ESAF (External Subsystem Attach Facility) interface problems
can be diagnosed by using the external trace data set. You can use
the IMS /TRACE command to direct
and control the tracing of internal IMS events.
The IMS external
subsystem (ESS) trace impacts performance. Activate the trace only
when you notice a problem or if you need to re-create a problem. IBM® software support may ask you
to provide dumps of IMS CTL, z/OS® SVC and the IMS online log data set.
To document problems
that involve the Db2 for z/OS ESS
interface:
Procedure
- Use the following TRACE command to turn
on the IMS ESS trace and to direct
its output to the external trace data set:
/TRACE SET ON TABLE SUBS OPTION LOG
- Obtain dumps of the IMS CTL
and involved dependent regions, before and after the failure, by issuing
this series of commands:
DUMP COMM=(dump title)
R id JOBNAME=(j1,j2,j3,j4,j5),
SDATA=(CSA,PSA,RGN,SQA,SUM,TRT),END
- Obtain a z/OS SVC
dump of the DB2® MSTR and DBM1
regions by issuing this series of commands:
DUMP COMM=(dump title)
R id JOBNAME=(dbtmstr,dbwdbm1),
SDATA=(CSA,PSA,RGN,SQA,SUM,TRT),END
- Save the IMS online
log data set that was active during the failure because IMS TYPE5501, 08, 07, 56 and other log records
can be critical to diagnosis. The IMS TYPE5501
records are updated by DB2.
The internal buffer for these records is stored at the location described
by the CDE entry named WAL in the IMS regions.
- If the IMS monitor
is started, issue the following command to monitor the IMS data set:
/TRACE SET ON MONITOR ALL