Snowbridge v2 - Inbound Queue #6697

claravanstaden · 2024-11-28T17:15:48Z

Description

Implements the Inbound Queue for Snowbridge v2.

Integration

This PR adds:

A new pallet called EthereumInboundQueueV2 with a submit extrinsic.
A new runtime API called InboundQueueApiV2 with a method dry_run, that takes an Ethereum command and converts it to Xcm and provides the execution fee on BH and static fee part of the AH execution.

Since this is a new pallet, no breaking changes are made to the EthereumInboundQueue pallet.

Review Notes

The Inbound Queue v2 pallet expects an abi-encoded envelope from Ethereum. Once decoded, the payload parameter is SCALE decoded into a v2::Message. The message contains:

origin: The origin address of the user on Ethereum
assets: A vector of assets that will be placed in the holding on AH. Can be a native ERC-20 or a foreign ERC-20 (Polkadot native assets). The XCM instructions that are used differ per asset type (ReserveAssetDeposited for native ERC-20 tokens and WithdrawAsset for Polkadot native assets). The user on Ethereum specifies additional xcm deposit the asset into a beneficiary account.
xcm: User provided xcms, to deposit the asset to a beneficiary, provide further fees or other instructions such as Transact (for example, to register a token)
claimer: The claimer on AH, in case funds are trapped so it can be claimed.

The relayer pays a DOT fee that covers:

The execution fees on BH
The static xcm part of the message on AH

The DOT fee is burnt from the relayer account and teleported to AH.

Messages may be processed out of order. Nonces are stored in a sparse bitmap for space-efficient storage.

Co-authored-by: Vincent Geddes <[email protected]>

Fix register token

bridges/snowbridge/pallets/inbound-queue-v2/src/lib.rs

franciscoaguirre · 2025-01-27T16:07:38Z

bridges/snowbridge/pallets/inbound-queue-v2/src/lib.rs

+
+		pub fn send_xcm(xcm: Xcm<()>, dest_para_id: u32) -> Result<XcmHash, Error<T>> {
+			let dest = Location::new(1, [Parachain(dest_para_id)]);
+			let (message_id, _) = send_xcm::<T::XcmSender>(dest, xcm).map_err(Error::<T>::from)?;


You're not charging the delivery fee here, is that intended?
I know this fee should be included in the relayer reward, but that's because it's charged here, right? Else it's just more reward for the relayer.

@yrong @vgeddes the delivery fee is charged from the BridgeHub sovereign account, right? Do we need to deduct it from the relayer account on BH?

I prefer to deduct it from the relayer account. Ideally in V2 we don't need to prefund BridgeHub sovereign account any more.

UI can estimate it off-chain with the runtime api

During v2 design I remember we found alternatives to remove all cases where v1 uses BH/AH SA accounts for covering costs.

So yes, let's charge the relayer directly. Either charge their account or better UX would be to subtract from the reward.

Good catch Francisco. @claravanstaden this delivery fee needs to be deducted from the relayer's account.

Its the responsibility of the user initiating the bridge transfer to provide a relayer reward that covers the delivery fee.

Also the delivery fee also needs to be included in the fee returned by the dry-run API

Fixed in 6f769cc

Delivery fee is deducted from relayers account on BH.

bridges/snowbridge/pallets/inbound-queue-v2/src/test.rs

bridges/snowbridge/primitives/router/src/inbound/v2.rs

franciscoaguirre · 2025-01-27T17:59:12Z

/cmd fmt

franciscoaguirre · 2025-01-27T18:03:15Z

bridges/snowbridge/pallets/inbound-queue-v2/src/lib.rs

+			let (xcm, _relayer_reward) =
+				Self::do_convert(envelope.message, origin_account_location.clone())?;
+
+			// Todo: Deposit fee(in Ether) to RewardLeger which should cover all of:


When will this be uncommented? Do we need to create a tracking issue?

