Nonstop Database, HiRDB Version 9 System Operation Guide

[Contents][Index][Back][Next]

3.9.6 Output of status information file for system log files (function for monitoring the free area)

This section describes how to use the status information file for system log files to identify the cause of an area insufficiency for the system log files.

Organization of this subsection
(1) Content of status information file for system log files
(2) How to identify the cause of an area insufficiency for the system log files

(1) Content of status information file for system log files

When the percentage of free area for the system log files falls below the warning value, a status information file for system log files is output. This file acquires the status of the system log files, synchronization points, and transactions. The status information file for system log files is created in the $PDDIR/spool/pdjnlinf directory. The file name is as shown below:

pdsnap.server-name.time-issued

Remarks
  • If acquisition of the date or time information fails, the time issued becomes 999999999999.
  • If a file with the same name already exists, that file is overwritten.
  • If you change the file name, it might not be possible to use the pdcspool command to delete the file.

Output contents of status information file for system log files
The following example lists the output contents of a status information file for system log files. In this example, the percentage of free area at back-end server bes1 is below the warning value.
pdsnap.bes1.030415124350    <- File name#1
Date:2003/04/15             <- Date issued#1
Time:12:43:50               <- Time issued#1
pdlogls -d sys -e -s bes1   <- Command issued internally by HiRDB
 Execution result of the pdlogls command
      :
 
pdlogls -d spd -e -s bes1
 Execution result of the pdlogls command
      :
 
pdls -d trn -a -s bes1
 Execution result of the pdls command
      :
 
pdls -d svr
 Execution result of the pdls command#2
      :
 

#1: If acquisition of the date or time failed, 99...99 is displayed. Change the file name as necessary.

#2: In the case of a HiRDB parallel server configuration, only information for units at the applicable server is displayed.

(2) How to identify the cause of an area insufficiency for the system log files

Use the status information file for system log files to identify the cause of the area insufficiency for the system log files. The method of identifying the cause is described below.

(a) Identify the status information file for the system log files

Identify the status information file for the system log files from the KFPS01160-E or KFPS01162-W message.

Example: When the KFPS01160-E message is output
KFPS01160-E Insufficient system log space. Transaction service stopped. Transactions terminate by force, server = bes1, output file name = pdsnap.bes1.030401084500
The following is the name of the status information file for the system log files in this case:
$PDDIR/spool/pdjnlinf/pdsnap.bes1.030401084500
(b) Acquire the status of the system log files

Refer to the contents of the status information file for the system log files and check the status of the system log files. The STATUS entry in the output results of the pdlogls -d sys command shows the status of the system log files.

(c) Causes of area insufficiency for system log files

The cause can be inferred from the status of the system log files. The following table shows the causes of insufficient free area for system log files.

Table 3-12 Causes of free area insufficiencies for system log files

Status of system log files Causes of insufficient free area for system log files
Many files are in unload wait status System log files might not have been unloaded. You must unload system log files regularly.
Many files are in overwrite disabled status One of the following causes might be applicable:
  • When numerous updates are performed within a single transaction, the update log size increases and the percentage of free area for system log files decreases. In such cases, split up such transactions into multiple transactions. If this is not possible, increase the available area for system log files.
  • If a transaction is not settled for a long period of time, no synchronization point will be acquired. During such a period, the update log size increases and the percentage of free area for system log files decreases. If there are such transactions, either revise the server's maximum wait time (value of the PDSWAITTIME operand in the client environment definition or value of the pd_watch_pc_client_time operand) or use the skipped effective synchronization point dump monitoring facility.
Many files are in extracting status Extraction processing might not have been applied to the system log files. Either revise the design of the system log files or revise the operation method for data linkage that uses the HiRDB Datareplicator.
Many files are in overwriting denied status for online reorganization Update processing of updatable online reorganization might not have been applied. Either revise the design of the system log files or revise the method of operating updatable online reorganization of the inner replica facility.