Debug Terraform Resource

This skill helps debug issues with Terraform resources in the DataRobot provider.

Common Issues

1. Resource Not Found After Create

• Check if the API returns the correct ID
• Verify the ID is stored in state: data.ID = types.StringValue(response.ID)
• Look for client.NotFoundError handling in Read

2. State Drift

• Ensure Read populates all computed fields
• Check for fields that should use UseStateForUnknown() plan modifier
• Verify Optional fields handle empty vs null correctly

3. Update Not Working

• Check if field has RequiresReplace() modifier (forces new resource)
• Verify Update sends only changed fields to API
• Look for proper error handling in Update

4. Import Fails

• Ensure ImportState uses resource.ImportStatePassthroughID
• Verify Read can reconstruct full state from just the ID

Debug Commands

# Enable Terraform debug logging
export TF_LOG=DEBUG
export TF_LOG_PATH=./terraform.log

# Run with trace API calls
export TF_LOG_PROVIDER=TRACE

# Test specific resource
go test -v ./pkg/provider -run TestAcc{ResourceName} -timeout 30m

Useful Patterns

// Add debug tracing
traceAPICall("CreateResourceName")

// Check for not found errors
if _, ok := err.(*client.NotFoundError); ok {
    resp.State.RemoveResource(ctx)
    return
}

// Add diagnostic details
resp.Diagnostics.AddError(
    "Error creating resource",
    fmt.Sprintf("Could not create: %s", err.Error()),
)