The reward functionality is added in #6578. Once that is merged, we can uncomment this. Added tracking issue: #7356

In the interim, I've added a dummy trait until the new pallet is ready: 6f769cc

Co-authored-by: Francisco Aguirre <[email protected]>

acatangiu · 2025-01-30T10:01:34Z

bridges/snowbridge/pallets/inbound-queue-v2/README.md

@@ -0,0 +1,3 @@
+# Ethereum Inbound Queue
+
+Reads messages from Ethereum and sends it to intended destination on Polkadot, using XCM.


Suggested change

Reads messages from Ethereum and sends it to intended destination on Polkadot, using XCM.

Reads messages from Ethereum and sends them to intended destination on Polkadot, using XCM.

This could also use more details like architecture overview, APIs, usage patterns, etc.

acatangiu · 2025-01-30T10:03:43Z

bridges/snowbridge/pallets/inbound-queue-v2/runtime-api/README.md

please also add at least some example(s) - how it can/should be used

bridges/snowbridge/pallets/inbound-queue-v2/runtime-api/src/lib.rs

bridges/snowbridge/pallets/inbound-queue-v2/src/api.rs

acatangiu · 2025-01-30T10:09:30Z

bridges/snowbridge/pallets/inbound-queue-v2/src/api.rs

+	// Calculate fee. Consists of the cost of the "submit" extrinsic as well as the XCM execution
+	// prologue fee (static XCM part of the message that is execution on AH).
+	let weight_fee = T::WeightToFee::weight_to_fee(&T::WeightInfo::submit());
+	let fee: u128 = weight_fee.try_into().map_err(|_| Error::<T>::InvalidFee)?;


I only see the submit() cost here, where is the "XCM execution prologue fee" added?

Turns out this runtime api implementation was incorrect. We don't need to estimate any fees here, since there are already other runtime-apis that can estimate the cost of calling submit()

Instead we actually only need a runtime api to convert the inbound message to an XCM that can be dry-run on AH.

New implemention in 6f769cc where is renamed to convert_message

acatangiu · 2025-01-30T11:12:57Z

bridges/snowbridge/pallets/inbound-queue-v2/src/lib.rs

+		pub fn account_to_location(account: AccountIdOf<T>) -> Result<Location, Error<T>> {
+			let account_bytes: [u8; 32] =
+				account.encode().try_into().map_err(|_| Error::<T>::InvalidAccount)?;
+			Ok(Location::new(0, [AccountId32 { network: None, id: account_bytes }]))
+		}


you can replace this fn with a bound on AccountIdOf<T>: Into<Location> as convertors already exist.

acatangiu · 2025-01-30T11:19:12Z

bridges/snowbridge/pallets/inbound-queue-v2/src/lib.rs

+
+		pub fn send_xcm(xcm: Xcm<()>, dest_para_id: u32) -> Result<XcmHash, Error<T>> {
+			let dest = Location::new(1, [Parachain(dest_para_id)]);
+			let (message_id, _) = send_xcm::<T::XcmSender>(dest, xcm).map_err(Error::<T>::from)?;


During v2 design I remember we found alternatives to remove all cases where v1 uses BH/AH SA accounts for covering costs.

So yes, let's charge the relayer directly. Either charge their account or better UX would be to subtract from the reward.

bridges/snowbridge/pallets/inbound-queue-v2/src/lib.rs

acatangiu · 2025-01-30T11:35:02Z

bridges/snowbridge/pallets/inbound-queue-v2/src/types.rs

I personally found this file confusing/obfuscating.

I would rather define pub type Nonce<T> = SparseBitmapImpl<crate::NonceBitmap<T>>; in the main inbound-queue-v2/src/lib.rs file where NonceBitmap<T> is also defined.

Seeing how this file doesn't define other types needed externally, it could then be dropped.

acatangiu · 2025-01-30T11:38:32Z

bridges/snowbridge/pallets/inbound-queue-v2/src/lib.rs

+		}
+	}
+
+	/// The nonce of the message been processed or not


Suggested change

/// The nonce of the message been processed or not

/// StorageMap used for encoding a [SparseBitmapImpl](<link-to-SparseBitmapImpl-file>) that tracks whether a specific nonce has been processed or not. Message nonces are unique and never repeated.

