1.2.3 Re-creating a database and changing its settings
- Organization of this subsection
(1) Re-creating the command execution log
To re-create the command execution log:
-
Stop JP1/IM - Manager.
-
Stop JP1/Base.
-
Delete the command execution log file, the action information file, and the action hosts file shown in the table below.
- In Windows:
-
Table 1‒12: Files to delete (Windows) Information type
Files to delete
Command execution log file
All files under Base-path\log\COMMAND\
All files under shared-folder\jp1base\log\COMMAND
Action information file
Console-path\log\action\actinf.log
shared-folder\jp1cons\log\action\actinf.log
Action hosts file
Console-path\log\action\acttxt{1|2}.log
shared-folder\jp1cons\log\action\acttxt{1|2}.log
- In UNIX:
-
Table 1‒13: Files to delete (UNIX) Information type
Files to delete
Command execution log file
All files under /var/opt/jp1base/log/COMMAND/
All files under shared-directory/jp1base/log/COMMAND/
Action information file
/var/opt/jp1cons/log/action/actinf.log
shared-directory/jp1cons/log/action/actinf.log
Action hosts file
/var/opt/jp1cons/log/action/acttxt{1|2}.log
shared-directory/jp1cons/log/action/acttxt{1|2}.log
-
Start JP1/Base.
-
Start JP1/IM - Manager.
Restarting JP1/Base and JP1/IM - Manager and executing a command from JP1/IM - View or an automated action re-creates the command execution log.
(2) Procedure for re-creating the monitored object database and the host information database
To re-create the monitored object database and the host information database:
-
Stop JP1/IM - Manager.
-
Back up the files.
Back up the Scope-path\database\ folder.
-
Re-create the monitored object database.
Executing the jcsdbsetup -f command deletes the existing monitored object database, and then re-creates the object database.
-
Re-create the host information database.
First, delete the files from the Scope-path\database\jcshosts\ folder, and then execute the following command:
jcshostsimport -r host-information-file (jcs_hosts)
-
Start JP1/IM - Manager.
(3) Procedure for re-creating the event database
The procedure differs depending on the version of JP1/Base that is installed on the target host whose event database you are re-creating.
(a) Manager (JP1/Base 09-00 or later)
The procedure differs depending on whether the integrated monitoring database is used.
- If the integrated monitoring database for JP1/IM - Manager is not used
-
Using the jevdbinit command of JP1/Base, initialize the event database.
For details about how to initialize the event database of JP1/Base, see the description about initializing the event database in the chapter that explains how to set up the Event Service environment in the JP1/Base User's Guide.
If you changed the serial numbers by executing the jevdbinit command with the -s option, you must re-create the command execution log.
For details about how to re-create the command execution log, see 1.2.3(1) Re-creating the command execution log.
- If the integrated monitoring database for JP1/IM - Manager is used
-
You can use the following procedure to initialize the event database:
-
Stop JP1/IM - Manager.
-
Execute the JP1/Base jevdbinit command without the -s option.
If you execute the jevdbinit command without the -s option, the serial numbers in the pre-initialization event database are inherited.
If you changed the serial numbers by executing the jevdbinit command with the -s option, you must set up the integrated monitoring database again and re-create the command execution log. Note that when you are using the Intelligent Integrated Management Base, after setting up the integrated monitoring database again, and then before start JP1/IM - Manager, you need to delete the file shown in the table below.
OS
Files to delete
Windows
Manager-path\data\imdd\actevent.ser
Manager-path\data\imdd\imdd_nodeStatus.ser
Manager-path\data\imdd\jddactseq.ser
shared-folder\data\imdd\actevent.ser
shared-folder\data\imdd\imdd_nodeStatus.ser
shared-folder\data\imdd\jddactseq.ser
UNIX
/var/opt/jp1imm/data/imdd/actevent.ser
/var/opt/jp1imm/data/imdd/imdd_nodeStatus.ser
/var/opt/jp1imm/data/imdd/jddactseq.ser
shared-directory/jp1imm/data/imdd/actevent.ser
shared-directory/jp1imm/data/imdd/imdd_nodeStatus.ser
shared-directory/jp1imm/data/imdd/jddactseq.ser
Before setting up the integrated monitoring database again, unset up the database by executing the jcodbunsetup command.
