8.2.10 Monitoring HiRDB using the TP1/Resource Manager Monitor facility

This section discusses important points about using the TP1/Resource Manager Monitor (RMM) with HiRDB when OpenTP1 is linked as OLTP. For details about how to operate RMM, see the manual OpenTP1 System Operation.

Organization of this subsection
(1) Notes on creating a command for obtaining process IDs subject to monitoring
(2) Notes on using RMM operations with a HiRDB/Parallel Server

(1) Notes on creating a command for obtaining process IDs subject to monitoring

When a command for obtaining the process IDs of the processes subject to monitoring is created, the processes to be monitored must be specified. For HiRDB, _scd must be specified as the process to be monitored. Following is an example of the command for HiRDB (shell script) provided by the RMM service.

#Watched Processes
PROCESSES="_scd"

To improve this command's reliability, the coding shown below should be added. When this is done, indeterminate process IDs will not be obtained before completion of HiRDB startup. This coding should be inserted before the line where process ID is obtained in the shell script (before the comment line "#These Lines Are The Description Of Get Process_ID Process").

#System status check
get_STATUS='$PDDIR/bin/pdls 2>/dev/null | \
       /usr/bin/awk'{print$4}' | /bin/grep -v STATUS'
for i in $get_STATUS
do
 if[$i!="ACTIVE"]
 then
   exit 2
 fi
done

(2) Notes on using RMM operations with a HiRDB/Parallel Server

In the case of a HiRDB/Parallel Server, a series of commands should be specified for the resource manager subject to monitoring in the definition of the RM subject to monitoring only for the OpenTP1 located on the same host as the unit where the system manager is located.