acatangiu · 2025-01-30T12:05:59Z

bridges/snowbridge/pallets/inbound-queue-v2/src/lib.rs

+			// Set nonce flag to true
+			log::info!(target: "snowbridge-inbound-queue:v2","💫 setting nonce to {:?}", envelope.nonce);
+			Nonce::<T>::set(envelope.nonce.into());
+
+			// Attempt to forward XCM to AH
+			let message_id = Self::send_xcm(xcm, T::AssetHubParaId::get())?;


any error will in practice revert/rollback storage changes, but let's be paranoid and mark nonce as processed only on certainty of success:

Suggested change

// Set nonce flag to true

log::info!(target: "snowbridge-inbound-queue:v2","💫 setting nonce to {:?}", envelope.nonce);

Nonce::<T>::set(envelope.nonce.into());

// Attempt to forward XCM to AH

let message_id = Self::send_xcm(xcm, T::AssetHubParaId::get())?;

// Attempt to forward XCM to AH

let message_id = Self::send_xcm(xcm, T::AssetHubParaId::get())?;

// Set nonce flag to true

log::info!(target: "snowbridge-inbound-queue:v2","💫 setting nonce to {:?}", envelope.nonce);

Nonce::<T>::set(envelope.nonce.into());

fixed in 6f769cc

@yrong mentioned if we increment the nonce after the send we risk re-entrancy attacks.

acatangiu · 2025-01-30T12:08:25Z

bridges/snowbridge/pallets/inbound-queue-v2/src/test.rs

+}
+
+#[test]
+fn test_submit_happy_path() {


maybe add another attempt here after the successful one and verify nonce verification fails

acatangiu · 2025-01-30T12:10:14Z

bridges/snowbridge/pallets/inbound-queue-v2/src/weights.rs

can we move these weights under mock.rs or tests.rs? I don't want them to be usable outside of tests.

acatangiu · 2025-01-30T12:27:31Z

bridges/snowbridge/primitives/core/src/fees.rs

+/// Burns the fees embedded in the XCM for teleports.
+pub fn burn_fees<AssetTransactor, Balance>(dest: Location, fee: Balance) -> DispatchResult


this is not used anywhere in this PR

you should call it in inbound_queue_v2::submit() before or within Self::send_xcm()

it could also be used in inbound-queue-v1 and you can remove old fn burn_fees()

Good catch, this was obsolete code. Deleted in 33f8e6d

In V2, we don't need to burn any asset for teleporting. No sovereign accounts, etc.

yes, with fees paid exclusively using eth, there are no more teleports involved so no need to locally burn anything on BH

acatangiu · 2025-01-30T12:37:48Z

bridges/snowbridge/primitives/core/src/fees.rs

+/// Burns the fees embedded in the XCM for teleports.
+pub fn burn_fees<AssetTransactor, Balance>(dest: Location, fee: Balance) -> DispatchResult


another note about it:

since it is defined in primitives, maybe make it less opinionated:

Suggested change

/// Burns the fees embedded in the XCM for teleports.

pub fn burn_fees<AssetTransactor, Balance>(dest: Location, fee: Balance) -> DispatchResult

/// Burns the fees embedded in the XCM for teleports.

pub fn burn_assets_for_teleport<AssetTransactor>(dest: Location, assets: Assets) -> DispatchResult

this way you can use it for various types of asset(s)

acatangiu · 2025-01-30T12:43:34Z

cumulus/parachains/integration-tests/emulated/tests/bridges/bridge-hub-westend/Cargo.toml

+penpal-emulated-chain = { workspace = true }
 rococo-westend-system-emulated-network = { workspace = true }


you don't need to take direct dependency to penpal-emulated-chain here, it is already included in rococo-westend-system-emulated-network

acatangiu · 2025-01-30T12:44:11Z

...rachains/integration-tests/emulated/tests/bridges/bridge-hub-westend/src/tests/snowbridge.rs

@@ -21,11 +21,13 @@ use codec::{Decode, Encode};
 use emulated_integration_tests_common::{PENPAL_B_ID, RESERVABLE_ASSET_ID};
 use frame_support::pallet_prelude::TypeInfo;
 use hex_literal::hex;
+use penpal_emulated_chain::PARA_ID_B;


