Troubleshooting Guide
General Troubleshooting
Logging
Logging
Logging levels and formats can be configured for Hyperlane tooling using the following two environment variables:
LOG_LEVEL: The log level to filter to. Defaults toinfo. Choices:debug | info | warn | error | offLOG_FORMAT: The format of the log output. Defaults topretty. Choices:pretty | json
The Hyperlane CLI also allows configuring logging via the --log and --verbosity flags.
Chain Configuration
Chain Configuration
Within your working directory, you may find a chains/ yaml files organized by chain name. These metadata.yaml files describe the information needed to use the chain in Hyperlane deployments and apps.
You can define a full configuration for any new chain in this file. The metadata that can be configured is defined in this example configuration. You can also find the chain metadata schema at chainMetadataTypes.ts.
Here’s an example configuration for two local anvil chains:
You can also extend a core chain config by providing the fields to be overridden:
Override RPC URLs
Override RPC URLs
You can override the RPC urls by extending the core chain config.
Example: Define RPC URLs array and adjust retry settings
Override transaction settings
Override transaction settings
Transaction overrides are any properties to include when forming transaction requests. For example:
gasPrice: number | stringmaxFeePerGas: number | stringmaxPriorityFeePerGas: number | stringnonce: number | stringtype: numberccipReadEnabled: boolean
In the example below we’re using a gas price of 7 gwei, hardcoding the nonce, and setting a maximum value for the base and priority fees.
If you are overriding the nonce in the chain configuration, ensure you are updating the value on successful transactions.
Long Message Delivery Times or Timeouts
Long Message Delivery Times or Timeouts
Delays or timeouts in message delivery often result from RPC issues, such as overloaded endpoints, or from chains that produce blocks only on demand.
To identify the issue, check for logs indicating RPC-related problems such as: Deprioritizing an inner provider in FallbackProvider
To address these delays:
- Check RPC Health: Ensure RPC endpoints are responsive.
- Configure Chains with On-Demand Blocks: For these chains, set the reorgPeriod to 0. This ensures that agents always look at the tip of the chain, avoiding delays caused by block finalization processes.
Agent Troubleshooting
Validator has not announced any storage locations Warning
Validator has not announced any storage locations Warning
- Reason: This occurs when the relayer cannot retrieve validator signatures required to process a message.
- Solution:
- Ensure the validators have announced their storage locations. Check validator logs for a message such as:
Validator has announced signature storage location, locations: ["file:///tmp/hyperlane-validator-signatures-local"] - Verify that each validator has a unique signature storage path (
--checkpointSyncer.path) to prevent overwriting. - Confirm that the relayer has read access to the storage paths.
- Ensure the validators have announced their storage locations. Check validator logs for a message such as:
No ISM Found for Origin Error
No ISM Found for Origin Error
- Reason: This occurs when the Interchain Security Module (ISM) does not recognize the origin chain.
- Solution:
- Ensure the ISM configuration includes the Validators for the origin chain.
- If it doesn’t, add validators for the origin chain to your ISM.
- Restart the relayer after updating the ISM configuration.
Checking Pending Messages in the Queue
Checking Pending Messages in the Queue
If messages are stuck in the relayer queue or not being processed, use the list_operations endpoint to inspect the relayer’s queue:
This endpoint provides the status of messages in the queue and can help identify why they are not being delivered.
Debugging with Logs
Debugging with Logs
For detailed insights, enable debug logging and monitor the relayer’s activity: HYP_LOG_LEVEL=debug. Once logs are captured, you can use them to locate issues with specific message IDs.
Retrying Messages
Retrying Messages
After reviewing the logs, you can trigger an immediate retry of all pending messages using the message_retry endpoint:
Make sure that you use the most recent version of the relayer and capture all logs for debugging.
Relayer Selecting Incorrect ISM
Relayer Selecting Incorrect ISM
Ensure the Interchain Security Module (ISM) on your chain is configured correctly. An incorrect ISM can result in messages not being processed. Verify the ISM address in your configuration and ensure it matches your deployment setup.
Advanced Troubleshooting
eth_getStorageAt() Compatibility
eth_getStorageAt() Compatibility
Some chains do not natively support the eth_getStorageAt() API. If you’re deploying on one of these chains and encounter issues, review the changes made to the Hyperlane codebase in this commit.
To deploy Hyperlane on these chains follow the steps: