A Cluster Services report displays information about cache and locks in a Cluster Services environment. Adabas writes the report information directly to the Review history file. This information is sent to Review on an interval basis. The interval is chosen when the report definition is created. This feature is only available for the z/OS and OS/390 platforms.
A subtask attached as separate unit of work from the Adabas nucleus address space collects Cluster Services statistics on a user-specified timer interval. The subtask will calculate the delta of lock and cache statistics over the interval. These delta values will be written directly to the Adabas Review repository file as history data at every interval. The statistics can be viewed and maintained online through the history data menus.
Approximately 2 kilobytes of data will be written to the repository file on an interval. For this reason, Software AG recommends that the DBA reexamine and closely monitor the size of this file when the report is running. Minimal data is written for those intervals that have no statistical change. The Cluster Services report timer interval can be set in the report options and must be specified in minutes. The minimum interval is one minute and the maximum is 1440 minutes. Once the report is started, the interval is in effect until the report is stopped. A new interval can be set prior to restarting the report.
Only one Cluster Services report may be started per Database (cluster). In the Review hub environment, you may have several Cluster Services reports active, but only one for each of the monitored databases.
The Cluster Services report can be started and stopped in the same way as any other Adabas Review report (see the section Saved and Started Reports). However, it cannot be viewed online from the LS screen. It can only be viewed online from the LH and LH EX screens (see the section Managing History Data).
To define the Cluster Services report, you choose the DBID where the report will run, and the interval (in seconds). You must also choose which database and file is to receive the report data, and which SVC to use.
To access the Edit Cluster Services Report function
Enter the EX code on the command line.
The following screen will be presented:
15:37:34 A D A B A S - R E V I E W 2003-03-11
Edit Cluster Services Report LOCL=00222
+-------------------------------------------------------+
| |
| Report Name: ..... CLUSTER SERVICES REPORT |
| DBID: ...................... _____ |
| Interval: (secs) ........... ____ |
| Autostarted: ............... N |
| |
| Repository DBID: ........... _____ |
| Repository File: ........... ____ |
| Repository Database SVC: ... ___ |
| |
+-------------------------------------------------------+
Command: ______________________________________________________________________
Enter-PF1---PF2---PF3---PF4---PF5---PF6---PF7---PF8---PF9---PF10--PF11--PF12--- |
The Edit Cluster Services Report screen provides the following input fields:
| Heading | Explanation |
|---|---|
| Report Name | The name of the report. This field cannot be modified. |
| DBID | Specifies which database the report is to run on. When a DBID is entered into this field, the DBID will be automatically appended to the report name. |
| Interval | The number of seconds between each report. Report data will be written to the Review history file once for each interval. The minimum interval is one minute and the maximum is 1440 minutes. |
| Autostarted | Specifies whether the report is started when Review is initialized. |
| Repository DBID | The database that contains the Review history file. |
| Repository File | The file number of the Review history file. |
| Repository Database SVC | The SVC currently being used by the Review history database. |
