9.2.1 Troubleshooting problems with setting up or starting a service
This subsection describes how to handle problems that are related to service setup or service startup.
- Organization of this subsection
(1) A Performance Management program service does not start
The following describes the possible causes and the resolutions.
-
PFM - Manager is not running.
If PFM - Manager and PFM - RM for Platform are both on the same host, the PFM - RM for Platform services cannot start if PFM - Manager is not running. Check if the PFM - Manager service is running; start it if it is not running. For details about how to start services, see the chapter that describes startup and termination of Performance Management in the Job Management Partner 1/Performance Management User's Guide.
-
The same port number is set for more than one Performance Management program service.
When the same port number is set for multiple Performance Management program services, none of those Performance Management program services can be started.
The default is for the system to assign port numbers automatically, in which case there cannot be any duplicated port numbers. If you set specific port numbers for Performance Management program services during Performance Management setup, check the port numbers that you specified. If you specified the same port number for more than one Performance Management program service, specify different port numbers. For details about how to set port numbers, see the chapter that describes installation and setup in the Job Management Partner 1/Performance Management Planning and Configuration Guide.
-
There is an error in the settings for the Store database storage directory.
If any of the directories listed below is set to be inaccessible or does not exist, the Remote Monitor Store service cannot start. Check the specified directory names and attribute settings and correct any errors.
-
Store database storage directory
-
Store database backup directory
-
Store database partial backup directory
-
Store database export directory
-
Store database import directory
If you specify these directories for multiple Remote Monitor Store services, the Remote Monitor Store services cannot start. Check the specified directory settings and correct any errors.
-
-
An unauthorized method was used to change a machine's host name.
For details about how to rename a host on a machine, see the chapter that describes installation and setup in the Job Management Partner 1/Performance Management Planning and Configuration Guide. If you rename a host using an unauthorized method, the Performance Management program services might not start.
-
An error occurred in the service control manager.
If you execute the jpcspm start command in Windows, the error message An error occurred in the Windows service control manager might be displayed and startup of the service might fail. If this has occurred, re-execute the jpcspm start command. If the same event occurs frequently, edit the jpccomm.ini file to change the retry interval and retry count for service startup processing during execution of the jpcspm start command. For details about how to change the retry interval and retry count, see the chapter that describes startup and termination of Performance Management in the Job Management Partner 1/Performance Management User's Guide.
(2) It takes too long for a service to start once startup is requested
It might take a long time for a service to actually start once you execute the jpcspm start command or you start a service by choosing the Services icon. If this is caused by either of the reasons listed below, the time required for starting the service will be reduced for subsequent startups of the service.
-
During the first startup after a new instance has been added, the indexes of the Store database are created, which might cause the services to take longer than usual to start.
-
If the Store service is unable to perform normal termination processing for a reason such as a power shut-off, the indexes of the Store database are rebuilt during the next restart, which might cause the services to take longer than usual to start.
(3) Communication is not performed properly when a program service starts immediately after a Performance Management program service has stopped
Immediately after a Performance Management program service has stopped, another program service might be started using the same port that the stopped service was using. In such a case, communication might not be performed properly. You can prevent this problem by doing the following:
-
Use fixed port numbers for the Performance Management program services.
For details about how to set port numbers, see the chapter that describes installation and setup in the Job Management Partner 1/Performance Management Planning and Configuration Guide.
-
Specify a value for TCP_TIMEWAIT.
In the TCP_TIMEWAIT value, set the connection wait time to the following default value:
In Windows Server 2003, Windows Server 2008, and Windows Server 2012: 2 minutes
(4) The message "The disk capacity is insufficient" is displayed, and then the Master Store service or the Remote Monitor Store service stops
Data storage in the Store database is interrupted when there is not enough free space on the disk used by the Store database. In such a case, the message The disk capacity is insufficient is displayed and the Master Store service or the Remote Monitor Store service stops.
If this message is displayed, take one of the following actions:
-
Provide sufficient disk space.
Determine the amount of disk space required for the Store database and change the Store database storage location to a disk with sufficient available space. For details about how to determine the disk space required for the Store database, see A. Estimating System Requirements. For details about how to change the Store database storage location, see 3.6.1 Changing performance data storage locations.