Suggested change

use penpal_emulated_chain::PARA_ID_B;

use rococo_westend_system_emulated_network::penpal_emulated_chain::PARA_ID_B;

acatangiu · 2025-01-30T12:47:01Z

...rachains/runtimes/assets/asset-hub-westend/src/weights/xcm/pallet_xcm_benchmarks_fungible.rs

+
+	pub fn exchange_asset() -> Weight {
+		// Proof Size summary in bytes:
+		//  Measured:  `159`
+		//  Estimated: `6196`
+		// Minimum execution time: 87_253_000 picoseconds.
+		Weight::from_parts(88_932_000, 6196)
+			.saturating_add(T::DbWeight::get().reads(9))
+			.saturating_add(T::DbWeight::get().writes(4))
+	}


This should be benchmarked eventually, but it's fine like this for Westend - it will get benchmarked outside of this PR at a later point

cumulus/parachains/runtimes/bridge-hubs/bridge-hub-westend/src/bridge_to_ethereum_config.rs

acatangiu · 2025-01-30T12:53:05Z

cumulus/parachains/runtimes/bridge-hubs/bridge-hub-westend/src/bridge_to_ethereum_config.rs

@@ -71,7 +74,9 @@ parameter_types! {
 	};
 	pub AssetHubFromEthereum: Location = Location::new(1,[GlobalConsensus(RelayNetwork::get()),Parachain(westend_runtime_constants::system_parachain::ASSET_HUB_ID)]);
 	pub EthereumUniversalLocation: InteriorLocation = [GlobalConsensus(EthereumNetwork::get())].into();
+	pub WethAddress: H160 = H160(hex_literal::hex!("fff9976782d46cc05630d1f6ebab18b2324d6b14"));


this is not used anywhere

Suggested change

pub WethAddress: H160 = H160(hex_literal::hex!("fff9976782d46cc05630d1f6ebab18b2324d6b14"));

acatangiu · 2025-01-30T16:40:37Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+	/// The full value of the assets.
+	pub value: u128,


can you add more info here? what is the full value of the assets?

There's an assets field above that can hold arbitrary number of different tokens with arbitrary amounts of each.
Is that field related to this value: u128 field? If yes, how? What is the relationship between them? If not, then what is this value?

pseudo-random guess: does it represent an estimate of the eth value of all assets above?

Oops, that comment is very incorrect - I should have picked it up in my reviews.

message.value is the amount of native ether that was bridged from Ethereum.

message.assets holds other kind of assets, ERC20, NFTs, etc.

acatangiu · 2025-01-30T16:48:33Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+			let xcm_string = hex::encode(message.xcm.clone());
+			log::info!(target: LOG_TARGET,"found xcm payload: {:x?}", xcm_string);


Suggested change

let xcm_string = hex::encode(message.xcm.clone());

log::info!(target: LOG_TARGET,"found xcm payload: {:x?}", xcm_string);

acatangiu · 2025-01-30T17:00:07Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+			if let Ok(versioned_xcm) = VersionedXcm::<()>::decode_with_depth_limit(
+				MAX_XCM_DECODE_DEPTH,
+				&mut message.xcm.as_ref(),
+			) {
+				if let Ok(decoded_xcm) = versioned_xcm.try_into() {
+					message_xcm = decoded_xcm;
+				} else {
+					log::error!(target: LOG_TARGET,"unable to decode xcm");
+				}
+			} else {
+				log::error!(target: LOG_TARGET,"unable to decode versioned xcm");
+			}


failing this should fail the whole conversion, no?

short-form suggestion:

Suggested change

if let Ok(versioned_xcm) = VersionedXcm::<()>::decode_with_depth_limit(

MAX_XCM_DECODE_DEPTH,

&mut message.xcm.as_ref(),

) {

if let Ok(decoded_xcm) = versioned_xcm.try_into() {

message_xcm = decoded_xcm;

} else {

log::error!(target: LOG_TARGET,"unable to decode xcm");

}

} else {

log::error!(target: LOG_TARGET,"unable to decode versioned xcm");

