CICS-z/OS® Communications Server actions after an emergency restart
When LU-LU sessions are re-established after an emergency restart (and subsequent processing), CICS® participates in a resynchronization protocol with logical units to discover if any messages, in either direction, were lost when CICS was terminated.
The logical units for which resynchronization is required will
have been marked in the TCTTEs. Resynchronization is not attempted
in the following cases:
- If the terminal was acquired by a main terminal operation specifying COLDACQ.
- If the terminal was acquired with the EXEC CICS SET TERMINAL ACQSTATUS(COLDACQ) command.
- If the session is a pipeline session.
- If the TCTTE is marked to perform a cold start of the session by the TCT assembly process. This is done for terminals such as 3270 terminals that do not support the set and test sequence number (STSN) command.
Note: If the previous session abended, the use of COLDACQ overrides CICS integrity control. This could
lead to data integrity problems. Also, you should check the CSMT log
for an activity keypoint after the restart of a session following
a CICS failure. If there is
no activity keypoint, you should issue COLDACQ again after the next
emergency restart.
For each logical unit that does require resynchronization, CICS issues an STSN command that
notifies the logical unit of the sequence numbers known to CICS—that
is, those numbers that backout processing placed in the TCTTE. The
logical unit can compare these sequence numbers with those that it
has logged for itself, and can thus determine if any messages were
lost.
- If an input message was lost, the logical unit should retransmit it to CICS.
- If an output message was lost, CICS retransmits the message from the resend slot and, in so doing, deletes the resend slot.