Receiving a file using IBM Sterling Connect:Direct

Use the CDInput node to receive files from an IBM® Sterling Connect:Direct® network.

Before you begin

  1. Read IBM Sterling Connect:Direct overview and concepts.
  2. Ensure that WebSphere® MQ is installed on the same machine as your integration node. Information about file transfers is held on storage queues that are controlled by WebSphere MQ, so you must install WebSphere MQ on the same computer as your integration node if you want to use the capabilities provided by the CDInput node.
  3. Ensure that a queue manager has been specified on the integration node; this queue manager will own the system queues that store information about the file transfer. For information about specifying a queue manager on the integration node, see mqsicreatebroker command.
  4. Create the system queues that will store the file transfer information. For more information about creating the system queues, see Creating the default IBM Integration Bus queues on a WebSphere MQ queue manager.

About this task

Multiple CDInput nodes can be deployed to the same integration server, or to different integration servers in the same integration node. Multiple CDInput nodes can read files transferred to the same directory without contention. Each file is processed only once, even if the nodes are deployed to separate integration servers.

On z/OS®, when the CDInput node receives notification of the arrival of a dataset that it should process, the node copies that dataset into Unix System Services temporarily, before processing.

Complete the following steps to create a CDInput node message flow using the default settings. The task includes some optional steps for additional configuration, but detailed information about configuring the CDInput node is given on the property panels for the node, in the IBM Integration Toolkit. The default settings assume that you have a Connect:Direct server running on the same machine as the integration node, and that you are using the default ports.

Procedure

  1. Drag a CDInput node onto a message flow and wire its Out terminal to an output node of your choice.
  2. Optional: To process only a defined subset of files, configure the Basic panel.
    This action allows multiple CDInput nodes in the same integration server to receive specific files, depending on the directory or file filters specified.

    You can also specify whether, after processing, the file is left in its directory, renamed, or deleted.

  3. To change how the node handles a message flow failure, configure the Retry panel.
  4. Optional: To change how records are identified in the input file, configure the Record detection property on the Records and Elements panel.
    For example, you might want to specify that a record is fixed length, and set the record length.
    1. If you set the Record detection property to anything other than Whole File, drag an additional output node to the flow, such as the MQOutput node. Wire the End of Data terminal on the CDInput node to the In terminal of the MQOutput node. The node connected to the End of Data terminal receives an empty message when the last record in the file is read.
  5. Optional: To process the file based on details of the transfer, place a node such as the Route node after the CDInput node.
    Details of the transfer are stored in the local environment, at LocalEnvironment.CD.
  6. Setup the username and password that is required for the CDInput node to connect to the primary Connect:Direct server using the mqsisetdbparms command.
  7. Add the flow to a BAR file and deploy the BAR file.

    For information on how to use configurable services with IBM Sterling Connect:Direct, see Advanced configuration properties when using IBM Sterling Connect:Direct nodes.

  8. Use IBM Sterling Connect:Direct to send a file to the local Connect:Direct server.
    The CDInput node processes this file.