Devices system values: Action to take when a device error occurs on the workstation

The Action to take when a device error occurs on the workstation system value is also known as QDEVRCYACN. You can use this system value to set the action to take when a device error occurs.

Quick reference
Location	From IBM® Navigator for i, select Configuration and Service > System Values. Right-click on Devices and click Properties, then select the Recovery tab.
Special authority	All object (ALLOBJ) and security administrator (SECADM).
Default value	Disconnect job, and send message to user's application after reconnecting.
Changes take effect	Immediately.
Lockable	Yes. (See Lock function of security-related system values for details.)

What can I do with this system value?

You can specify the action to take when an input/output (I/O) error occurs for an interactive job's workstation.

The device recovery action is not performed until the next I/O operation is performed by the job. In a LAN or WAN environment, this allows one device to disconnect and another to connect, using the same device description, before the next I/O operation for the job occurs. The job can recover from the I/O error message and continue running to the second device. To avoid this, specify a device recovery action of Disconnect job, and return to previous request level after reconnecting or End the job and send message to QHST log. These device recovery actions are performed immediately when an I/O error, such as a power off operation, occurs.

You can select the following options:

Send error message to user's application (*MSG): Sends the I/O error message to the user's application program. The application program performs error recovery.
Disconnect job, and send message to user's application after reconnecting (*DSCMSG): Disconnects the job. When the user signs on again, an error message is sent to the application program.
Disconnect job, and return to previous request level after reconnecting (*DSCENDRQS): Disconnects the job. When the user signs on again, a cancel request function is performed to return control of the job back to the last request level.
End the job and send message to QHST log (*ENDJOBNOLIST): Ends the job. A message is sent to the QHST log indicating that the job ended because of the device error. To minimize the performance impact of the ending job, the job's priority is lowered by 10, the time slice is set to 100 milliseconds and the purge attribute is set to yes.
Select Produce a job log (*ENDJOB), if you want the message sent to the job log and the QHST log.