For details about how to re-create the command execution log, see 1.2.3(1) Re-creating the command execution log.
For details about the jevdbinit command, see the chapter on commands in the JP1/Base User's Guide.
Note that if you execute the jevdbinit command with the -s option specified, you must use Central Scope to select the root monitoring node, change the status, and delete the status change event logs.
-
(b) Agent (JP1/Base 07-51 or earlier)
Using the jevdbinit command of JP1/Base, initialize the event database. There is no need to delete and re-create the event database.
For details about how to initialize the event database of JP1/Base, see the description about initializing the event database in the chapter that explains how to set up the Event Service environment in the JP1/Base User's Guide.
- Important
-
If an agent initializes the event database, JP1/Base discards the events without registering them in the event database. Consequently, if the correct procedure is not followed, it might become impossible to transfer some of the events after the event database is initialized.
(c) Agent (JP1/Base 07-00 or earlier)
When an event database is re-created, the following problem occurs:
-
At the JP1 event forwarding destination host, the processing performance for accepting, registering, and acquiring JP1 events deteriorates.
This is because re-creation initializes the event database at the forwarding source, creating a mismatch with the management information in the event database at the forwarding destination.
- Important
-
If an agent initializes the event database, JP1/Base discards the events without registering them in the event database. Consequently, if the correct procedure is not followed, it might become impossible to transfer some of the events after the event database is initialized.
To prevent this problem from occurring, re-create event databases using the following procedure.
To re-create event databases:
-
Stop JP1/Base.
-
Stop JP1/Base at all forwarding destination hosts defined in the forwarding setting file (forward) of the JP1/Base you stopped in Step 1.
If data is forwarded from the JP1/Base at the forwarding destination host to yet another host, stop this forwarding destination as well. If JP1/IM - Manager has been installed on the host that is to be stopped, stop JP1/IM - Manager beforehand.
For details about the forwarding setting file (forward), see the sections that describe the settings for JP1 event forwarding in the chapter that explains how to set up an Event Service environment in the JP1/Base User's Guide.
-
Delete the event databases of the JP1/Bases you stopped in Steps 1 and 2.
If you need to view the content of the event databases, use the jevexport command of JP1/Base to output this content to a CSV file. Note that you cannot re-create an event database from an output CSV file.
For details about the jevexport command, see the chapter on commands in the JP1/Base User's Guide.
-
Start the JP1/Base (and JP1/IM - Manager) that you stopped in Step 2.
-
Start the JP1/Base that you stopped in Step 1.
Starting JP1/Base in Steps 4 and 5 re-creates the event databases.
For this example, assume that event databases will be re-created in the system configuration shown in the following figure.
|
To re-create (delete) the event database of host A, it is necessary to delete the event databases of Hosts C and F, which are the forwarding destination hosts for JP1 events.
(4) Procedure for re-creating the file for accumulated response-waiting events
-
Stop JP1/IM - Manager.
-
Delete the file for accumulated response-waiting events.
Table 1‒14: Files to delete OS
Files to delete
Windows
Console-path\log\response\resevent.dat
shared-folder\jp1cons\log\response\resevent.dat
UNIX
/var/opt/jp1cons/log/response/resevent.dat
shared-directory/jp1cons/log/response/resevent.dat
-
Start JP1/IM - Manager.
(5) Procedures for expanding the IM database size
This subsection explains how to expand the IM database size on a physical host, and in a cluster environment. If you create the IM database with L specified for the database size in the setup information file (jimdbsetupinfo.conf), the IM database size cannot be expanded.