-
Change the data retention conditions for the Store database.
Change the data retention conditions for the Store database by adjusting the maximum data capacity. For details about how to change the data retention conditions for the Store database, see the chapter that describes management of operation monitoring data in the Job Management Partner 1/Performance Management User's Guide.
If the Master Store service or the Remote Monitor Store service will not start after you have taken these actions, an unrecoverable logical conflict has occurred in the Store database. In such a case, restore the Store database from a backup, and then start the Master Store service or the Remote Monitor Store service. If no backup is available, initialize the Store database, and then start the Master Store service or the Remote Monitor Store service. To initialize the Store database, you must delete all the following files from the storage directory for the Store database:
-
Files with the extension .DB
-
Files with the extension .IDX
For details about the storage directory for the Store database, see 3.6.1 Changing performance data storage locations.
(5) The Remote Monitor Collector service of PFM - RM does not start
If the PFM - RM host is running Windows, startup of the Remote Monitor Collector service might fail during PFM - RM startup, and one of the following messages might be displayed in the Windows event log when Windows is restarted:
-
The Service Name service hung on starting.
-
Service Name service hung on startup.
Because this problem is caused by a timeout in Windows service control manager, it tends to occur when PFM - Manager's communication load is high and a response from PFM - Manager takes time. This problem occurs when all of the following conditions are satisfied:
-
JP1/PFM - Manager's communication load is high.
For example, many copies of PFM - RM are being started concurrently.
-
In Windows Services applet, the startup type is set to Automatic for PFM - RM services.
-
The OS is restarted.
To avoid this problem, take either of the following steps:
-
When starting a service at the time the OS is restarted, start it by executing the jpcspm start command instead of starting it from Windows service control manager.
-
Shorten the PFM - RM startup time by using the setting described below for the PFM - RM host.
This setting shortens the reconnection processing during startup of the PFM - RM service if PFM - Manager cannot be connected to. In this case, the probability of the PFM - RM service starting in a stand-alone mode increases.
To shorten the startup time of PFM - RM, change [Agent Collector x Section]# in the startup information file (jpccomm.ini) and the NS Init Retry Count label of [Agent Store x Section]# from NS Init Retry Count =2 to NS Init Retry Count =1.
- #
-
The PFM - RM product ID is entered for x. For details about product IDs, see B. List of Identifiers. If multiple copies of PFM - RM are installed on the same host, specify an NS Init Retry Count label value for each product ID.
The startup information file (jpccomm.ini) is stored in the following directory:
- If the PFM - RM host is a physical host
-
installation-folder\jpccomm.ini
- If the PFM - RM host is a logical host
-
environment-directory#\jp1pc\jpccomm.ini
- #
-
Indicates the directory on the shared disk specified when the logical host was created.
(6) Failure Audit (Event ID: 4625 or 4776) is recorded in the Windows security event log.
When the monitored host is running Windows, Failure Audit (Event ID: 4625 or 4776) might be recorded in the Windows security event log.
PFM - RM for Platform makes a WMI connection to the monitored host by using the user name and password specified in the account information#1 when the monitoring target was set up. However, WMI also tries to connect to the monitored host by using the process-executing account information#2 (in which a user name and password were specified when the instance environment was set up), resulting in this problem. Even when Failure Audit (Event ID: 4625 or 4776) is displayed in the Windows security event log, there is no problem if performance information has been collected.
To avoid this problem, take the following steps:
-
Create account information having the same user name and password on the PFM - RM host and the monitored host.
-
Specify the user name and password (in the account information created in step 1) in the setting items for the instance environment and the setting items for the monitoring target as described below.
-
RMHost_User#2 setting item in the account information for the instance environment: User name
-
RMHost_Password#2 setting item in the account information for the instance environment: Password
-
User#1 setting item in the account information for the monitoring target: User name
-
Password#1 setting item in the account information for the monitoring target: Password
For details about how to specify these settings, see 3.1.4 Setup procedure for the Windows edition.
-
#1: This is a setting item (User or Password) in common account information (wmi) when common account information is used.
#2: This is a setting item (User or Password) in common account information (pfmhost) when common account information is used.