If the execution of a job does not advance beyond the range between 0% (the system under the control of the managing server does not receive the job) and 50% (the job does not reach beyond the relay system), the following problems can be suspected:

(a) PC/WS for relay system or client stopped

The service (Client Install Service) for the relay system or the client may have stopped.

(b) Client's own host name invalid

A client's own host name cannot begin with a numeric character.

If the client is under direct control of the managing server, the client's own host name should not be more than 64 characters long.

If the host name of the managing server is more than 64 characters, the relay system cannot return job results to the managing server.

Host names are not case sensitive. Therefore, make sure that a case-sensitive host name is not assigned (e.g., DMP011 vs. dmp011).

(c) Inconsistent IP addresses between the hosts file for the managing server and a lower system

A job cannot be executed if the IP address of the hosts file for the managing server and the IP address of a lower system are not in agreement.

(d) Invalid hosts file

In some cases, remote installation fails because the IP address of a higher system is unknown. If this problem occurs, check the following items:

• Whether the host name of the higher system is defined in the hosts file
• Whether the last line in the hosts file ends with a line feed

(e) Inconsistency between client's own host name and the destination specified in managing server

If the name of the client is different from the name specified in the system configuration of the managing server or in a destination, the remote installation process does not terminate. In this case, you should make sure that the correct client name is used.

(f) Abnormal termination of remote installation post-processing, AIT file, or recorder-file processing

If remote installation post-processing or the processing of the AIT file or recorder file does not terminate normally, the installation on the client side falls into an endless wait state.

If the remote installation process hangs, the following problems can be suspected:

• During the termination process, a RegisterWindowMessage is not issued to the AIT or recorder file.
• The program-termination code specified in the installation script is not correctly specified.

(g) Invalid host-name specification in DNS environment

If the client belongs in a domain, the host name of the client managed by JP1/Software Distribution is hostname.domainname. If, for example, the client's host name is client-111, and the name of the domain in which the client belongs is net02.abc.co.us, then the host name managed by JP1/Software Distribution is:

client-111.net02.abc.co.us

(h) There is a duplicate destination with different host IDs

If a duplicated client exists during operations in which host IDs are used, a job may be executed for an incorrect destination, resulting in a hung job.

If this is the case, delete the unneeded client and then re-execute the job. For details about how to search for and delete unneeded clients, see 9.1.6(3) Example of deleting duplicate clients in the manual Setup Guide.

If the installation process has completed on the client side but the execution of the job stops at 70% on the managing server side, it is possible that the user program that is started before or after the installation process has not terminated.

If a remote installation job fails due to error, the cause of the error is displayed as a maintenance code. Check the maintenance code in the Detailed Information dialog box that is displayed from the Job Status window. For details about maintenance codes, see 6.2.3 List of maintenance codes.

Packages that are stored in the managing server are automatically deleted after a certain period of time. If a retention period is not specified during the packaging or job creation, the package is deleted the day after it is installed. Even if a retention period is specified, if the day after the remote installation falls on a holiday or the client has been turned off for several days, the package may be deleted. In this case, a situation arises in which the job exists in the relay system but the associated package does not.

To avoid such a situation, be sure to specify an adequate retention period for the package during the packaging or job creation. Note, however, that excessively long retention periods can waste relay system hard-disk space.

The following maintenance code is displayed:

Maintenance code: 300097030000

Even when the registration of an icon is specified during the installation, in some cases the icon fails to be registered. In this case, it is possible that the Windows Shell (or Explorer under Windows) for the client PC is acting as a program separate from user programs.

In situations in which an English OBDCINST.DLL file with relatively new data is installed, the screen is not correctly displayed and the installation of the ODBC driver fails. In this case, you should delete the OBDCINST.DLL file and then execute the remote installation process.

(a) Presence/absence of a coprocessor not correctly indicated

The presence or absence of a coprocessor is determined according to the type of CPU being used, as follows:

• i486DX, i486DX2, or Pentium
  A coprocessor is installed.
  
• i386SX, i386DX, or i486SX
  A coprocessor is not installed.
  

Add-on coprocessors are not recognized. The correct display of a coprocessor is not guaranteed for Intel-compatible CPUs.

(a) The database area is insufficient

JP1/Software Distribution may not operate normally if there is insufficient database area. When the database area for Embedded RDB is insufficient, use Upgrade database in Database Manager to expand the size of the database area.

