See the KFPS01220-E message that is output to syslogfile or the message log file to identify the back-end server where the system log file shortage occurred.

Example

 
KFPS01220-E Request to swap sys(bes1) log file unable to be executed
            because there is no standby log file group available.
 

The underlined text shows the back-end server where the system log file shortage occurred. In this example, the system log file shortage occurred at the back-end server bes1.

Determine whether release of checking of system log file unload status is being used at the back-end server (bes1). If N is specified for the pd_log_unload_check operand, release of checking of system log file unload status is being used.

If release of checking of system log file unload status is being used, you need to add a system log file, and then restart the unit. Instead of proceeding to the next step, follow the procedure explained in 20.17.2 Unit-restarting procedure (restarting the unit after adding a system log file).

If release of checking of system log file unload status is not being used, proceed to the next step.

Execute the pdlogls command to check whether there is a system log file that is waiting to be unloaded. Specify the back-end server name determined in step (1) for the -s option of the pdlogls command.

Example

The Status column shows whether a system log file is waiting to be unloaded. System log files for which u or a is shown are waiting to be unloaded. In this example, log002 to log005 are waiting to be unloaded.

If there is a system log file that is waiting to be unloaded, proceed to the next step.

If there is no system log file that is waiting to be unloaded, you need to add a system log file, and then restart the unit. Instead of proceeding to the next step, follow the procedure explained in 20.17.2 Unit-restarting procedure (restarting the unit after adding a system log file).

The action to take differs depending on whether the automatic log unloading facility is being used. If the pd_log_auto_unload_path operand is specified. the automatic log unloading facility is being used.

• If the automatic log unloading facility is being used
  The automatic log unloading facility might have stopped, and as a result, there is a possibility that the system log files were not unloaded. If the automatic log unloading facility stops, the KFPS01150-E message is output. Check whether the KFPS01150-E message has been output to syslogfile or the message log file.
  If the KFPS01150-E message has been output, take the action described in this message. Also see 3.8.5 Error handling and take the necessary action.
  

  Reference note

  When the cause of abnormal termination of the automatic log unloading facility is corrected and the unit is restarted, the facility runs during the unit restart, unloading all system log files that were waiting to be unloaded.

  If the KFPS01150-E message is not output (the automatic log unloading facility is not stopped), you must add a system log file and, then restart the unit. Instead of proceeding to the next step, follow the procedure explained in 20.17.2 Unit-restarting procedure (restarting the unit after adding a system log file).
  
• If the automatic log unloading facility is not being used
  Unload the system log files that are waiting to be unloaded. For details about the procedure, see 20.17.5 Changing a system log file to unloaded status.
  If you cannot unload any of the system log files that are waiting to be unloaded, you must add a system log file, and then restart the unit. Instead of proceeding to the next step, follow the procedure explained in 20.17.2 Unit-restarting procedure (restarting the unit after adding a system log file).
  

Execute the pdstart command to restart the unit.

After the unit restarts, transactions can be accepted. However, the system logs that are output by processing of new transactions might cause a system log file shortage again. Therefore, resume the job after the operation in step (7) is completed.

When the synchronization point validation process has completed, the KFPS02183-I message is output to syslogfile or the message log file.

Example

 
KFPS02183-I HRDB unt1 Syncpoint dump for bes1 has been acquired to file group
            spd001. log file information:log001, 7, 287. start time=12:45:00,
            end time=12:45:12
 

Confirm that the underlined server name is the server name identified in step (1). In this example, synchronization point dump validation processing is complete at the back-end server bes1.

If N is specified for the server definition of the server where the system log file shortage occurred or for the pd_spd_assurance_msg operand of the server common definition (corresponding to the back-end server definition for bes1 in this example), the KFPS02183-I message is not output. In this case, see the execution result of the pdlogls command to check whether synchronization point dump validation processing is complete.

Example

• Execution result of the pdlogls command executed immediately after unit restart
  

  

  System log files that have d in the Status column are overwrite disabled. In this example, log001 and log006 cannot be overwritten.
  
• Execution results of the pdlogls command executed after synchronization point dump validation processing is complete
  

  

  When synchronization point dump validation processing finishes, the system log files are released from overwrite-disabled status. The value in the Status column changes from d to -. In this example, the system log files log001 and log006 are released from overwrite-disabled status.
  

Database recovery processing executed during unit restart also outputs system logs. If merely unloading the system log files cannot secure a system log file size necessary for a restart, the unit will terminate abnormally again before synchronization point dump validation processing finishes. In this case, instead of proceeding to the next step, follow the procedure explained in 20.17.2 Unit-restarting procedure (restarting the unit after adding a system log file).

Take the following steps before resuming the job:

• Check whether there is a sufficient number of swappable system log files.
  Execute the pdlogls command to determine the number of swappable system log files. Make sure at least 2/3 of the total files are available. If the number is not sufficient, increase the number of swappable system log files by taking an action such as unloading system log files.
  
• Check whether the automatic log unloading facility is running.
  If the automatic log unloading facility is being used, execute the pdlogatul -d sys -i command to check whether this facility is running.
  If the automatic log unloading facility is stopped, check the KFPS01150-E message, which is output after unit restart, and take the action prescribed in the message. Then, execute the pdlogatul -d sys -b command to restart the automatic log unloading facility.
  
• Acquire backup information and log point information.
  If operation without unloading system log is being used, use the pdcopy command to back up all RDAREAs within the server. When doing so, specify the -z option to collect the log point information file.
  After making a backup, execute the pdlogchg -z command to change the system log file status to unloaded status.
  

After you have eliminated the system log file shortage, take the actions described in 20.17.10 Actions to take after eliminating the system log file shortage.