Rosetta CLI

CLI to validate the correctness of Rosetta API implementations

Overview

The rosetta-cli is used by developers to test the correctness of their Rosetta API implementations. The CLI also provides the ability to look up block contents and account balances.

Documentation

Before diving into the CLI, we recommend taking a look at the Rosetta API Docs:

• Overview
• Data API
• Construction API

Install

To download a binary for the latest release, run:

curl -sSfL https://raw.githubusercontent.com/coinbase/rosetta-cli/master/scripts/install.sh | sh -s

The binary will be installed inside the ./bin directory (relative to where the install command was run).

Downloading binaries from the Github UI will cause permission errors on Mac.

Installing in Custom Location

To download the binary into a specific directory, run:

curl -sSfL https://raw.githubusercontent.com/coinbase/rosetta-cli/master/scripts/install.sh | sh -s -- -b <relative directory>

Usage

CLI for the Rosetta API

Usage:
  rosetta-cli [command]

Available Commands:
  check:construction           Check the correctness of a Rosetta Construction API Implementation
  check:data                   Check the correctness of a Rosetta Data API Implementation
  configuration:create         Create a default configuration file at the provided path
  configuration:validate       Ensure a configuration file at the provided path is formatted correctly
  help                         Help about any command
  utils:asserter-configuration Generate a static configuration file for the Asserter
  utils:train-zstd             Generate a zstd dictionary for enhanced compression performance
  version                      Print rosetta-cli version
  view:balance                 View an account balance
  view:block                   View a block
  view:networks                View all network statuses

Flags:
      --block-profile string        Save the pprof block profile in the specified file
      --configuration-file string   Configuration file that provides connection and test settings.
                                    If you would like to generate a starter configuration file (populated
                                    with the defaults), run rosetta-cli configuration:create.

                                    Any fields not populated in the configuration file will be populated with
                                    default values.
      --cpu-profile string          Save the pprof cpu profile in the specified file
  -h, --help                        help for rosetta-cli
      --mem-profile string          Save the pprof mem profile in the specified file

Use "rosetta-cli [command] --help" for more information about a command.

Configuration

All rosetta-cli parameters are populated from a configuration file (--configuration-file) provided at runtime. If a configuration file is not provided, the default configuration is used. This default configuration can be viewed here. Note, there is no default configuration for running check:construction as this is very network-specific. You can view a full list of all configuration options here.

In the examples/configuration directory, you can find examples configuration files for running tests against a Bitcoin Rosetta implementation (config) and an Ethereum Rosetta implementation (config).

Using environment variables

It's possible to set the configuration file path using an environment variable instead of using a CLI flag. See an example below:

ROSETTA_CONFIGURATION_FILE=/path/to/cli/config rosetta <command>

CLI flags take precedence over environment variables.

Writing check:construction Tests

The new Construction API testing framework (first released in rosetta-cli@v0.5.0) uses a new design pattern to allow for complex transaction construction orchestration. You can read more about the design goals here.

Most teams write their Construction API tests using the Rosetta Constructor DSL. We have examples of a DSL files written for UTXO-based chains and account-based chains.

Terminology

When first learning about a new topic, it is often useful to understand the hierarchy of concerns. In the automated Construction API tester, this "hierarchy" is as follows:

Workflows -> Jobs
  Scenarios
    Actions

Workflows contain collections of Scenarios to execute. Scenarios are executed atomically in database transactions (rolled back if execution fails) and culminate in an optional broadcast. This means that a single Workflow could contain multiple broadcasts (which can be useful for orchestrating staking-related transactions that affect a single account).

To perform a Workflow, we create a Job. This Job has a unique identifier and stores state for all Scenarios in the Workflow. State is shared across an entire Job so Actions in a Scenario can access the output of Actions in other Scenarios. The syntax for accessing this shared state can be found here.

Actions are discrete operations that can be performed in the context of a Scenario. A full list of all Actions that can be performed can be found here.

If you have suggestions for more actions, please open an issue in rosetta-sdk-go!

Workflows

To use the automated Construction API tester (without prefunded accounts), you must implement 2 required Workflows:

• create_account
• request_funds

If you don't implement these 2 Workflows, processing could stall.

Please note that create_account can contain a transaction broadcast if on-chain origination is required for new accounts on your blockchain.

If you plan to run the automated Construction API tester in CI, you may wish to provide prefunded accounts when running the tester (otherwise you would need to manually fund generated accounts).

Optionally, you can also provide a return_funds workflow that will be invoked when exiting check:construction. This can be useful in CI when you want to return all funds to a single accout or faucet (instead of black-holing them in all the addresses created during testing).

