Proposal to Integrate Chainlink Price Feeds

In general it would be a good idea to create an autonomous proposal. But I think the chances, that it will go through, are very low. So I don’t think it’s worth the efforts.

If you want to understand the reason why, check here. Compound is not governed by a democracy, but is in the hands of a few. The top 6 together have more than 50% voting power. These are Compound investors, are affiliated with the Compound founders or are the Compound founders. It seems, that the Compound founders are not willing to switch to Chainlink and I’m pretty sure, that the other 4 won’t vote against them.

2 Likes

I would disagree, as your post seems to be merely a conjecture?
An autonomous proposal has already passed before, namely this

If the top holders reject a proposal to use a better oracle without submitting an alternative solution, I would probably be weary of leaving and keeping holdings in the platform.

So certainly after this DAI debacle, a solution needs to be put forward.

1 Like

I don’t say, that autonomous proposals haven’t a chance to pass in general. I meant this specific proposal. Perhaps you are right, that it’s worth trying it and make it transparent that way, what the different parties think about it.

And yes, the oracle needs to be fixed asap.

In the first step at least 100,000 delegated votes are needed, so the autonomous proposal can be converted into a governance proposal. Not sure, if it is already necessary to have the needed code changes available. Would be great, if someone who knows this exactly, would give some hints.

1 Like

This has already been discussed and explained at length here.

Chainlink is not a good oracle design, as the company is incentivezed to obfuscates a clear problem, by adding unnecessary layers (aggregators, nodes) that increase the attack surface of any system that integrates with them. With regards to improving the Open Oracle, the path forward is to add more high-liquidity exchange reporters to the current oracle view. This should not be the highest priority action, as it does not address the security issue at hand which can only be resolved by reducing the amount DAI outstanding debt in the system.

3 Likes

As I already explained to you here, none of these claims about Chainlink is true. I will repost my comment here to provide context to everyone.

This is simply not true and a bit disingenuous. Chainlink simply wants to provide the DeFi ecosystem with the most secure price oracle solution with the highest quality data and this involves pulling from data aggregators who have full time monitoring teams to ensure manipulation is prevented 24/7. Taking a simple median from a select few exchanges is simply not an adequate oracle solution as it will always be vulnerable to market coverage issues and manipulation attacks around volume shifts and consolidations. As I described in the post linked above, the three layers of aggregation (data level, node level, network level) prevents any single source of truth. Relying on a single source of true is what played a significant factor in the $100M false liquidations of Compound users, and at the very least made the attack much easier and cheaper to pull off.

These are not “unnecessary layers” but layers of redundancy to ensure smart contracts always receive price data that reflects the true market wide price and not that of a single or a few exchanges. As I described in my proposal, Chainlink is directly secured by cryptoeconomics through an opportunity cost of losing future income if a node is malicious, as well as losing their reputation as an DevOps infrastructure provider. This is why we have never seen a successful attack against the Chainlink network, because the incentives work and ensure correct data is always posted on-chain.

As I also described, taking a simple median across a select few exchanges who change their infrastructure to support signed data compatible with Ethereum still does not provide adequate market coverage because it doesn’t take into account volume and liquidity differences across exchanges like data aggregators do. Here is a repost of my comment for more context.

I think you’re missing the key point of market coverage. The issue with Compound’s oracle during this DAI manipulation attack was that it did not report the true market wide price. If an attacker manipulates the entire market (across every single exchange), then yes all oracles would be affected at that point, but that’s only because the true market wide price was changed, but that’s not what happened during this event, only a single exchange (Coinbase) was manipulated. That’s the nuance here, market coverage raises the cost of attack to highest degree possible, and while it doesn’t prevent market manipulation altogether, but does make it as expensive as possible and ensures protocols always receive the true market wide price.

Like we discussed at length in the governance discord channel, the DAI liquidation event was a mixture of two factors. Too much DAI debt taken out AND a lack of price oracle market coverage. Coinbase is not the only liquid market for DAI and the DAI/USD Coinbase trading pair only tracks 4.75% of DAI’s daily volume according to CoinGecko. The vast majority of the volume is from Uniswap, which Chainlink adequately tracks today through its usage of data aggregators. Both of these factors (debt and market coverage) need to be solved, but we shouldn’t be ignoring the latter whatsoever. We can solve both issues at the same time, because market coverage played a significant factor in the false liquidation of $100M in user funds. The longer the Compound protocol goes without ensuring market coverage, the more exposed user funds are to further oracle manipulation.

2 Likes

If only alot more high-liquidity exchanges had/adopted the open oracle reporting api. Seems your solutions is to make this high liquidity exchanges provide tooling for us, what incentive is there for them to spend resources for this?? I know coinbase has, because they’re an early investor to compound.

How do you intend to do that? By artificially limiting market size?

1 Like

How do you intend to do that? By artificially limiting market size?

This is being discussed in this thread DAI Market Risk

1 Like

This is all incorrect and outright false advertising from a company with obvious interests. Adding layers of obfuscation to what should be a simple and transparent oracle design is detrimental to security. I’ll restate what I’ve said numerous times before “Full market coverage” is a silly idea, in reality most markets are low volume/liquidity and a lot of CEXs have fake volume and liquidity. You want to be very careful in selecting what markets to include in a price feed, and medianizing the set of carefully selected high-liquidity exchanges is the best possible way to do this.

A bit off-topic, but It’s qiute lamentable, and time-consuming, that this industry has to deal now with a clear and obvious campaign of aggressive astroturfing by a company to promote a subpar technical solution constantly and everywhere. The arguments on this topic have become repetitive and tiresome, so I’ll just leave it at that.