(b) A database access error occurs during Inventory Viewer or CSV output utility operations

The work table area may be insufficient. Check if the following message was output to CSVODBC.log or INVODBC.log:

 
KFPA11713-E Unable to expand work file due to insufficient HiRDB file system area path-of-work-table-area/rdsys03

path-of-work-table-area

Path for the work table area specified when the database was created.

When this message is output it means that there is insufficient space for the database work table area. Re-estimate the size of the work table area, and then use Upgrade database in Database Manager to expand the size of the work table area.

If the currently allocated work table area size is the same as the estimated size, use the automatic expansion setting.

(c) RD area of the database is shut down

If an error occurs during access to the database (for example, an error occurs during database maintenance or you are not able to open the Remote Installation Manager screen), check the event log to see if either of the following messages is output:

KFPH00306-E RDAREA "RD area name" held due to error

KFPH00307-E RDAREA "RD area name" HELD(CMD) due to error

If the message is output, the RD area is shut down. Perform the following steps to open the shut-down RD area.

1. If the KFPH00306-E message is output, use the facility Recover the database from backup of Database Manager to restore data from a backup file created by the netmdb_backup.bat.
   

   Restriction:

   You can perform this step only if you have backed up data.

2. Start the command prompt.
3. Execute the pdntcmd.bat command that is stored in JP1/Software-Distribution-Manager-installation-directory\BIN.
4. Execute the following command.
   
   • To resolve the KFPH00306-E error:
     Execute pdrels -r ALL -o
     
   • To resolve the KFPH00307-E error:
     Execute pdrels -r RD area name .
     
   When the command succeeds, the message KFPH00110-I pdrels command completed.is displayed and the command prompt becomes active again.
   

(a) Screen drawing and job creation by Remote Installation Manager suddenly slows down

If a job was deleted before the slowdown, it is possible that the deletion of the job is being executed in the background, in which case you should select the Specify when jobs will be deleted check box in the JP1/Software Distribution Manager setup, and specify a time that does not interfere with normal operations. However, jobs waiting for execution are immediately deleted upon the specification of job deletion, irrespective of whether a delayed job deletion is specified.

It is also possible that an applications program is running in the background.

(b) JP1/Software Distribution server and Remote Installation Manager hang

It is possible that a database is damaged, in which case you should execute the following DBCC statement.

To check the condition of the database:

1. From the Microsoft SQL Server program group, start the Query analyzer.
2. Use a DBCC statement to check the integrity of the database:
   

   DBCC CHECKDB (JP1/Software Distribution database-name)

If an error message is output, it is an indication that the database is damaged, in which case you should restore the database from its backup or notify the system administrator.

(c) A management database access error message is displayed during the operation of Remote Installation Manager

Check to see whether the following message is found in the RDBSRV.LOG file for JP1/Software Distribution:

1105,17: 'segment-name' segment full; area for object 'object-name' for database 'database-name' cannot be allocated. If the area in Syslogs is exhausted, dump the transaction log. Otherwise, use either ALTER DATABASE or sp_extendsegment to increase segment size.

If this message is displayed, it is an indication that the transaction log is full, in which case you should dump the transaction log. The log can be deleted as follows:

• Microsoft SQL Server 2012, Microsoft SQL Server 2008 and Microsoft SQL Server 2005
  From Management Studio, right-click the database. Then, from the displayed menu, choose Task and then Backup to display the Database Backup dialog box. From the General page of the Database Backup dialog box, obtain a backup of the transaction log. The transaction log for the portion that was backed up is truncated.
  
• Microsoft SQL Server 2000
  From the Database backup menu for Enterprise Manager, create a backup for the transaction log. The transaction log equal in size to the backup that is created is truncated.
  
• Microsoft SQL Server 7.0
  From the Truncate log menu for Enterprise Manager, truncate the transaction log. Alternatively, code a DUMP statement (DUMP TRANSACTION database-name WITH NO_LOG) in the Query analyzer to delete the transaction log.
  

If there is no need to periodically create a backup for the transaction log, the truncation of the log can also be automatically executed. For details about how to set up automatic execution, see 7.3.2(7) Setting the transaction log in the manual Setup Guide.

(d) Errors occur only when large software is packaged using Packager

It is possible that the database has run out of space. Review the database size; increase the device size as necessary.

(a) If the maximum number of processes connected to the database server is exceeded

If the maximum number of processes connected to the database server is exceeded, JP1/Software Distribution outputs the following message to RDBSRV.LOG:

maximum number of processes xxx exceeded

In xxx, the current number of processes is output.

If this message is output, change the value of processes in the initialization parameter file (INIT.ORA) to increase the maximum number of connected users. You must restart the Oracle service to apply the change.