H06.03 ReadMe

3.12 Storage Subsystem and SCF STATUS ADAPTER *, DETAIL Command
----------------------------------------------------------------
In rare instances, depending on the number of adapters that match
the specified adapter name wildcard, and on the number of devices
configured on each adapter, a Storage Subsystem helper process
(object file TZSTOSRV) can abend. For example, if you enter:
SCF STATUS ADAPTER $*.* , DETAIL
The workaround is to either:
- Issue the SCF STATUS ADAPTER command on each
adapter name
- Use a more restrictive wildcard name
3.13 Measure Subsystem
-----------------------
Do not stop and then immediately restart the Measure subsystem.
In rare instances, a %5004 halt can occur when a Measure subsystem
shutdown is in progress and certain chronological events occur when
Measure is restarted. Essentially, the Measure subsystem is started
again, but no new process with the old pin of Measctl is created.
When an Awake is called on an non
-existent Measure process, a halt
can occur.
If this halt should occur, stop the Measure subsystem. Reload the
processor.
3.14 Snapshot Server and DMON with SAVEABEND ON
------------------------------------------------
On a rare occasion, if both the Snapshot server
($SYSTEM.SYSnn.INSPSNAP) and DMON have SAVEABEND turned ON and both
abend at the same time, both processes might hang and then neither
can be stopped.
To recover, diver and reload the processor on which these hung
processes are executing.
3.15 CNSL (T8171) Stays Blank During System Load
--------------------------------------------------
Occasionally, during a system load (or coldload), the CNSL will stay
blank and not display any EMS messages. (The CLCI will be fine.) You
can always use Web ViewPoint to view EMS messages.
To enable the CNSL to view messages, use OSM and reset the
Maintenance Entity (ME). After performing the ME reset, use the other
fabric connection to the ME. In other words, if you were connected
Page
7
of
12