Skip to main content

Using AssetHub

The relay chain does not natively support assets beyond its native token. This functionality exists in parachains. On both Polkadot and Kusama, this parachain is called Asset Hub.

The Asset Hub provides a first-class interface for creating, managing, and using fungible and non-fungible assets. The fungible interface is similar to Ethereum's ERC-20 standard. However, the data structures and stateful operations are encoded directly into the chain's runtime, making operations fast and fee-efficient.

Beyond merely supporting assets, integrating an Asset Hub into your systems has several benefits for infrastructure providers and users:

  • Support for on-chain assets.
  • Significantly lower transaction fees (about 1/10) than the relay chain.
  • Significantly lower deposits (1/100) than the relay chain. This includes the existential deposit and deposits for proxy/multisig operations.
  • Ability to pay transaction fees in certain assets. As in, accounts would not need DOT to exist on-chain or pay fees.

The Asset Hub will use DOT as its native currency. Users can transfer DOT from the relay chain into the Asset Hub and use it natively. The relay chain will also accept DOT transfers from the Asset Hub back to the relay chain for staking, governance, or any other activity.

Using the Asset Hub for DOT/KSM balance transfers will be much more efficient than the relay chain and is highly recommended. Until domain-specific parachains are built, the relay chain will still need to be used for staking and governance.

Assets Basics​

See the Assets pallet for the most up-to-date info and reference documentation.

Assets are stored as a map from an ID to information about the asset, including a management team, total supply, total number of accounts, its sufficiency for account existence, and more. Additionally, the asset owner can register metadata like the name, symbol, and number of decimals for representation.

Some assets, as determined by on-chain governance, are regarded as “sufficient”. Sufficiency means that the asset balance is enough to create the account on-chain, with no need for the DOT/KSM existential deposit. Likewise, you cannot send a non-sufficient asset to an account that does not exist. Sufficient assets can be used to pay transaction fees (i.e. there is no need to hold DOT/KSM on the account).

Assets do have a minimum balance (set by the creator), and if an account drops below that balance, the dust is lost.

Asset Operations​

The Assets pallet has its interface for dealing with assets. See the Integration section below for how to fetch information and construct transactions.

The main functions you will probably interact with are transfer and transfer_keep_alive. These functions transfer some amount (balance) of an AssetId (a u32, not a contract address) to another account.

The Assets pallet also provides an approve_transfer, cancel_approval, and transfer_approved interface for non-custodial operations.

Asset transfers will result in an assets.transferred event. The same instructions for monitoring events and not transactions applies to asset transfers.

Note that you can use the same addresses (except pure proxies!) on the Asset Hub that you use on the relay chain. The SS58 encodings are the same; only the chain information (genesis hash, etc.) will change on transaction construction.

Paying Transaction Fees in Another Asset​

Users in the Asset Hub can pay the fees of their transactions with assets other than DOT. The only requirement is that a liquidity pool of the relevant asset against DOT should already exist as a storage entry of the Asset Conversion pallet.

Technically speaking, this is enabled by the ChargeAssetTxPayment signed-extension implemented in the Asset Hub runtime. This signed-extension extends transactions to include an optional AssetId that specifies the asset to be used for payment of both the execution fees and the optional tip. It defaults to the native token when it is set to None. In case it is given, this AssetId has to be an XCM Multilocation. Once the transaction is executed in the block, it will emit an AssetTxFeePaid event, informing of the account paying the fees, the amount in the asset paid as fee, the tip (if any), and the asset ID of the asset paying the fees.

Handling Pools with Low Liquidity

Wallets and UIs enabling this functionality should ensure that the user is prompted with the necessary warnings, such that they do not accidentally spend all of their funds to perform a swap on a pool with no or low liquidity.

How to Build Transactions Paying Fees with Other Assets​
  • This repository contains the complete workflow on how to create a liquidity pool for a given asset, add liquidity to it and then build a transaction to pays fees with this asset (including fees estimation). It is done with several libraries: Polkadot.js API and Subxt.
  • Example using Asset Transfer API to do a cross-chain transfer in Polkadot Asset Hub paying fees with GLMR.
  • A simple script using Polkadot.js API to do a local transfer of bridged KSM in Polkadot Asset Hub paying fees with USDT.

Foreign Assets​

Foreign assets are those assets in Asset Hub whose native blockchain is not Asset Hub. These are mainly native tokens from other parachains or bridged tokens from other consensus systems (such as Ethereum). Once a foreign asset has been registered in Asset Hub (by its root origin), users are enabled to send this token from its native blockchain to Asset Hub and operate with it as if it were any other asset.

Practically speaking, foreign assets are handled by the foreign-assets pallet in Asset Hub, which is an instance of the Assets pallet. Hence, this pallet exposes the same interface to users and other pallets as the Assets pallet.

The main difference to take into account for foreign assets is their identifier. Instead of using integers as identifiers like in the Assets pallet, assets stored in the foreign-assets pallet are identified by their XCM multilocation.

Integration​

The Asset Hub will come with the same tooling suite that Parity Technologies provides for the Relay Chain, namely API Sidecar and TxWrapper Polkadot, as well as the Asset Transfer API. If you have a technical question or issue about how to use one of the integration tools, please file a GitHub issue so a developer can help.

