Error Formatting

It is impossible to list all the possible ways an error can occur. Some may occur in the transport layer (for example, loss of network connectivity), in which case the results vary depending on what client and transport you are using. However, if the rippled server successfully receives your request, it tries to respond in a standardized error format.

Caution: When your request results in an error, the entire request is copied back as part of the response, so that you can try to debug the error. However, this also includes any secrets that were passed as part of the request. When sharing error messages, be very careful not to accidentally expose important account secrets to others.

Some example errors:

{
  "id": 3,
  "status": "error",
  "type": "response",
  "error": "ledgerIndexMalformed",
  "request": {
    "account": "r9cZA1mLK5R5Am25ArfXFmqgNwjZgnfk59",
    "command": "account_info",
    "id": 3,
    "ledger_index": "-",
    "strict": true
  }
}
HTTP Status: 200 OK
{
    "result": {
        "error": "ledgerIndexMalformed",
        "request": {
            "account": "r9cZA1mLK5R5Am25ArfXFmqgNwjZgnfk59",
            "command": "account_info",
            "ledger_index": "-",
            "strict": true
        },
        "status": "error"
    }
}
{
    "result": {
        "error": "ledgerIndexMalformed",
        "request": {
            "account": "r9cZA1mLK5R5Am25ArfXFmqgNwjZgnfk59",
            "command": "account_info",
            "ledger_index": "-",
            "strict": true
        },
        "status": "error"
    }
}

WebSocket Format

Field Type Description
id (Varies) ID provided in the Web Socket request that prompted this response
status String "error" if the request caused an error
type String Typically "response", which indicates a successful response to a command.
error String A unique code for the type of error that occurred
request Object A copy of the request that prompted this error, in JSON format. Caution: If the request contained any account secrets, they are copied here!

JSON-RPC Format

Some JSON-RPC request respond with an error code on the HTTP layer. In these cases, the response is a plain-text explanation in the response body. For example, if you forgot to specify the command in the method parameter, the response is like this:

HTTP Status: 400 Bad Request
Null method

For other errors that returned with HTTP status code 200 OK, the responses are formatted in JSON, with the following fields:

Field Type Description
result Object Object containing the response to the query
result.error String A unique code for the type of error that occurred
result.status String "error" if the request caused an error
result.request Object A copy of the request that prompted this error, in JSON format. Caution: If the request contained any account secrets, they are copied here! Note: The request is re-formatted in WebSocket format, regardless of the request made.

Universal Errors

All methods can potentially return any of the following values for the error code:

  • unknownCmd - The request does not contain a command that the rippled server recognizes.
  • jsonInvalid - (WebSocket only) The request is not a proper JSON object.
    • JSON-RPC returns a 400 Bad Request HTTP error in this case instead.
  • missingCommand - (WebSocket only) The request did not specify a command field.
    • JSON-RPC returns a 400 Bad Request HTTP error in this case instead.
  • tooBusy - The server is under too much load to do this command right now. Generally not returned if you are connected as an admin.
  • noNetwork - The server is having trouble connecting to the rest of the XRP Ledger peer-to-peer network (and is not running in stand-alone mode).
  • noCurrent - The server does not know what the current ledger is, due to high load, network problems, validator failures, incorrect configuration, or some other problem.
  • noClosed - The server does not have a closed ledger, typically because it has not finished starting up.
  • wsTextRequired - (WebSocket only) The request's opcode is not text.