Broadcast Invocation

If you'd like to broadcast a transaction at the end of a Scenario, you must populate the following fields:

• <scenario>.network
• <scenario>.operations
• <scenario>.confirmation_depth (allows for stake-related transactions to complete before marking as a success)

Optionally, you can populate the following field:

• <scenario>.preprocess_metadata

Once a transaction is confirmed on-chain (after the provided <scenario>.confirmation_depth, it is stored by the tester at <scenario>.transaction for access by other Scenarios in the same Job.

Dry Runs

In UTXO-based blockchains, it may be necessary to amend the operations stored in <scenario>.operations based on the suggested_fee returned in /construction/metadata. The automated Construction API tester supports running a "dry run" of a transaction broadcast if you set the follow field:

• <scenario>.dry_run = true

The suggested fee will then be stored as <scenario>.suggested_fee for use by other Scenarios in the same Job. You can find an example of this in the Bitcoin config.

If this field is not populated or set to false, the transaction will be constructed, signed, and broadcast.

End Conditions

When running the rosetta-cli in a CI job, it is usually desired to exit when certain conditions are met (or before then with an exit code of 1). We provide this functionality through the use of "end conditions" which can be specified in your configuration file.

check:data

A full list of check:data end conditions can be found here. If any end condition is satisifed, we will exit and output the results in results_output_file (if it is populated).

check:construction

The check:construction end condition is a map of workflow:count that indicates how many of each workflow should be performed before check:construction should stop. For example, {"create_account": 5} indicates that 5 create_account workflows should be performed before stopping.

Unlike check:data, all check:construction end conditions must be satisifed before the rosetta-cli will exit.

Disable Complex Checks

If you are just getting started with your implementation, you may want to disable balance tracking (did any address balance go below zero?) and reconciliation (does the balance I calculated match the balance returned by the /account/balance endpoint?). Take a look at the simple configuration for an example of how to do this.

Status Codes

If there are no issues found while running check, it will exit with a 0 status code. If there are any issues, it will exit with a 1 status code. It can be useful to run this command as an integration test for any changes to your implementation.

Commands

version

Print rosetta-cli version

Usage:
  rosetta-cli version [flags]

Flags:
  -h, --help   help for version

Global Flags:
      --block-profile string        Save the pprof block profile in the specified file
      --configuration-file string   Configuration file that provides connection and test settings.
                                    If you would like to generate a starter configuration file (populated
                                    with the defaults), run rosetta-cli configuration:create.

                                    Any fields not populated in the configuration file will be populated with
                                    default values.
      --cpu-profile string          Save the pprof cpu profile in the specified file
      --mem-profile string          Save the pprof mem profile in the specified file

check:data

Check all server responses are properly constructed, that
there are no duplicate blocks and transactions, that blocks can be processed
from genesis to the current block (re-orgs handled automatically), and that
computed balance changes are equal to balance changes reported by the node.

When re-running this command, it will start where it left off if you specify
some data directory. Otherwise, it will create a new temporary directory and start
again from the genesis block. If you want to discard some number of blocks
populate the start_index filed in the configuration file with some block index.
Starting from a given index can be useful to debug a small range of blocks for
issues but it is highly recommended you sync from start to finish to ensure
all correctness checks are performed.

By default, account balances are looked up at specific heights (instead of
only at the current block). If your node does not support this functionality,
you can disable historical balance lookups in your configuration file. This will
make reconciliation much less efficient but it will still work.

If check fails due to an INACTIVE reconciliation error (balance changed without
any corresponding operation), the cli will automatically try to find the block
missing an operation. If historical balance disabled is true, this automatic
debugging tool does not work.

To debug an INACTIVE account reconciliation error without historical balance lookup,
set the interesting accounts to the path of a JSON file containing
accounts that will be actively checked for balance changes at each block. This
will return an error at the block where a balance change occurred with no
corresponding operations.

If your blockchain has a genesis allocation of funds and you set
historical balance disabled to true, you must provide an
absolute path to a JSON file containing initial balances with the
bootstrap balance config. You can look at the examples folder for an example
of what one of these files looks like.

Usage:
  rosetta-cli check:data [flags]

Flags:
  -h, --help   help for check:data

Global Flags:
      --block-profile string        Save the pprof block profile in the specified file
      --configuration-file string   Configuration file that provides connection and test settings.
                                    If you would like to generate a starter configuration file (populated
                                    with the defaults), run rosetta-cli configuration:create.