Parachain Node​

Using the Asset Hub will require running a parachain node to sync the chain. This is very similar to running a relay chain node, with the addition of some extra flags. You can follow these guidelines to set up an Asset Hub node.

Asset Transfer API​

Asset-transfer-api is a library focused on simplifying the construction of asset transfers for Substrate-based chains that involve system parachains like Asset Hub (Polkadot and Kusama). It exposes a reduced set of methods that facilitate users to send transfers to other (para) chains or locally. You can refer to this table for the current cross-chain support and here for the complete documentation, including installation guide and usage examples.

Sidecar​

API Sidecar is a REST service for relay chain and parachain nodes. It comes with endpoints to query information about assets and asset balances on the Asset Hub.

  • Asset lookups always use the AssetId to refer to an asset class. On-chain metadata is subject to change and thus unsuitable as a canonical index.
  • Please refer to docs for full usage information. Details on options like how to make a historical query are not included here.

Here are the available public instances:

The purpose of these instances is to allow anyone to check and get a quick overview of the info that the asset-related endpoints provide.

caution

These instances should only be used for ad-hoc checks or tests and not for production, heavy testing or any other critical purpose.

Tx Wrapper Polkadot​

TxWrapper Polkadot is a library designed to facilitate transaction construction and signing in offline environments. It comes with asset-specific functions to use on the Asset Hub. When constructing parachain transactions, you can use txwrapper-polkadot exactly as on the relay chain, but construct transactions with the appropriate parachain metadata like genesis hash, spec version, and type registry.

XCM Transfer Monitoring​

Monitoring of XCM deposits​

Thanks to XCM and a growing number of parachains, the relay chain native token can exist across several blockchains, which means the providers need to monitor cross-chain transfers on top of local transfers and corresponding balances.transfer events.

Currently, DOT can be sent and received in the relay chain and in the Asset Hub either with a Teleport from system parachains or with a Reserve Backed Transfer from any other parachain. In both cases, the event emitted when processing the transfer is the balances.minted event. Hence, providers should listen to these events, pointing to an address in their system. For this, the service provider must query every new block created, loop through the events array, filter for any balances.minted event, and apply the appropriate business logic.

Tracking back XCM information​

What has been mentioned earlier should be sufficient to confirm that DOT has arrived in a given account via XCM. However, in some cases, it may be interesting to identify the cross-chain message that emitted the relevant balances.minted event. This can be done as follows:

  1. Query the relevant chain at the block the balances.minted event was emitted.
  2. Filter for messageQueue(Processed) events. These can be emitted during any phase of the block, not just initialization. This event has a parameter Id. The value of Id identifies the cross-chain message received in the relay chain or in the Asset Hub. It can be used to track back the message in the origin parachain if needed. Note that a block may contain several messageQueue(Processed) events corresponding to several cross-chain messages processed for this block.

Additional Examples of Monitoring XCM Transfers​

The two previous sections outline the process of monitoring XCM deposits to specific account(s) and then tracing back the origin of these deposits. However, the process of tracking an XCM transfer (hence the events to look for) may vary based on the direction of the XCM message. Here are some examples to showcase the slight differences:

  1. For an XCM transfer from a Parachain to a relay chain (example):

    • The event to look for in the Parachain side is called parachainsystem (UpwardMessageSent), and the parameter message_hash in this event identifies the XCM transfer.
    • The event to track in the relay chain side is called messagequeue (Processed), and the parameter id of the event should be the same as the message_hash found in the Parachain event.

  2. For an XCM transfer from a relay chain to a parachain (example):

    • The event to look for in the relay chain side is called xcmPallet (sent), and the parameter message_id in this event identifies the XCM transfer.
    • The event to look for in the Parachain side is called dmpqueue (ExecutedDownward), and the parameter that identifies the XCM message is either called message_hash or message_id.

  3. For an XCM transfer from a System Parachain to a Parachain (example):

    • The event to look for in the System Parachain side is called xcmpqueue (XcmpMessageSent), and again the message_hash is one of the parameters of the event.
    • The corresponding event in the Parachain side is the xcmpqueue (Success) and the message_hash found in that event should have the same value as the one in the System parachain.

Monitoring of Failed XCM Transfers​

In case that an XCM transfer fails to complete successfully, then we will notice some different parameters in the events emitted or different events. Below are some examples:

  1. From a relay chain to a System Parachain (example):

    • We will see the event dmpqueue (ExecutedDownward) in the System Parachain side with the following parameters:
      • outcome with value Incomplete and with the type of error which in this example is UntrustedReserveLocation.
      • message_id which shows the hash of the XCM Transfer.

  2. From a Parachain to another Parachain (example):

    • We will see the event xcmpqueue (Fail) in the destination Parachain with the following parameters:
      • error which in this example is TooExpensive.
      • message_hash which identifies the XCM Transfer.
    • Note: there might be another event called polkadotxcm (AssetsTrapped) which indicates that some assets have been trapped (and hence can be claimed).

A great resource to learn more about Error Management in XCM is the Polkadot blog post from Gavin Wood, XCM Part III: Execution and Error Management.

Last updated on by Filippo