}

if let Ok(decoded_xcm) = VersionedXcm::<()>::decode_with_depth_limit(

MAX_XCM_DECODE_DEPTH,

&mut message.xcm.as_ref(),

).map(|versioned_xcm| versioned_xcm.try_into()) {

message_xcm = decoded_xcm;

} else {

let xcm_string = hex::encode(message.xcm);

log::debug!(target: LOG_TARGET, "unable to decode xcm: {:x?}", xcm_string);

return Err(ConvertMessageError::InvalidXcm);

}

@acatangiu should we not rather fail silent on invalid xcm, so that assets can be trapped on AH? Otherwise the funds will be locked on Ethereum but not claimable on Polkadot. This was @alistair-singh 's idea.

acatangiu · 2025-01-30T17:01:01Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+			}
+		}
+
+		log::info!(target: LOG_TARGET,"xcm decoded as {:?}", message_xcm);


Suggested change

log::info!(target: LOG_TARGET,"xcm decoded as {:?}", message_xcm);

log::trace!(target: LOG_TARGET,"xcm decoded as {:?}", message_xcm);

acatangiu · 2025-01-30T17:01:57Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+		let network = EthereumNetwork::get();
+
+		// use eth as asset
+		let fee_asset = Location::new(2, [GlobalConsensus(EthereumNetwork::get())]);


nit:

Suggested change

let fee_asset = Location::new(2, [GlobalConsensus(EthereumNetwork::get())]);

let fee_asset_id = Location::new(2, [GlobalConsensus(EthereumNetwork::get())]);

acatangiu · 2025-01-30T17:06:55Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+		let eth: XcmAsset =
+			(fee_asset.clone(), message.execution_fee.saturating_add(message.value)).into();
+		let mut instructions = vec![
+			DescendOrigin(PalletInstance(InboundQueuePalletInstance::get()).into()),


rather than InboundQueuePalletInstance: Get<u8> as generic on this fn, I would use the full location InboundQueueLocation: Get<InteriorLocation> and here:

Suggested change

DescendOrigin(PalletInstance(InboundQueuePalletInstance::get()).into()),

DescendOrigin(InboundQueueLocation::get()),

acatangiu · 2025-01-30T17:10:50Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+			// If the claimer can be decoded, add it to the message. If the claimer decoding fails,
+			// do not add it to the message, because it will cause the xcm to fail.


yes, not clear if we should fail the whole thing or just continue with default claimer...

I have an idea of a more general solution for any type of failures on inbound queue that are caused by protocol incompatibilities (bridge's fault, not the user's):

the inbound queue verifies the authenticity and exact validity of the message through the prover, so if it fails on some decoding step we can consider the message invalid, it will always fail, and hopefully we can revert it the Ethereum side.
So maybe we can mark it in some other InvalidMessage: SparseBitmap storage item or simply emit an InvalidMessage(nonce) event that can be proved back to Ethereum so that it can be rolled back there in case of these kinds of bridge protocol issues.

This is a further enhancement idea to be done outside of this PR, but worth exploring.

This makes sense 👍

Same comment as here: #6697 (comment) It might be better to allow the invalid claimer, so that the funds can be trapped on AH.

The future improvement sounds cool!

acatangiu · 2025-01-30T17:19:38Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+			if let Ok(claimer) = Junction::decode(&mut claimer.as_ref()) {
+				let claimer_location: Location = Location::new(0, [claimer.into()]);


why is the claimer just a Junction / InteriorLocation? seems like a gratuitous limitation. You should leave it as full Location specifiable on Ethereum side.

Could be a local account (just a Junction like now) or it could be a remote location that will be derived into a sov acc.

acatangiu · 2025-01-30T17:20:09Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+				refund_surplus_to = claimer_location.clone();
+				instructions.push(SetHints {
+					hints: vec![AssetClaimer { location: claimer_location }].try_into().unwrap(),
+				}); // TODO