(a) Procedure for expanding the IM database size on a physical host
The procedure for expanding the IM database size differs depending on whether you need to continue system monitoring without using the IM database during the expansion process. The procedure for each scenario is described below.
-
Procedure when monitoring events without using the IM database during the expansion process
-
Isolate the integrated monitoring database and the IM Configuration Management database.
Isolate the integrated monitoring database and the IM Configuration Management database so that Central Console only uses the JP1/Base event database.
Execute the following command, and then restart JP1/IM - Manager:
jcoimdef -db OFF
For details about the jcoimdef command, see jcoimdef in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
If JP1/IM - MO is being used, also stop the JP1/IM - Message Optimizer service of JP1/IM - MO on the connection source.
-
Back up the database.
Execute the jimdbbackup command with the -m EXPAND option specified.
For details about the jimdbbackup command, see jimdbbackup in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
-
Unset up both the integrated monitoring database and the IM Configuration Management database.
Unset up only those databases that have been set up.
-
Edit the setup information file.
Change the size specified in the database size (IMDBSIZE) of the setup information file.
-
Stop JP1/IM - Manager.
-
Stop JP1/Base.
-
Restart the OS.
-
Start JP1/Base.
-
Start JP1/IM - Manager.
-
Set up both the integrated monitoring database and the IM Configuration Management database.
Set up only those databases that were unset up in Step 3.
During setup, you need to specify a database size that is larger than the backup size and the same database directory that was used during the backup.
-
Recover the database.
Execute the jimdbrecovery command with the -m EXPAND option specified.
For details about the jimdbrecovery command, see jimdbrecovery in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
-
Restart the JP1/IM3-Manager service.
Execute the following command and then restart JP1/IM - Manager:
jcoimdef -db ON
For details about the jcoimdef command, see jcoimdef in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
If JP1/IM - MO is being used, also start the JP1/IM - Message Optimizer service of JP1/IM - MO on the connection source.
-
When you are using the Intelligent Integrated Management Base, execute the jddupdatetree command in new and rebuilding mode.
For details about the jddupdatetree command, see jddupdatetree in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
-
Procedure when stopping system monitoring via Central Console
-
Stop the JP1/IM3-Manager service.
-
Back up the database.
Execute the jimdbbackup command with the -m EXPAND option specified.
For details about the jimdbbackup command, see jimdbbackup in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
-
Unset up the integrated monitoring database and the IM Configuration Management database.
Unset up only those databases that have been set up.
-
Edit the setup information file.
Change the size specified in the database size (IMDBSIZE) parameter in the setup information file.
-
Stop JP1/Base.
-
Restart the OS.
-
Start JP1/Base.
-
Set up the integrated monitoring database and the IM Configuration Management database.
Set up only those databases that you unset up in Step 3.
During setup, you need to specify a database size that is larger than the size of the database you backed up, and the same database directory that was used during the backup.
-
Recover the database.
Execute the jimdbrecovery command with the -m EXPAND option specified.
For details about the jimdbrecovery command, see jimdbrecovery in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
-
Start the JP1/IM3-Manager service.
-
When you are using the Intelligent Integrated Management Base, execute the jddupdatetree command in new and rebuilding mode.
For details about the jddupdatetree command, see jddupdatetree in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
(b) Procedure for expanding the IM database size in a cluster environment
The procedure for expanding the database size differs depending on whether you need to continue system monitoring via Central Console during the expansion process. The procedure for each scenario is described below.
-
Procedure when continuing system monitoring via Central Console (with limited functionality)
-
Isolate the integrated monitoring database and the IM Configuration Management database.
Isolate the integrated monitoring database and the IM Configuration Management database so that Central Console only uses the JP1/Base event database.
