Persistent Connections and Keeping State

If you are carrying out batched queries, your database client can request a persistent connection. The server will not close this connection after sending a reply to your client. This avoids having to set up a new TCP connection for every query.

The client can request this by sending the -k or --persistent-connection query flag to the server. This flag may be sent without a query argument to start the connection. Or it may be included as a query flag with the first query.

During a persistent connection, the server operates a "stop-and-wait" protocol. This means that the next query cannot be sent until the reply has been received to the previous query. If you want to be able to send queries in batch mode, you must use a third-party tool, for example netcat or telnet, or the whois client with the file name ripe-whois-client-3.2.2.tar.gz that can be downloaded and compiled. The default whois client that is available with most operating systems may not allow this.

To exit a persistent connection, send the -k or --persistent-connection flag with no query argument or an empty query (\n) to the server. The connection will also time out after a period of inactivity.