                                    Any fields not populated in the configuration file will be populated with
                                    default values.
      --cpu-profile string          Save the pprof cpu profile in the specified file
      --mem-profile string          Save the pprof mem profile in the specified file

check:construction

The check:construction command runs an automated test of a
Construction API implementation by creating and broadcasting transactions
on a blockchain. In short, this tool generates new addresses, requests
funds, constructs transactions, signs transactions, broadcasts transactions,
and confirms transactions land on-chain. At each phase, a series of tests
are run to ensure that intermediate representations are correct (i.e. does
an unsigned transaction return a superset of operations provided during
construction?).

Check out the https://github.com/coinbase/rosetta-cli/tree/master/examples
directory for examples of how to configure this test for Bitcoin and
Ethereum.

Right now, this tool only supports transfer testing (for both account-based
and UTXO-based blockchains). However, we plan to add support for testing
arbitrary scenarios (i.e. staking, governance).

Usage:
  rosetta-cli check:construction [flags]

Flags:
  -h, --help   help for check:construction

Global Flags:
      --block-profile string        Save the pprof block profile in the specified file
      --configuration-file string   Configuration file that provides connection and test settings.
                                    If you would like to generate a starter configuration file (populated
                                    with the defaults), run rosetta-cli configuration:create.

                                    Any fields not populated in the configuration file will be populated with
                                    default values.
      --cpu-profile string          Save the pprof cpu profile in the specified file
      --mem-profile string          Save the pprof mem profile in the specified file

configuration:create

Create a default configuration file at the provided path

Usage:
  rosetta-cli configuration:create [flags]

Flags:
  -h, --help   help for configuration:create

Global Flags:
      --block-profile string        Save the pprof block profile in the specified file
      --configuration-file string   Configuration file that provides connection and test settings.
                                    If you would like to generate a starter configuration file (populated
                                    with the defaults), run rosetta-cli configuration:create.

                                    Any fields not populated in the configuration file will be populated with
                                    default values.
      --cpu-profile string          Save the pprof cpu profile in the specified file
      --mem-profile string          Save the pprof mem profile in the specified file

configuration:validate

Validate the correctness of a configuration file at the provided path

Usage:
  rosetta-cli configuration:validate [flags]

Flags:
  -h, --help   help for configuration:validate

Global Flags:
      --configuration-file string   Configuration file that provides connection and test settings.
                                    If you would like to generate a starter configuration file (populated
                                    with the defaults), run rosetta-cli configuration:create.

                                    Any fields not populated in the configuration file will be populated with
                                    default values.

view:networks

While debugging a Data API implementation, it can be very
useful to view network(s) status. This command fetches the network
status from all available networks and prints it to the terminal.

If this command errors, it is likely because the /network/* endpoints are
not formatted correctly.

Usage:
  rosetta-cli view:networks [flags]

Flags:
  -h, --help   help for view:networks

Global Flags:
      --block-profile string        Save the pprof block profile in the specified file
      --configuration-file string   Configuration file that provides connection and test settings.
                                    If you would like to generate a starter configuration file (populated
                                    with the defaults), run rosetta-cli configuration:create.

                                    Any fields not populated in the configuration file will be populated with
                                    default values.
      --cpu-profile string          Save the pprof cpu profile in the specified file
      --mem-profile string          Save the pprof mem profile in the specified file

view:balance

While debugging, it is often useful to inspect the state
of an account at a certain block. This command allows you to look up
any account by providing a JSON representation of a types.AccountIdentifier
(and optionally a height to perform the query).

For example, you could run view:balance '{"address":"interesting address"}' 1000
to lookup the balance of an interesting address at block 1000. Allowing the
address to specified as JSON allows for querying by SubAccountIdentifier.

Usage:
  rosetta-cli view:balance [flags]

Flags:
  -h, --help   help for view:balance

Global Flags:
      --block-profile string        Save the pprof block profile in the specified file
      --configuration-file string   Configuration file that provides connection and test settings.
                                    If you would like to generate a starter configuration file (populated
                                    with the defaults), run rosetta-cli configuration:create.

                                    Any fields not populated in the configuration file will be populated with
                                    default values.
      --cpu-profile string          Save the pprof cpu profile in the specified file
      --mem-profile string          Save the pprof mem profile in the specified file

view:block

While debugging a Data API implementation, it can be very
useful to inspect block contents. This command allows you to fetch any
block by index to inspect its contents. It uses the
fetcher (https://github.com/coinbase/rosetta-sdk-go/tree/master/fetcher) package
to automatically get all transactions in the block and assert the format
of the block is correct before printing.

If this command errors, it is likely because the block you are trying to
fetch is formatted incorrectly.

