TS/MP 2.5 Release Supplement

dynamics, and changes to management algorithms and some data points. The differences in
statistical results must be minimal.
Error Handling
The ACS subsystem provides the following error handling components.
ACS Auto-Restart
The ACS subsystem detects any failure of its core processes. All the core processes on the processor
are automatically restarted when any core process fails on that processor. No manual intervention
is required from the operators. Also, the Pathsend requestors can continue to run without being
restarted. To use the ACS auto-restart feature, the Pathsend requestors must follow the guidelines
described in the Programming for Failure Recovery section of the HP NonStop TS/MP 2.5 Pathsend
and Server Programming Manual.
Pathsend Errors
Some TS/MP 2.0 errors are no longer generated but that does not cause compatibility problems.
New errors generated by the ACS subsystem can easily be handled by applications and are highly
unlikely to occur. The new errors always indicate that the transaction has not been processed and
that the transaction can be aborted, which is the standard action for unhandled errors.
For information on the Pathsend API, link manager, and PDMCOM errors, see the TS/MP 2.5
System Management Manual.
Event Management System (EMS)
The TS/MP PATHMON does not generate any new EMS messages. However, the ACS subsystem
does generate EMS messages, which might require changes in message tokenization to conform
to current standards.
All operator messages, for TS/MP and for ACS, are documented in the Operator Messages Manual.
For a definition of the EMS tokens in messages returned by the ACS subsystem, see Section 14 of
the TS/MP 2.5 Management Programming Manual.
Subsystem Control Facility (SCF)
For SCF errors specific to the ACS subsystem, see “SCF Error Messages” (page 85).
Error Handling 27