TODO what? :)

acatangiu · 2025-01-30T17:27:40Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+		let appendix = vec![
+			RefundSurplus,
+			// Refund excess fees to the claimer, if present, otherwise to the relayer.
+			DepositAsset {
+				assets: Wild(AllOf { id: AssetId(fee_asset.into()), fun: WildFungible }),
+				beneficiary: refund_surplus_to,
+			},
+		];
+
+		instructions.extend(appendix);


using instructions.push() for each of the appendix elements would avoid allocating another intermediary appendix vec

acatangiu · 2025-01-31T08:55:12Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+		pub const EthereumNetwork: xcm::v5::NetworkId = xcm::v5::NetworkId::Ethereum { chain_id: 11155111 };
+		pub const GatewayAddress: H160 = H160(GATEWAY_ADDRESS);
+		pub const InboundQueuePalletInstance: u8 = 84;
+		pub AssetHubLocation: InteriorLocation = Parachain(1000).into();


Suggested change

pub AssetHubLocation: InteriorLocation = Parachain(1000).into();

acatangiu · 2025-01-31T08:57:55Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+		let instructions = vec![
+			DepositAsset { assets: Wild(AllCounted(1).into()), beneficiary },


Suggested change

let instructions = vec![

DepositAsset { assets: Wild(AllCounted(1).into()), beneficiary },

let instructions = vec![

RefundSurplus,

DepositAsset { assets: Wild(AllCounted(1).into()), beneficiary },

acatangiu · 2025-01-31T09:00:50Z

bridges/snowbridge/primitives/router/src/inbound/v2.rs

+		let mut descend_origin_found = 0;
+		let mut reserve_deposited_found = 0;
+		let mut withdraw_assets_found = 0;
+		while let Some(instruction) = instructions.next() {


nice test! please also add checks for the inner xcm instructions being included

acatangiu · 2025-01-31T09:10:07Z

...hains/integration-tests/emulated/tests/bridges/bridge-hub-westend/src/tests/snowbridge_v2.rs

+		assert_ok!(<AssetHubWestend as AssetHubWestendPallet>::ForeignAssets::force_create(
+			RuntimeOrigin::root(),
+			token_location.clone().try_into().unwrap(),
+			assethub_sovereign.clone().into(),


why use this account as the token admin?

this account is AH's sov acc as derived on BH (has meaning of SA in BH context); then this is used as local account on AH (account has no meaning on AH, it's just a "random account").

acatangiu · 2025-01-31T09:19:21Z

...hains/integration-tests/emulated/tests/bridges/bridge-hub-westend/src/tests/snowbridge_v2.rs

+	});
+}
+
+pub(crate) fn set_up_eth_and_dot_pool(asset: v5::Location) {


you can drop this function and use this existing one instead

acatangiu · 2025-01-31T10:12:23Z

...hains/integration-tests/emulated/tests/bridges/bridge-hub-westend/src/tests/snowbridge_v2.rs

+		assert_expected_events!(
+			AssetHubWestend,
+			vec![
+				RuntimeEvent::ForeignAssets(pallet_assets::Event::Issued { .. }) => {},


check asset id - also check for all assets

example here how to use this macro to also check inner values

acatangiu · 2025-01-31T10:12:36Z

...hains/integration-tests/emulated/tests/bridges/bridge-hub-westend/src/tests/snowbridge_v2.rs

+		// Check that the token was received and issued as a foreign asset on PenpalB
+		assert_expected_events!(
+			PenpalB,
+			vec![RuntimeEvent::ForeignAssets(pallet_assets::Event::Issued { .. }) => {},]


check for all assets including their IDs

acatangiu · 2025-01-31T10:15:07Z

...hains/integration-tests/emulated/tests/bridges/bridge-hub-westend/src/tests/snowbridge_v2.rs

+	let asset_id: Location =
+		[PalletInstance(ASSETS_PALLET_ID), GeneralIndex(RESERVABLE_ASSET_ID.into())].into();
+
+	register_foreign_asset(eth_location());


you should just make ETH (and maybe WETH) part of AH genesis for these tests and not have to register for every test: e.g.

polkadot-sdk/cumulus/parachains/integration-tests/emulated/chains/parachains/assets/asset-hub-westend/src/genesis.rs

Line 78 in 698d9ae

assets: vec![

acatangiu · 2025-01-31T10:17:00Z

...hains/integration-tests/emulated/tests/bridges/bridge-hub-westend/src/tests/snowbridge_v2.rs