Usage:
  rosetta-cli view:block [flags]

Flags:
  -h, --help           help for view:block
      --only-changes   Only print balance changes for accounts in the block

Global Flags:
      --block-profile string        Save the pprof block profile in the specified file
      --configuration-file string   Configuration file that provides connection and test settings.
                                    If you would like to generate a starter configuration file (populated
                                    with the defaults), run rosetta-cli configuration:create.

                                    Any fields not populated in the configuration file will be populated with
                                    default values.
      --cpu-profile string          Save the pprof cpu profile in the specified file
      --mem-profile string          Save the pprof mem profile in the specified file

utils:asserter-configuration

In production deployments, it is useful to initialize the response
Asserter (https://github.com/coinbase/rosetta-sdk-go/tree/master/asserter) using
a static configuration instead of intializing a configuration dynamically
from the node. This allows a client to error on new types/statuses that may
have been added in an update instead of silently erroring.

To use this command, simply provide an absolute path as the argument for where
the configuration file should be saved (in JSON).

Usage:
  rosetta-cli utils:asserter-configuration [flags]

Flags:
  -h, --help   help for utils:asserter-configuration

Global Flags:
      --block-profile string        Save the pprof block profile in the specified file
      --configuration-file string   Configuration file that provides connection and test settings.
                                    If you would like to generate a starter configuration file (populated
                                    with the defaults), run rosetta-cli configuration:create.

                                    Any fields not populated in the configuration file will be populated with
                                    default values.
      --cpu-profile string          Save the pprof cpu profile in the specified file
      --mem-profile string          Save the pprof mem profile in the specified file

utils:train-zstd

Zstandard (https://github.com/facebook/zstd) is used by
rosetta-sdk-go/storage to compress data stored to disk. It is possible
to improve compression performance by training a dictionary on a particular
storage namespace. This command runs this training and outputs a dictionary
that can be used with rosetta-sdk-go/storage.

The arguments for this command are:
<namespace> <database path> <dictionary path> <max items> (<existing dictionary path>)

You can learn more about dictionary compression on the Zstandard
website: https://github.com/facebook/zstd#the-case-for-small-data-compression

Usage:
  rosetta-cli utils:train-zstd [flags]

Flags:
  -h, --help   help for utils:train-zstd

Global Flags:
      --block-profile string        Save the pprof block profile in the specified file
      --configuration-file string   Configuration file that provides connection and test settings.
                                    If you would like to generate a starter configuration file (populated
                                    with the defaults), run rosetta-cli configuration:create.

                                    Any fields not populated in the configuration file will be populated with
                                    default values.
      --cpu-profile string          Save the pprof cpu profile in the specified file
      --mem-profile string          Save the pprof mem profile in the specified file

Correctness Checks

This tool performs a variety of correctness checks using the Rosetta Server. If any correctness check fails, the CLI will exit and print out a detailed message explaining the error.

Response Correctness

The validator uses the autogenerated Go Client package to communicate with the Rosetta Server and assert that responses adhere to the Rosetta interface specification.

Duplicate Hashes

The validator checks that a block hash or transaction hash is never duplicated.

Non-negative Balances

The validator checks that an account balance does not go negative from any operations.

Balance Reconciliation

Active Addresses

The CLI checks that the balance of an account computed by its operations is equal to the balance of the account according to the node. If this balance is not identical, the CLI will exit.

Inactive Addresses

The CLI randomly checks the balances of accounts that aren't involved in any transactions. The balances of accounts could change on the blockchain node without being included in an operation returned by the Rosetta Data API. Recall that all balance-changing operations should be returned by the Rosetta Data API.

Development

• make deps to install dependencies
• make test to run tests
• make lint to lint the source code (included generated code)
• make release to run one last check before opening a PR
• make compile version=RELEASE_TAG to generate binaries

Helper/Handler

Many of the packages use a Helper/Handler interface pattern to acquire required information or to send events to some client implementation. An example of this is in the reconciler package where a Helper is used to get the account balance and the Handler is called to indicate whether the reconciliation of an account was successful.

Repo Structure

cmd
examples // examples of different config files
pkg
  logger // logic to write syncing information to stdout/files
  processor // Helper/Handler implementations for reconciler, storage, and syncer
  tester // test orchestrators

Troubleshoot

• While running check:data or check:construction option if you get the following error:

  dial tcp 127.0.0.1:8080: socket: too many open files: unable to sync to 1902533: unable to sync to 1902533

  Please run ulimit -n 10000 to increase the max concurrent opened file limit

  Note: MacOS users, if you face ulimit: setrlimit failed: invalid argument error while setting ulimit, please run sudo launchctl limit maxfiles 10000 200000 before setting the ulimit

License

This project is available open source under the terms of the Apache 2.0 License.

© 2020 Coinbase