2.4.5 Settings for handling errors
JP1/IM - EG for NNMi provides various measures for handling errors in JP1/IM - EG for NNMi, such as a tool for collecting data needed for investigation of errors and a function for automatically restarting the process in the event of an abnormal process termination.
This subsection describes the settings for handling JP1/IM - EG for NNMi errors.
- Organization of this subsection
(1) Preparations for collecting data in the event of an error (Windows only)
If a JP1/IM - EG for NNMi process stops due to an application error, while the error dialog box is displayed, use the following procedure to collect a user dump:
-
Start Task Manager.
You can use either of the following procedures to start Task Manager:
-
Right-click a blank area on the task bar and choose Task Manager.
-
Press Ctrl + Shift + Esc keys to start Task Manager.
-
-
Click the Process tab.
-
Right-click the name of the JP1/IM - EG for NNMi process that was stopped by an application error, and then choose Create Dump File.
-
When a dialog box showing the user dump output destination path opens, collect a dump from there.
- Important
-
If the error dialog box is closed, a normal dump cannot be collected, and consequently you will not be able to collect a user dump. If you closed the error dialog box by mistake (by clicking OK, for example) before collecting a user dump, reproduce the error and then collect a user dump.
(2) Preparations for collecting data in the event of an error (UNIX only)
JP1/IM - EG for NNMi provides a shell script (jegn_log.sh) as a tool for collecting data in the event of a problem. This tool enables you to collect data needed for resolving problems in batch mode.
The data collection tool of JP1/IM - EG for NNMi can collect troubleshooting data for JP1/IM - EG for NNMi and JP1/Base. For details about the data that can be collected, see 10.3 Data to be collected in the event of an error.
- About the data collection tool
-
For details about the jegn_log.sh command, see jegn_log.sh (UNIX only) in Chapter 7.
In the event of a problem, you might need to obtain a core dump to facilitate investigation of the cause. Output of a core dump depends on the user environment settings. Check the settings described below.
(a) Setting the kernel parameters regarding core dump (Linux only)
In kernel parameter of Linux (kernel.core_pattern), when the output destination of core dump file is set to directory other than collection-target log file directory, or when the name of the core dump file is changed from the default setting, the data collection tool would not be able to acquire the core dump file when the tool is executed.
The data collection tool collects files whose file names start with core in the following default collection-target log file directories.
- For physical hosts:
-
/var/opt/jp1eg4nnmi/log
- For logical hosts:
-
shared-directory/jp1eg4nnmi/log
Depending on the setting of kernel.core_pattern, it might be necessary to check and address the following points before executing the data collection tool.
-
When the output directory of a core dump file is changed
Make a copy of the core dump file in the default output directory.
-
When the file name of a core dump file is changed
Change the file name of the core dump file to a name beginning with core.
(b) Setting ABRT for core dump files (Linux only)
In a Linux with Automatic Bug Reporting Tool (ABRT) installed, ABRT can be configured to allow limited processes, OS user accounts, or user groups to generate core dump files. In such a case, you cannot investigate problems because a core dump file might not be generated in case of core dump output events such as a segmentation failure in a JP1/IM - EG for NNMi process, a bus failure.
Depending on your operation, you should change the ABRT settings to ensure that processes or OS user accounts or user groups that run JP1/IM - EG for NNMi are allowed to generate core dump files. For details, see the documentation for your Linux.
(c) The systemd settings related to core dump files (Linux only)
These settings apply to Linux environments where the settings file for core dump file names (/proc/sys/kernel/core_pattern) begins with the character string "|/usr/lib/systemd/systemd-coredump".
If the operation settings file for core dump files (/etc/systemd/coredump.conf) includes a setting that specifies that no core dump files are to be created, no core dump file will be output and users will not be able to investigate the failure in situations such as when a segmentation fault or a bus failure occurs in a JP1/IM - EG for NNMi process.
Based on operations to be performed, revise the settings in the operation settings file for core dump files (/etc/systemd/coredump.conf) so that core dump files are created. For details, see the documentation for your Linux.
(3) Settings for restarting after an abnormal process termination
A process is generated when JP1/IM - EG for NNMi starts. JP1/IM - EG for NNMi enables you to make settings so that the process will restart automatically after it has terminated abnormally for some reason.
To specify the settings for process restart after abnormal termination of the process:
-
Specify process restart.
Edit the extended startup process definition file (jegn_service_0700.conf). This file is stored at the following locations:
- In Windows:
-
• Physical host: EG-for-NNMi-path\conf
• Logical host: shared-folder\JP1EG4NNMI\conf
- In UNIX:
-
• Physical host: /etc/opt/jp1eg4nnmi/conf
• Logical host: shared-directory/jp1eg4nnmi/conf
The relevant parameter is the fourth value of the values separated by the vertical bar (|). Set either 0 (do not restart (default)) or 1 (restart). Do not change any other parameter values.
For details about the extended startup process definition file, see F.1 Extended startup process definition file (jegn_service_0700.conf).
-
Apply the settings.
Either restart JP1/IM - EG for NNMi or execute the jegn_spmd_reload command to apply the settings. For details about the jegn_spmd_reload command, see jegn_spmd_reload in 7. Commands.
-
Suppress error reporting to Microsoft (Windows only).
In the case of Windows, you must specify settings so that the Microsoft error reporting dialog box, which is displayed in the event of error detection, will not be displayed. This is because a process cannot restart if this dialog box is displayed.
1. In the Run text box, enter gpedit.msc, and then click OK.
2. In the Local Group Policy Editor, click Computer Configuration, Administrative Templates, Windows Components, and then Windows Error Reporting.
3. Right-click Disable Windows Error Reporting in the right pane of the window, and then select Edit.
4. In the settings window, select the Enabled radio button.
5. Click OK.
(4) Setting JP1 event issuance in the event of an abnormal process termination
JP1/IM - EG for NNMi enables you to issue JP1 events in the event of automatic recovery from a process error or abnormal termination. By issuing JP1 events, you can manage the error logs of JP1/IM - EG for NNMi itself.
To specify settings for JP1 event issuance:
-
Set JP1 event issuance.
Edit the JP1/IM - EG for NNMi parameter definition file (jegn_param_V7.conf). This file is stored at the following locations:
- In Windows:
-
• Physical host: EG-for-NNMi-path\conf
• Logical host: shared-folder\JP1EG4NNMI\conf
- In UNIX:
-
• Physical host: /etc/opt/jp1eg4nnmi/conf
• Logical host: shared-directory/jp1eg4nnmi/conf
In this file, SEND_PROCESS_TERMINATED_ABNORMALLY_EVENT and SEND_PROCESS_RESTART_EVENT are the JP1 event issuance setting parameters. To issue JP1 events, change the values of these parameters to dword:1.
For details about the JP1/IM - EG for NNMi parameter definition file, see F.2 JP1/IM - EG for NNMi parameter definition file (jegn_param_V7.conf).
-
Execute the jbssetcnf command to apply the definition information.
For details about the jbssetcnf command, see the JP1/Base User's Guide.
-
Restart JP1/IM - EG for NNMi.
The specified settings take effect after the restart.