MigrationProcess.md

Migration Lifecycle

Migration files are executed in ascending order, sorted by the ordinal file number (the number in the file name). Parameters such as filter.from, filter.to, filter.only, and filter.skip influence the selection of migration files.

Transactions

We have introduced the capability to assign a specific name to each transaction, enhancing its entropy. This feature varies depending on the framework used.

Ethers.js Usage:

In Ethers.js, you can specify the transaction name using the customData field within the overrides. A special field, txName, is dedicated for this purpose.

Here's an example of how to set a transaction name using Ethers.js:

await govToken.transferOwnership(TOKEN_OWNER, { customData: { txName: "Transfer Ownership" }});

This method helps avoid potential collisions and ensures a smoother recovery process.

Purpose

The primary purpose of naming transactions is to facilitate the deployment process. If an error occurs, you can use the --continue flag to resume the deployment from the point of failure. The Migrator will utilize these names to distinguish between identical transactions

Namespaces

Instead of having all deployment scripts in the deploy folder, you can separate those into subfolders:

deploy
├── l1-deployment
│   ├── 1_core.migration.ts
│   └── 2_setup.migration.ts
└── l2-testnet
    ├── 1_prepare.migration.ts
    └── 2_deploy.migration.ts

And when running the migration, you can specify the namespace like this:

npx hardhat migrate --namespace l1-deployment

Verifier

For a list of parameters that affect the verification process, see Parameter Explanation.

If verification fails, the verification.verifyAttempts parameter indicates how many additional requests will be made before the migration process is terminated.