterprise, perhaps even buy NK missiles.

But the job isn’t finished.

I support a hard fork to restore the funds to the people who actually lost them. Many of these were real-world savings pool. Many of these users were holding EUR savings in EURe/sDAI and spending them via Gnosis Pay and Monerium. They weren’t degen yield farmers chasing a high. They were ordinary people who were introduced to blockchain through Gnosis as a low-risk, real-use system. We have all (me included) been pushing people to Gnosis for this.

If we want DeFi to be taken seriously by normal users, then we need to treat their losses seriously. That means making them whole if we can. And now we can.

We’ve already had the “code is law” debate. In 2016 the rigid camp chose to stay on ETC and fade into obscurity. The rest of the ecosystem moved on. Today, Gnosis and the broader EVM world operate on social consensus: when the chain is exploited, the community decides how to respond. That is the reality. And that is a strength.

The soft fork froze the attacker. The logical next step is a hard fork that removes their illicit gains and returns the funds to the victims. If we don’t do this, we signal that “low-risk DeFi” isn’t actually low-risk at all. And we undermine the trust that makes this ecosystem useful beyond speculation.

If we want adoption, we can’t have normal users swallow a 100% loss of their principal in what was supposed to be low risk defi savings. Hard fork, fix it and move on.

14 Likes
3

I’m all for a hard fork as it would go down in blockchain history as a huge example of decentralized coordination against malicious actors.

Just have to mention that this “low risk defi” narrative is off-putting to a lot of people, implying that people using experimental/innovative protocols deserve to lose their money whereas users of protocols that have solidified their presence should always be given priority.

4 Likes
4

I truly do not understand…

Why does Gnosis ltd seem to want to prevent this particular vulnerability, whereas for the oracle’s one in the EURe/sDAI pool, you decided to hide it from your users, thereby causing hundreds of thousands of euros in additional losses.

In the same way that I support Kraken doing everything to rescue the money it holds in custody, I also support Gnosis Ltd. using its multisig power to protect people.

My question is why you decided to deliberately cause your users to lose thousands of euros and are now claiming you want to save them.

Thank you for your response.

4 Likes
5

Congratulations on achieving this. I can well imagine the stress of the last few days. For me, as a regular, curious user of blockchain technology who observes and tries out many things but ultimately lacks in-depth knowledge of the technology, it was quite an eye-opener.

I had no liquidity in the affected pools. I had lost quite a bit in previous hacks, but I wasn’t affected by that this time. And even if I had been, it wouldn’t have bothered me; I’m sufficiently diversified, and when it comes to money, in my opinion, this is no different with DeFi than with TradeFi—you always have to keep the total loss of any position (with varying probabilities) in mind.

What personally affected me in this case wasn’t a potential loss, but the inability to be part of the solution, despite my involvement as a validator of Gnoschain. As a Dappnode user, I could only watch and, if I’m in favor of a soft fork, leave the work to others.

Therefore, I’ll probably have to delve a bit deeper into the technology in the future if I truly want to be part of the system. Hopefully, I’ll find the time.

Regarding the soft fork and potential hard fork: I’m not sure if this will have a general impact on the acceptance of blockchain technology in general or the Gnosischain ecosystem in particular. Ideally, the system will continue to improve, making such events increasingly unlikely, which, in my opinion, strengthens trust in the technology. If I have to rely on measures like the ones currently being implemented, it might limit my acceptance, as these measures become increasingly difficult to implement with growing decentralization.

7 Likes
6

maybe cause it wasn’t @Karpatkey who is responsible for this;-)

1 Like
7

Congratulations on the soft fork! I’m still not sure whether this is good or bad, but leaving opinions aside, it’s impressive to see the capabilities of the team behind Gnosis and the speed at which they delivered a fix in record time.

If the HF is approved, Gnosis Chain will set a precedent and differentiate itself from other chains, even Ethereum. Also, if this goes through, I think Gnosis should consider developing mechanisms to act quickly in case of future hacks, ideally having the ability at the client level to lock any address involved in an incident. I don’t want to go deep into implementation details, but as an example, consensus could read from a specific contract to block certain addresses from operating on the chain. I know this is controversial, but since it has already been done, it doesn’t make much sense to go through the same stressful situation again if another hack ever happens.

6 Likes
8

Fixing Balancer? It’s an easy win.
An anonymous exploit, no politics, no pride. Patch it, move on. Job done.

The pool? That’s a different beast.
Admitting failure here means conceding that kpk isn’t just bad at DeFi, they’re structurally incompetent. $20M+ in fees paid? Burnt capital. Equity stake? Zero.

And yet Gnosis still lets kpk skim 2% AUM off the shared treasury with SafeDAO - including native Safe tokens that make up 90%+ of it. They do nothing. Literally. See for yourself: etherscan

The fact that kpk isn’t even embarrassed to siphon funds for zero value? That’s the real scandal.

Gnosis needs to wake up.
You don’t nurture a tumor. You excise it. Delay only compounds dysfunction.

4 Likes
9

This was only due to the fact that there was a need to act fast to get the soft fork code to as many big validators as possible without showing its code to the public and alert the hacker before it’s too late for them.

