HP Integrity NonStop H-Series - Manual (Page 72)

Usage Guidelines
HP NonStop AutoTMF Software User’s Guide—429952-013
5-8
NonStop AutoTMF and Process Pairs
NonStop AutoTMF and Process Pairs
If AutoTMF is used, no error 75 occurs on a takeover because AutoTMF creates new
automatic transactions in the backup process to access audited files.
However, when a primary process fails, automatic transactions may or may not have
been aborted, depending on the cause of the failure.
If the failure is due to a programming error, such as a trap or a call to STOP or ABEND,
AutoTMF commits all outstanding transactions before the program stops.
If the failure is due to an environmental error, such as the loss of a CPU, then the
active transactions are unilaterally aborted.
After the primary process failure, the backup process takes over at some point in the
processing, without knowing if all the previous updates have been committed.
In theory this situation could be remedied by making AutoTMF commit all active
transactions whenever the primary process does a CHECKPOINT to the backup
process. However, to commit transactions at every checkpoint would require that no
record locks or file locks be held by the program when it calls CHECKPOINT. If the
program still held locks, AutoTMF would have to abend the process after committing
the transaction.
How to Use AutoTMF with Process Pairs
In T0581AAJ, AutoTMF was enhanced to intercept the CHECKMONITOR procedure
and insure that the backup process of a process pair always initializes the AutoTMF
internal segment upon startup to insure the AutoTMF environment in the backup
process is ready in the event of a takeover.
However, be aware that
•
No reliable way exists to checkpoint transactions.
•
In some cases, a CPU failure may cause the database updates to be rolled back.
When the backup process takes over, AutoTMF starts generating new automatic
transactions without knowledge of the transactions that occurred in the failed primary
process. The backup process behaves as if the work had been completed and does
not reapply an update that was rolled back.
Recommendation
To limit the scope of the recovery required in the backup process, you should set the
ATMFProgram attribute MAXUPDATES be set to 1.Setting MAXUPDATES to 1 forces
AutoTMF to commit transactions immediately after an update. Then if a failure occurs,
at most one update must be recovered by the backup process.