+
+	BridgeHubWestend::execute_with(|| {
+		type RuntimeEvent = <BridgeHubWestend as Chain>::RuntimeEvent;
+		let instructions = vec![DepositAsset { assets: Wild(AllCounted(2)), beneficiary }];


Suggested change

let instructions = vec![DepositAsset { assets: Wild(AllCounted(2)), beneficiary }];

let instructions = vec![RefundSurplus, DepositAsset { assets: Wild(AllCounted(2)), beneficiary }];

acatangiu · 2025-01-31T10:18:57Z

...hains/integration-tests/emulated/tests/bridges/bridge-hub-westend/src/tests/snowbridge_v2.rs

+			events.iter().any(|event| matches!(
+				event,
+				RuntimeEvent::Assets(pallet_assets::Event::Burned { owner, .. })
+					if *owner == ethereum_sovereign.clone(),


why is it burned from ethereum_sovereign?

@yrong correct me if I am wrong, but Polkadot native assets are locked in the Ethereum sovereign account, and when sent back to Ethereum it is then burnt.

…b.rs Co-authored-by: Adrian Catangiu <[email protected]>

Co-authored-by: Adrian Catangiu <[email protected]>

claravanstaden and others added 30 commits November 20, 2024 16:31

inbound queue v2

07808bd

adds sparse bitmap

6f238eb

use sparse bitmap

b0115cd

use payfees

c27def2

Update bridges/snowbridge/pallets/inbound-queue-v2/src/lib.rs

93b4516

Co-authored-by: Vincent Geddes <[email protected]>

Update bridges/snowbridge/pallets/inbound-queue-v2/src/lib.rs

3a13134

Co-authored-by: Vincent Geddes <[email protected]>

rename NonceBitmap

7e23462

config T::XcmPrologueFee: Balance

68f5a8e

burn fees

02e8081

fix westend config

ee0bc27

use sendcontroller

4309272

runtime config for sendcontroller

de00eb6

wip

3653b7f

tests

990bf47

xcm fee

9f7829c

adds comments

11839dc

revert sendcontroller

8974696

adds test for xcm register command

75d3c06

update test

c666d23

runtime api

42d688a

adds comment

92e3419

update test

37ecd53

update tests

69a660b

Merge branch 'master' into inbound-queue-v2-clean

43f0d8d

prdoc

71929f6

adds register token test

5735960

enable exchange asset on AH

a935ff2

fix transact

373d63a

Fix for register token

a61ceef

Merge pull request #6 from yrong/ron/inbound-queue-v2-clean

fae2947

Fix register token

claravanstaden added 6 commits January 22, 2025 15:37

fixes

959c662

fix encoding

61e2a28

complete abi encoded payload

1d7ffa0

fix tests

1d3fb32

print register token values

09dfe47

update gateway contract address and adds logs

c9f15ac

franciscoaguirre reviewed Jan 27, 2025

View reviewed changes

franciscoaguirre added the T15-bridges This PR/Issue is related to bridges. label Jan 27, 2025

Merge branch 'master' into inbound-queue-v2-clean

5432cbc

paritytech-review-bot bot requested a review from a team January 27, 2025 17:59

Update from franciscoaguirre running command 'fmt'

