The Monitor runs as a Natural subtask under Entire System Server or as a batch job and does all the work of generating, printing and distributing reports and bundles.
This section covers the following topics:
To select Monitor Start/Close:
Enter 6
in the command line of the System
Administration menu.
The Monitor Management screen is displayed:
17:14:52 **** ENTIRE OUTPUT MANAGEMENT **** 2007-05-16 User ID XYZ - Monitor Management - Status Idle at 17:14:15 2007-05-16 S Start Monitor C Close Monitor L Display Monitor Log D Display Monitor Log by Date/Time P Purge Monitor Buffer Pool E Purge a single Buffer Pool Entry +-------------------------------------------------+ : ---------- Main Task Parameters ----------- : : Monitor Node ........ 40 : : Minimum Wait ........ 10__ (in seconds) : : Maximum Wait ........ 50__ (in seconds) : : Wait Increment ...... 5___ (in seconds) : : Current Wait ........ 60 (in seconds) : +-------------------------------------------------+ Command => ___________________________________________________________________ Enter-PF1---PF2---PF3---PF4---PF5---PF6---PF7---PF8---PF9---PF10--PF11--PF12--- Help Exit Flip Tasks Wake Menu |
Key | Name | Function |
---|---|---|
PF8 | Tasks | Display monitor subtask status. |
PF10 | Wake | Activate the Monitor before the next cycle. |
The Monitor Management screen enables the system administrator to start, wake or close the Entire Output Management Monitor manually, display the Monitor Log and purge the Monitor Buffer Pool.
These functions are explained below.
Field | Explanation |
---|---|
Status |
Monitor status. Possible values:
|
at | The time when the Monitor was last active. |
Monitor Node | The node under which Entire Output Management is running. |
Minimum Wait | The minimum time in seconds the Monitor is to wait between two consecutive monitoring cycles. You can modify this value by entering a new value. |
Maximum Wait | The maximum time in seconds the Monitor is to wait between two consecutive monitoring cycles. You can modify this value by entering a new value. |
Wait Increment | The number of seconds by which the wait time increases. If there is no activity during the minimum wait time, the wait time is increased by this value, until the maximum is reached. When activity occurs, the wait time returns to the minimum. You can modify this value by entering a new value. |
Current Wait | The wait time in effect for the current cycle. |
To start the Monitor, the Entire System Server Node specified for start must be active.
Enter an S
in the command line.
A message confirms the start, and the value of the Status field changes accordingly.
To activate the Monitor before the next scheduled activity cycle, see Wait parameters above).
Press PF10 (Wake) on the Monitor Management screen.
The Monitor is activated.
When you press ENTER again, the at field (see above) displays the time when the Monitor became active.
If there was any pending work, the Status changes. When the activity cycle is completed, Monitor status changes back to Idle.
To close the Monitor:
Enter a C
in the command line of the
Monitor Management screen.
A window is displayed, asking you for confirmation.
Enter SHUTDOWN to confirm the shutdown.
The Monitor status changes to "Shutdown In Progress". This means that the Monitor has not yet detected the close, since it is in wait status. The next time it is active, the Monitor detects the close and performs the normal close.
The value of the Status field changes to "Closed".
You can change the default wait time between two monitoring cycles, in order to reflect the load at your site, by modifying the Wait fields:
when starting the Monitor;
when the Monitor is already active;
Change the wait parameters by entering new values (in seconds).
For descriptions of these fields, see Wait Factor.
To display the monitor log in descending chronological order (most recent event first):
Enter L
in the command line of the
Monitor Management screen.
A screen appears displaying all Monitor log records, ordered by descending time.
To display the monitor log in ascending chronological order (oldest event first):
Enter D
in the command line of the
Monitor Management screen.
A screen appears displaying all Monitor log records, ordered by ascending time.
You can display more information about a log entry by entering the IN line command in the command line preceding the entry.
A user that is a non-administrator may also Display Log information via the profile setting "Display Monitor" set to Y on the User Profile Definition screen. This enables them to only display log information in the system administration sub-system using option 6 "Monitor Start/Close".
Note:
Sometimes errors require user intervention. The monitor will try
to re-execute its tasks each monitor cycle. To reduce logging during such
transient error situations, repeated identical log messages will be
automatically suppressed using the following scheme:
Identical messages 1 to 9 are written to monitor log as usual;
Message 10 is preceded by a warning, that future identical messages will be suppressed;
Next identical messages that are written to log (reminders) are 20, 30, 40, … 100, 200, 300, … 1000, 2000, 3000, …;
Suppression of duplicates is reset whenever the monitor task is restarted. In addition, duplicates are not suppressed, if there is a delay of 23 hours (or more) between occurrences.
No parameter settings are needed to influence suppression.
For further details, see the section LO - Display Log Information for an Object.
To purge the monitor buffer pool:
Enter P
in the command line of the
Monitor Management screen.
All entries in the Natural Buffer Pool are purged.
To purge a single buffer pool entry:
Enter E
in the command line of the
Monitor Management screen.
A window is displayed.
In this window, you enter the data for the object to be purged:
Field | Explanation |
---|---|
Library | Enter the name of the library where the object to be purged is located. |
Object | Enter the name of the object to be purged. |
DBID | Enter the ID of the database where the object to be purged is located. |
FNR | Enter the file number of the object to be purged. |
Only the object specified will be purged from the Monitor Buffer Pool.
As monitor performance is significantly worse when tracing is active, you should only activate it under the direction of your Software AG support representative.
To start or stop the monitor activity trace:
Enter the command TRACE ON or T+ (to start the trace) or TRACE OFF or T- (to stop the trace) respectively in the command line.
A message is written to the monitor's output file to record that tracing was started or stopped.
17:23:28 **** ENTIRE OUTPUT MANAGEMENT **** 2007-05-16 UserId UKSJU - Monitor Task Management - Cmd # ----- Task Status ------ Action Last Active Wait Factors Min Max Incr Curr _ 01 Idle M 2005-07-24 17:23:26 30__ 120_ 10_ 30 _ 02 Process SPOOL Queue S 2005-07-24 17:23:28 60__ 300_ 30_ 60 _ 03 Idle C 2005-07-24 17:23:28 120_ 3600 120 120 _ 04 Process Bundles R 2005-07-24 17:23:28 30__ 180_ 10_ 30 _ 05 Idle P 2005-07-24 17:23:28 40__ 240_ 20_ 40 Action values : M Main task, S Scan source queues, C Copy to container, R Create reports/bundles, P Manage printouts Valid commands: C Close, W Wake, P Purge buffer, E Purge single, L Display log Command => ___________________________________________________________________ Enter-PF1---PF2---PF3---PF4---PF5---PF6---PF7---PF8---PF9---PF10--PF11--PF12--- Help Exit Flip Do Undo Wait Menu |
This screen shows the current status of the monitor subtasks.
Field | Explanation | |
---|---|---|
# | Task number 01 to 05 | |
Task Status | Current task status | |
Action | Processing performed by this task | |
Last Active | Date and time the task was last active | |
Wait Factors | The Minimum, Maximum, Increment and Current wait times for this task. These values (except current) may be modified by pressing PF8 and overtyping with the required new value. | |
Cmd | Line command, which may take one of the following values: | |
C | Close the task. If you close task 1, all subtasks will be
closed. For any other subtask, task 1 will take over its work. |
|
W | Wake the task to perform its processing cycle. | |
P | Purge the Natural buffer pool of the task. | |
E | Purge a single object from the Natural buffer pool of the task. | |
L | Display log entries for the task. |