The Integration Service writes the workflow and session logs as binary files on the node where the service process runs. It adds a .bin extension to the log file name you configure in the session and workflow properties.

When you run a session on a grid, the Integration Service creates one session log for each DTM process. The log file on the primary node has the configured log file name. The log file on a worker node has a .w<Partition Group Id> extension:

For example, if you run the session s_m_PhoneList on a grid with three nodes, the session log files use the names, s_m_PhoneList.bin, s_m_PhoneList.w1.bin, and s_m_PhoneList.w2.bin.

When you rerun a session or workflow, the Integration Service overwrites the binary log file unless you choose to save workflow logs by time stamp. When you save workflow logs by time stamp, the Integration Service adds a time stamp to the log file name and archives them.

To view log files for more than one run, configure the workflow or session to create log files.

A workflow or session continues to run even if there are errors while writing to the log file after the workflow or session initializes. If the log file is incomplete, the Log Events window cannot display all the log events.

The Integration Service starts a new log file for each workflow and session run. When you recover a workflow or session, the Integration Service appends a recovery.time stamp extension to the file name for the recovery run.

For real-time sessions, the Integration Service overwrites the log file when you restart a session in cold start mode or when you restart a JMS or WebSphere MQ session that does not have recovery data. The Integration Service appends the log file when you restart a JMS or WebSphere MQ session that has recovery data.

To convert the binary file to a text file, use the

infacmd

convertLog or the

infacmd

GetLog command.