Snowbridge - Tests refactor #8014
Conversation
| @@ -44,7 +44,7 @@ fn transfer_and_transact_in_same_xcm( | |||
| let asset_hub_location = BridgeHubWestend::sibling_location_of(AssetHubWestend::para_id()); | |||
|
|
|||
| // TODO(https://github.com/paritytech/polkadot-sdk/issues/6197): dry-run to get local fees, for now use hardcoded value. | |||
| let ah_fees_amount = 90_000_000_000u128; // current exact value 79_948_099_299 | |||
| let ah_fees_amount = 700_000_000_000u128; // TODO not sure why this increased so much | |||
There was a problem hiding this comment.
Hmm, I don't like the uncertainty with this TODO. We should try understand what is happening here.
Was this test not failing in #7402?
There was a problem hiding this comment.
Extracting the Weth to a common place had some side effects. Some tests used a different weth address, with an asset owner that was not Snowbridge sovereign. Using the same Weth address (that is registered in the emulated tests genesis config), resulted in having to make some changes to the test setup and asset owner. The fee increase doesn't make sense though, will investigate.
The test was passing in #7402.
There was a problem hiding this comment.
This was an interesting one. The root cause is the min_balance of the registered Weth asset. In the original version of this test, the min_balance of Weth was 1_000. In the updated test, the min_balance was 1_000_000_000. The weight purchased should be at least the minimum balance of the asset provided:
- https://github.com/paritytech/polkadot-sdk/blob/master/cumulus/primitives/utility/src/lib.rs#L178-L182
- https://github.com/paritytech/polkadot-sdk/blob/master/cumulus/parachains/common/src/xcm_config.rs#L61
- https://github.com/paritytech/polkadot-sdk/blob/master/substrate/frame/assets/src/types.rs#L346-L360
I set the min_balance back to 1_000, which lowered the required weight fee again.
There was a problem hiding this comment.
We need to think about the impact on mainnet for using Ether and Weth to pay fees. Both their min_balance are 15_000_000_000_000. Maybe this is not a problem, just thinking about it explicitly. It means the min amount provided to buy weight needs to be at least 15_000_000_000_000. I am not sure if the PayFees instruction works the same way.
Added internal issue to track this: SNO-1440
There was a problem hiding this comment.
I'd assume on mainnet we will create a liquidity pool (ETH, DOT) and rely on SwapFirstAssetTrader as the primary trader. IIUC, TakeFirstAssetTrader here is just a fallback without a liquidity pool, in this case, ED acts as a simplified version of the swap rate.
Resolves: SNO-1427
Closes: #7977