3 Likes

I don’t the gaslighting here is necessary, this is simply a conversation where we can discuss market coverage and why Compound’s current oracle is not sufficient for the value it secures today. The Chainlink network doesn’t obfuscate anything, but is incredibly transparent into its operation (feeds.chain.link, market.link, reputation.link). The usage of data aggregators ensures price data tracks all trading environments, preventing market coverage issues and this includes tracking fake volume and market manipulation. You want multiple layers of redundancy to ensure there is no single point of failure.

I’m not going to repeat myself on the importance of market coverage because Chainlink has already proven its ability to properly secure a wide range of dApps like Aave and Synthetix for long periods of time, which have never experienced market coverage or data quality issues like Compound did. This isn’t a Chainlink shill, this is “please fix your oracle before it breaks again and users lose more money” plea that’s coming from someone who has seen these oracle manipulation issues time and time again. Taking a median across exchanges where volume can shift overnight to exchanges not being tracked doesn’t solve the problem.

4 Likes

My biggest concern is that an event involving the current oracle system will happen again. If Chainlink provides a solution that prevents this from occurring in the future, we should take a hard look at it. As a user, I want to make sure that I’m protected from this from happening in the future and that my positions aren’t liquidated when they should not have been.

Chainlink had better market coverage, which is obvious because it did not report this crazy DAI price. The bottom line is Compound’s current infrastructure got us here, and something needs to change.

I don’t know of any successful protocol that recognizes a problem and does not seek to address it.

4 Likes

I don’t understand your concerns about obfuscation. All nodes are run by known, established entities and you can see each price update they post on-chain. It’s not like they can post prices then hide from everyone, as they would be immediately identifiable. Full market coverage also doesn’t mean you are just taking an average of all exchanges. It means you are taking in data from all exchanges and then weighting it by volume, while also removing outliers, fake exchange volume, etc. By having such a setup, you can account for volume shifts between exchanges or if it starts flowing to a new exchange. This is much more scalable and doesn’t require constant shifts in the underlying price calculation to account for changes in the trading market.

Also, Compound can still keep a circuit breaker in place should they want a fallback mechanism, similar to what they have today in Uniswap.

6 Likes

Great points @The_Crypto_Oracle. Honestly I just want to know my funds are safe, and Chainlink’s data sources/methods seem as reliable as it gets for mitigating risk in similar situations, and we can implement them quickly.

I’m happy to see that we have a COMP distribution proposal in the works. These sort of changes will be very useful. However, I can’t help but feeling like the oracle issue is a more pressing issue to resolve right now.

4 Likes

Agreed. @rleshner mentioned wanting to implement additional price feed safeguards on Discord last night, which is great, but I didn’t hear any specifics around what he is envisioning from a technical standpoint. It would be very helpful to get some more clarity here so we can get this conversation moving forward (and a fix in place ASAP).

4 Likes

How can happen outstanding debt in DAI?
The Variable Interest Rate Mechanism was not effective?
Because DAI borrowing interest rate was around 4% all time. Higher interest rates on borrowing assets incentivize users to repay high expensive positions.
I was personally borrow DAI because of lower interest rates than other stablecoins.
Can you, if it is not a problem, put a link where the DAI market is analyzed precisely because of that anomaly? And where are the potential risks presented? I would like to take a closer look at your argument.
Thank you

1 Like

no, but they have had problems with it earlier.

1 Like

@TennisBowling Could you provide more context? I hadn’t heard about this, what oracle or liquidation issues did Aave have recently?

3 Likes

@tennisbowling please write also some details, what you mean by this.

3 Likes

I’ll echo the others in that it would be helpful if you could expand upon these concerns in any detail @TennisBowling.

I feel it is pretty clear the current oracle framework needs a rethink (I’m not seeing many here argue against this), and so far Chainlink sounds like the best solution proposed thus far due to the massively increased market coverage w/ minimal development. There have been other ideas floated (more reporters, tightening bounds etc.) that could improve the current situation, but IMO only nominally by comparison.

I’ve heard concerns about centralization, but it’s hard to understand those concerns when people seem ok using Coinbase Pro as our reporter, meanwhile CL is using a large network of independent data sources. I’ve also heard concerns about obfuscation, but again that is hard to understand when I can easily view granular node/feed data and reputation info.

I’ll admit the LINK marines/frogs can be a little much! Part of me feels like the negative sentiment here could be more a knee-jerk reaction to that than the technicals, which I feel that is a shame because from what I can tell the actual team is one of the most professional in the space, and the technicals are all that really matters.

Again if there is something I’m missing, I would love to know more details so we can discuss.

4 Likes

Chainlink is the best solution, but if in Chainlink exists some point of failure how we can improve that with the existing price discovery model?
Existing price discovery solution is centralized and every argument against failed because at some point in time DAI price on Coinbase Open Price Feed wasn’t aligned with the global DAI price. We can’t say that is because of outstanding debt in DAI on the protocol because in that scenario protocol has big impact on price at existing oracle model. If Compound caused stablecoin volatility on an existing oracle solution then it just doesn’t work.

Agree, there may be more constructive reactions when something like this happens again

3 Likes

FYI - Coinbase experienced a 60% deviation from the market-wide price yesterday on the SNX/USD pair:

Compare this to Coingecko:

Had SNX been a collateral type on Compound we likely would have seen another false liquidation event. We need to expand market coverage via Chainlink, plain and simple.

7 Likes