Execute the following command, and then restart JP1/IM - Manager:
jcoimdef -db OFF -h logical-host-name
For details about the jcoimdef command, see jcoimdef in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
If JP1/IM - MO is being used, also stop the JP1/IM - Message Optimizer service of JP1/IM - MO on the connection source.
-
Stop the cluster service (JP1/IM3-Manager DB Cluster Service_logical-host-name) of the IM database.
Stop the cluster service (JP1/IM3-Manager DB Cluster Service_logical-host-name) of the IM database registered in the cluster software.
-
Back up the database.
Execute the jimdbbackup command with the -m EXPAND option specified.
For details about the jimdbbackup command, see jimdbbackup in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
-
Unset up the integrated monitoring database and the IM Configuration Management database.
Unset up only those databases that have been set up.
-
Edit the cluster setup information file.
Change the size specified in the database size (IMDBSIZE) of the cluster setup information file.
-
Stop JP1/IM - Manager.
-
Stop JP1/Base.
-
Restart the OS.
-
Start JP1/Base.
-
Start JP1/IM - Manager.
-
Set up the integrated monitoring database and the IM Configuration Management database.
Set up only those databases that were unset up in Step 4.
During setup, you need to specify a database size that is larger than the backup size and the same database directory that was used during the backup.
-
Recover the database.
Execute the jimdbrecovery command with the -m EXPAND option specified.
For details about the jimdbrecovery command, see jimdbrecovery in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
-
Start the cluster service (JP1/IM3-Manager DB Cluster Service_logical-host-name) of the IM database.
Start the cluster service (JP1/IM3-Manager DB Cluster Service_logical-host-name) of the IM database you stopped in Step 2.
-
Restart the JP1/IM3-Manager service.
Execute the following command and then restart JP1/IM - Manager:
jcoimdef -db ON -h logical-host-name
For details about the jcoimdef command, see jcoimdef in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
If JP1/IM - MO is being used, also start the JP1/IM - Message Optimizer service of JP1/IM - MO on the connection source.
-
When you are using the Intelligent Integrated Management Base, execute the jddupdatetree command in new and rebuilding mode.
For details about the jddupdatetree command, see jddupdatetree in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
-
Procedure when stopping system monitoring via Central Console
-
Stop the JP1/IM3-Manager service and the cluster service (JP1/IM3-Manager DB Cluster Service_logical-host-name) of the IM database.
Stop the JP1/IM3-Manager service and the cluster service (JP1/IM3-Manager DB Cluster Service_logical-host-name) of the IM database registered in the cluster software.
-
Back up the database.
Execute the jimdbbackup command with the -m EXPAND option specified.
For details about the jimdbbackup command, see jimdbbackup in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
-
Unset up the integrated monitoring database and the IM Configuration Management database.
Unset up only those databases that have been set up.
-
Edit the cluster setup information file.
Change the size specified in the database size (IMDBSIZE) parameter in the cluster setup information file.
-
Stop JP1/Base.
-
Restart the OS.
-
Start JP1/Base.
-
Set up the integrated monitoring database and the IM Configuration Management database.
Set up only those databases you unset up in Step 3.
During setup, you need to specify a database size that is larger than the size of the database you backed up, and the same database directory that was used during the backup.
-
Recover the database.
Execute the jimdbrecovery command with the -m EXPAND option specified.
For details about the jimdbrecovery command, see jimdbrecovery in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
-
Start the JP1/IM3-Manager service and the cluster service (JP1/IM3-Manager DB Cluster Service_logical-host-name) of the IM database.
Start the JP1/IM3-Manager service and cluster service (JP1/IM3-Manager DB Cluster Service_logical-host-name) of the IM database you stopped in Step 1.
-
When you are using the Intelligent Integrated Management Base, execute the jddupdatetree command in new and rebuilding mode.
For details about the jddupdatetree command, see jddupdatetree in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
(6) Procedure for changing the IM database port
To change the IM database port:
-
Stop JP1/IM3-Manager service.
