7.5.1 In Windows
- Organization of this subsection
(1) Collecting dump information
To collect dump information:
-
Open Task Manager.
-
Select the Processes tab.
-
Right-click the process name for which you want to collect dump information, and then select Create Dump File.
Dump files are stored in the following folder:
system-drive\Users\user-name\AppData\Local\Temp
-
Obtain the dump file from the folder created in step 3.
If you have changed the environment variable settings so that dump files are output to a different folder, obtain the dump file from that folder.
(2) Executing the data collection command
To collect data for troubleshooting, use the jpcras command. The procedure for executing the data collection command follows. The operations described here must be executed by an OS user who has Administrator permissions.
To execute the data collection command:
-
Log on to the host on which the service is installed whose data you want to collected.
-
At the command prompt, execute the following command to enable the command extension function of the command interpreter:
cmd /E:ON
-
Specify the data you want to collect and the storage folder, and then execute the jpcras command.
In the following command specification example, all information that can be collected using the jpcras command is saved in the c:\tmp\jpc\agt folder:
jpcras c:\tmp\jpc\agt all all
When the jpcras command is executed, the jpctool service list -id * -host * command is internally executed in order to obtain a PFM service list and check the startup status. If a firewall is installed between the command execution host and the host of another Performance Management system, or if the system configuration is large, execution of the jpctool service list -id * -host * command may take a long time. In such a case, by setting the environment variable JPC_COLCTRLNOHOST to 1, you can suppress the processing of the jpctool service list -id * -host * command, thereby shortening the command execution time.
For details about the jpcras command, see the chapter that explains commands in the manual JP1/Performance Management Reference.
- Note on executing the command
-
If the user account control functionality (UAC) is enabled in the operating system, the User Account Control dialog box might be displayed during command execution. If it is displayed, click the Continue button to continue data collection, or click the Cancel button to cancel data collection.
(3) Executing the data collection command (in a logical host environment)
When Performance Management is running on a logical host, data is stored on a shared disk, and must therefore be collected from both the active server and the standby server.
To collect data for troubleshooting, use the jpcras command. The procedure for executing the data collection command follows. The operations described here must be executed by an OS user who has Administrator permissions.
To execute the data collection command in a logical host operation environment:
-
Bring the shared disk online.
The data of the logical host is stored on the shared disk. On the executing node, make sure that the shared disk is online and then collect the data.
-
On both the active server and standby server, specify the data you want to collect and the storage folder, and then execute the jpcras command.
In the following command specification example, all information that can be collected using the jpcras command is saved in the c:\tmp\jpc\agt folder:
jpcras c:\tmp\jpc\agt all all
If you execute the jpcras command without specifying the lhost argument, all Performance Management data is collected from the node's physical host and logical host. If a Performance Management instance is running in a logical host environment, log files on the shared disk are collected.
If you execute the jpcras command on a node at which the shared disk is offline, files on the shared disk cannot be collected, but the process terminates normally without causing an error.
- Note:
-
Collect data from both the executing node and the standby node by executing the data collection command. To investigate what happened before and after a failover, you need data from both the active server and the standby server.
For details about the jpcras command, see the chapter that explains commands in the manual JP1/Performance Management Reference.
-
Collect cluster software data.
This data is required in order to determine whether the problem occurred in the cluster software or in Performance Management. Collect data that can be used to investigate start, stop, and other control requests issued by the cluster software to Performance Management, and the results.
(4) Collecting a Windows event log
In the Windows Event Viewer window, collect the content of System and Applications.
(5) Checking the operation content
Check the operational details having to do with the problem and record them. You need to check the following types of information:
-
Detailed operation content
-
Time at which the problem occurred
-
Machine configuration (including the version of each OS, host name, and configuration of PFM - Manager and PFM - RM for Virtual Machine)
-
Whether the problem is reproducible
-
If the user logged on from PFM - Web Console, the Performance Management user name used for the logon
(6) Collecting error information from windows
Obtain a hardcopy of the following information:
-
The window operation when the application error occurred
-
The error message dialog box
If there is detailed information, also make a hardcopy of that information.
-
The Command Prompt window or the Administrator Console window, if the error occurred during command execution
To print the Command Prompt window or the Administrator Console window, specify the following settings for the Command Prompt Properties window:
-
Edit options on the Options page
Select Quick Edit Mode.
-
Layout page
For Screen buffer size, set Height to 500.
-
(7) Collecting other information
Collect other necessary information.
-
Content of Accessories > System Tools > System Information