Unfortunately to give it to dappnode users it would have needed to be totally public as is now and would alert the hacker to move the funds and it would all end up having been in vain.

4 Likes
10

sure, this wasn’t meant as a complain against the procedure but against my missing skills to run a node without dappnode…

edit: but ofc…as more validators are run by people not personally known by the inner circle makes a procedure like this more complicated, even if they are not dappnode users.

1 Like
11

There needs to be an EMERGENCY vote regarding Karpatkey’s engagement.

The fees are absolutely atrocious, the yield is barely above stETH + sUSDS if at all, and we’re exposed to incestuous misalignment from excessive exposure to Balancer because of Karpatkey’s relationship there.

The fact that multiple GNO holders are now speaking up about this issue, and the dissatisfaction with Karpatkey has been mentioned numerous times now just signals the need for this referendum. I wouldn’t want there to be any implication that the Gnosis team is turning a blind eye to these concerns.

I would like to ask @ernst @StefanGeorge (and Martin, I am only able to mention 2 users in my first post) for an emergency referendum regarding Karpatkey’s duties, with ALL options on the table. Otherwise, I will be engaging with the proposal process myself to ensure that GNO holders are given the choice to disengage with this deeply -EV relationship.

I envisage that a large GNO voting block will be supportive of punitive action, and a regression if not full halt to Karpatkey’s responsibilities.

Thank you.

6 Likes
12

My own take on @Karpatkey : the are doing a good job in general but don’t communicate about things that failed (like the EURe Issue) and don’t involve the community about decisions where funds, voting power and possibly voting incentives (not sure if these are placed by Karpatkey or Gnosis ltd). If the mind behind these decisons, from whom ever they came, are posted here for open discussion it would be, imho, much better as only afterwards minutes. Could also attract ppl interested in this stuff to join this forum.

2 Likes
13

The problem isn’t so much whether we like kpk or not; I was one of the biggest supporters of kpk, Gnosis, GnosisChain, Safe, and CowSwap.
I was one of the first French speakers to actively promote the use of Safe and Cowswap, I’ve been involved with Gnosis for years, I even participated in the launch of Conditional Tokens back in the day (my first airdrop).

There are two aspects to the problem.

First, the hypocrisy of the Gnosis Ltd. leadership. They claim to have their users’ best interests at heart and want to create a fairer world, yet the facts show the opposite. We’ve had almost no contact between the DAO, which represents the community, and Gnosis LTD for months, except to sign checks. They hid a vulnerability for months to protect their personal interests, and I’ll also remind you that the vulnerability was still present; $4M of the $9M would not have been lost if they had acted. The inner circle of Gnosis VIPs wasn’t even participating in votes anymore; it was a struggle to get anything passed.

Second, regarding this specific exploit, we still haven’t had clear communication on who lost what. In a hypothetical example, of course, where more than 50% of the losses belong to the DAO and two or three multi-millionaires linked to Gnosis, would the hard fork still be the best option? Because I see comparisons being made between TheDAO hard fork and GnosisChain, but the Ethereum wasn’t held in custody in a multisig controlled mainly by Gnosis or companies spun off from Gnosis. I’m all for a non-neutral chain; I love the Cosmos-style app-chain approach, with circuit breakers on the exits and more active, direct community involvement. But for that, you need a strong bond between the different stakeholders, a bond we haven’t had with Gnosis LTD for months.

To summarize, I am against Gnosis’s hypocritical communication, I want Gnosis, kpk, and Balancer to be held accountable and explain to us why they denied the problem for almost a year, and finally, I am worried about the legal repercussions as well as the slippery slope of a hardfork on GnosisChain.

4 Likes
14

i brought quite a number of people to gnosischain and gnosispay.

most did the sdai/eure liquidity-providing thing and used their funds through gnosispay card.

everyone assumed this be a mature defi solution with trustworthy services.

i have even one guy referred who sold a house and put it all into eure/sdai LP.

you can imagine how my phone rings now for being the one recommending this path.

i would really love to see the LPs made whole.

on-chain it looks like the funds are still there, just the LP-tokens worth shifted, so i still have hope.

maybe its to consider to have some kind of help-line or discord place where affected LPs can talk to the higher-ups in charge.

thank you for all you did yet and i hope this to turn out fine for everyone.

1 Like
15

We’ve been following the discussion regarding the Soft Fork and welcome the position taken by Gnosis leadership.

As noted in the Balancer post, the Gnosis community was affected in the following pools:

  • EURA/EURe
  • WETH/wstETH
  • GNO/osGNO
  • EURe/sDAI

The USDC.e/USDT/sDAI pool was not impacted, and kpk successfully withdrew 100% of the funds. Other smaller pools in which GnosisDAO held no liquidity were neither targeted nor affected.

As liquidity managers, we’re responsible for maintaining conditions that allow Gnosis Chain to settle, operate, and remain usable for the broader community. We remain committed to this purpose and will continue to work with the community to the best of our ability.

3 Likes
16

Congrats on the flawless strategy: 100% withdrawn from unaffected pools and 0% from affected ones.

The lack of engagement with any of the criticism raised just proves kpk is grandfathered into GnosisDAO and immune to accountability.

6 Likes
17

ETA of snapshot voting for hard fork?