4e0fb26

franciscoaguirre reviewed Jan 27, 2025

View reviewed changes

claravanstaden and others added 2 commits January 28, 2025 10:54

Update bridges/snowbridge/pallets/inbound-queue-v2/src/test.rs

cece49a

Co-authored-by: Francisco Aguirre <[email protected]>

Update bridges/snowbridge/primitives/router/src/inbound/v2.rs

5b28126

Co-authored-by: Francisco Aguirre <[email protected]>

paritytech-review-bot bot requested a review from a team January 28, 2025 08:55

claravanstaden mentioned this pull request Jan 28, 2025

Add relayer rewards to Snowbridge Inbound Queue v2 #7356

Open

claravanstaden added 2 commits January 28, 2025 10:58

pr comments

bec9f2c

replace operating mode

687e45e

acatangiu reviewed Jan 30, 2025

View reviewed changes

vgeddes mentioned this pull request Jan 30, 2025

Snowbridge V2 #7402

Draft

acatangiu reviewed Jan 30, 2025

View reviewed changes

acatangiu reviewed Jan 31, 2025

View reviewed changes

claravanstaden and others added 2 commits February 3, 2025 14:24

Update bridges/snowbridge/pallets/inbound-queue-v2/runtime-api/src/li…

70a9e42

…b.rs Co-authored-by: Adrian Catangiu <[email protected]>

Update bridges/snowbridge/pallets/inbound-queue-v2/src/api.rs

f7cdc2c

Co-authored-by: Adrian Catangiu <[email protected]>

paritytech-review-bot bot requested a review from a team February 3, 2025 12:25

claravanstaden and others added 2 commits February 3, 2025 14:29

Update bridges/snowbridge/pallets/inbound-queue-v2/src/envelope.rs

fde022b

Co-authored-by: Adrian Catangiu <[email protected]>

Update bridges/snowbridge/pallets/inbound-queue-v2/src/lib.rs

08f76d3

Co-authored-by: Adrian Catangiu <[email protected]>

claravanstaden mentioned this pull request Feb 5, 2025

Inbound Queue Improvements vgeddes/polkadot-sdk#1

Merged

2 tasks

		@@ -0,0 +1,3 @@
		# Ethereum Inbound Queue

		Reads messages from Ethereum and sends it to intended destination on Polkadot, using XCM.

	Reads messages from Ethereum and sends it to intended destination on Polkadot, using XCM.
	Reads messages from Ethereum and sends them to intended destination on Polkadot, using XCM.

	/// The nonce of the message been processed or not
	/// StorageMap used for encoding a [SparseBitmapImpl](<link-to-SparseBitmapImpl-file>) that tracks whether a specific nonce has been processed or not. Message nonces are unique and never repeated.

		/// Burns the fees embedded in the XCM for teleports.
		pub fn burn_fees<AssetTransactor, Balance>(dest: Location, fee: Balance) -> DispatchResult

		penpal-emulated-chain = { workspace = true }
		rococo-westend-system-emulated-network = { workspace = true }

	use penpal_emulated_chain::PARA_ID_B;
	use rococo_westend_system_emulated_network::penpal_emulated_chain::PARA_ID_B;

		let xcm_string = hex::encode(message.xcm.clone());
		log::info!(target: LOG_TARGET,"found xcm payload: {:x?}", xcm_string);

	log::info!(target: LOG_TARGET,"xcm decoded as {:?}", message_xcm);
	log::trace!(target: LOG_TARGET,"xcm decoded as {:?}", message_xcm);

	let fee_asset = Location::new(2, [GlobalConsensus(EthereumNetwork::get())]);
	let fee_asset_id = Location::new(2, [GlobalConsensus(EthereumNetwork::get())]);

	DescendOrigin(PalletInstance(InboundQueuePalletInstance::get()).into()),
	DescendOrigin(InboundQueueLocation::get()),

		// If the claimer can be decoded, add it to the message. If the claimer decoding fails,
		// do not add it to the message, because it will cause the xcm to fail.

		if let Ok(claimer) = Junction::decode(&mut claimer.as_ref()) {
		let claimer_location: Location = Location::new(0, [claimer.into()]);

		let instructions = vec![
		DepositAsset { assets: Wild(AllCounted(1).into()), beneficiary },

	let instructions = vec![DepositAsset { assets: Wild(AllCounted(2)), beneficiary }];
	let instructions = vec![RefundSurplus, DepositAsset { assets: Wild(AllCounted(2)), beneficiary }];

Snowbridge v2 - Inbound Queue #6697

Are you sure you want to change the base?

Snowbridge v2 - Inbound Queue #6697

Conversation

claravanstaden commented Nov 28, 2024 • edited Loading

Description

Integration

Review Notes

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

franciscoaguirre commented Jan 27, 2025

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

acatangiu Feb 3, 2025 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

claravanstaden commented Nov 28, 2024 •

edited

Loading

acatangiu Feb 3, 2025 •

edited

Loading