TMF Operations and Recovery Guide (G06.24+)

Recovery Methods
HP NonStop TMF Operations and Recovery Guide—522417-002
7-25
Responding to Accidental Loss of an Active-Audit
Volume
The following RECOVER FILES command illustrates the recovery of all audited logical
files located on the logical volumes $L1, $L2, $L3, and $L4, which reside on the
physical volume $P7:
TMF 40> RECOVER FILES ($L1.*.*, $L2.*.*, $L3.*.*, $L4.*.*), &
>>>WHEREPHYSVOLIS $P7, FROMARCHIVE
.
.
.
For more information about the SMF product, see Storage Management Foundation
User’s Guide. For more information about the RECOVER FILES command, see the
TMF Reference Manual.
Responding to Accidental Loss of an Active-Audit Volume
Loss of an active-audit volume causes TMF to abruptly halt. If the primary volume of a
mirrored disk becomes unusable, the system uses the mirror volume. If the active-
audit volume is not mirrored, or if both mirrors of the volume are unusable, contact the
Global Customer Support Center (GCSC) or your service provider for assistance.
Responding to Abnormal Transaction Activity
Routine monitoring of your TMF system prepares you to identify abnormal activity. If a
transaction processing application is added or changed on your system, watch
carefully to make sure that the application is working properly.
After installing a new or changed application, you should:
•
Use the TMFCOM STATUS TMF or STATUS AUDITTRAIL command to monitor
the rate at which the active-audit trails are filling for any unexpected increase in the
rate of audit generation.
•
Use the TMFCOM STATUS TMF command to monitor any unexpected increase in
the TMF transaction rate.
•
Use the TMFCOM STATUS TRANSACTION, PROCESS process command to
watch for any long-running transactions that cannot be attributed to batch or query
activity.
•
Use the TMFCOM STATUS TRANSACTION, STATE ABORTING command to
monitor the rate of aborting transactions.
If you observe unexpected transaction-related activity on the system, consider one of
the following actions to verify the correctness of the database updates:
•
Manually check randomly-selected database records (rows) that are known to
have been updated by a new or changed application for correctness of data fields
and logical consistency. You can retrieve data with an application browser (if one
exists), with FUP, or with an SQL conversational interface query.