If JP1/IM - MO is being used, also stop the JP1/IM - Message Optimizer service of JP1/IM - MO on the connection source.
-
Back up the database.
Execute the jimdbbackup command with the -m MAINT option specified.
For details about the jimdbbackup command, see jimdbbackup in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
-
Unset up both the integrated monitoring database and the IM Configuration Management database.
Unset up only those databases that have been set up.
-
Edit the setup information file.
Change the port number described in the setup information file.
-
Stop JP1/Base.
-
Restart the OS.
-
Start JP1/Base.
-
Set up both the integrated monitoring database and the IM Configuration Management database.
Set up only those databases that were unset up in Step 3.
-
Recover the database.
Execute the jimdbrecovery command with the -m MAINT option specified.
For details about the jimdbrecovery command, see jimdbrecovery in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
-
Start JP1/IM - Manager.
If JP1/IM - MO is being used, also start the JP1/IM - Message Optimizer service of JP1/IM - MO on the connection source.
-
When you are using the Intelligent Integrated Management Base, execute the jddupdatetree command in new and rebuilding mode.
For details about the jddupdatetree command, see jddupdatetree in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
(7) Procedure for rebuilding the IM database
The procedure below explains the procedure for rebuilding the IM database that is required when you change the manager's host name. After you change the host name of a physical or logical host, you need to rebuild the IM database. Note that when the host name of a logical host is changed, you need to re-register the service created in this procedure in the IM database service to be registered in the cluster software.
To rebuild the IM database:
-
Stop JP1/IM3-Manager service.
If JP1/IM - MO is being used, also stop the JP1/IM - Message Optimizer service of JP1/IM - MO on the connection source.
-
Unset up both the integrated monitoring database and the IM Configuration Management database.
Unset up only those databases that have been set up.
-
Change the name of the host on which JP1/IM - Manager has been installed.
-
This step is not necessary if the host name is not changed.
-
Stop JP1/Base.
-
Restart the OS.
-
Start JP1/Base.
-
Set up both the integrated monitoring database and the IM Configuration Management database.
Set up only those databases that were unset up in Step 2.
When you are setting up a logical host, you need to edit the logical host name in the cluster setup information file.
-
When you are using the Intelligent Integrated Management Base, delete the file shown in the table below.
OS
Files to delete
Windows
Manager-path\data\imdd\actevent.ser
Manager-path\data\imdd\imdd_nodeStatus.ser
Manager-path\data\imdd\jddactseq.ser
shared-folder\data\imdd\actevent.ser
shared-folder\data\imdd\imdd_nodeStatus.ser
shared-folder\data\imdd\jddactseq.ser
UNIX
/var/opt/jp1imm/data/imdd/actevent.ser
/var/opt/jp1imm/data/imdd/imdd_nodeStatus.ser
/var/opt/jp1imm/data/imdd/jddactseq.ser
shared-directory/jp1imm/data/imdd/actevent.ser
shared-directory/jp1imm/data/imdd/imdd_nodeStatus.ser
shared-directory/jp1imm/data/imdd/jddactseq.ser
-
Start JP1/IM - Manager.
Start the JP1/IM - Manager of the host to be changed.
If JP1/IM - MO is being used, also start the JP1/IM - Message Optimizer service of JP1/IM - MO on the connection source.
-
When you are using the Intelligent Integrated Management Base, execute the jddupdatetree command in new and rebuilding mode.
For details about the jddupdatetree command, see jddupdatetree in Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
- Important
-
If you rebuild the IM database after changing the host name, you cannot recover the database. Therefore, as needed, use the jcoevtreport command to output and save JP1 events, and use the jcfexport command to save the IM configuration management information. For details about commands, see Chapter 1. Commands in the JP1/Integrated Management 3 - Manager Command, Definition File and API Reference.
If you do not change the host name, you can recover the database. For details, see 1.2.2 Database backup and recovery.