Nonstop Database, HiRDB Version 9 System Operation Guide

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

3.10.4 Actions to be taken by the HiRDB administrator when system log files cannot be automatically extended

The HiRDB administrator must check the status of system log files and take appropriate action according to the cause of each problem.

Organization of this subsection
(1) Identifying system log files that can no longer be automatically extended
(2) Identifying the problem cause
(3) Modify the system log file(s) according to the identified cause

(1) Identifying system log files that can no longer be automatically extended

Refer to message KFPS01390-I to identify system log files that can no longer be automatically extended. Message KFPS01390-I shows the server name, file group name, and system type in the case of a duplexed system in which a system log file that can no longer be automatically extended is located.

(2) Identifying the problem cause

Refer to the reason code shown in message KFPS01390-I and identify the cause of the problem that is preventing automatic extension. Also, execute the pdlogls -d sys -e -E command on the system log file of the applicable server to check the status of the system log file. By executing the pdlogls -d sys -e -E command, you can check the following points:

(3) Modify the system log file(s) according to the identified cause

Take corrective action according to the cause that is preventing automatic extension.

If the disk containing the HiRDB file system area in which the system log files were created has become full
If automatic extension cannot be carried out because the disk containing the HiRDB file system area in which the system log files are created has become full (reason code 205 in message KFPS01390-I), take the following steps:
  1. Use an OS command to check whether the disk on which the applicable system log file was created has become full.
  2. Take an action such as deleting unnecessary files to increase available disk space.

If the capacity of a system log file has reached the upper limit for system log files
If automatic extension cannot be carried out because a system log file has reached its upper limit (reason code 1692 in message KFPS01390-I), take the following steps:
  1. Use the pdfls command to check whether the applicable system log file has reached the upper limit for system log files (100 GB).
  2. Reassess the total capacity and the number of generations of system log files. For details about the total capacity and the number of generations of system log files, see Designing system log files in the HiRDB Version 9 Installation and Design Guide.
  3. Add a system log file group.

If the capacity of a system log file has reached the extension upper limit specified by the pd_log_auto_expand_size operand
If automatic extension cannot be carried out because the extension upper limit specified by the pd_log_auto_expand_size operand has been reached (reason code 1694 in message KFPS01390-I), take the following steps:
  1. Use the pdfls command to check whether the applicable system log file has reached the extension upper limit specified by the pd_log_auto_expand_size operand.
  2. Modify the extension upper limit specified by the pd_log_auto_expand_size operand, or omit specification of this operand.
    To modify the HiRDB system definition, use one of the following methods:
    [Figure] Use the pdchgconf command.
    [Figure] Normally terminate HiRDB, modify the HiRDB system definition, and then restart HiRDB.

If an input/output error has caused system log file extension to fail
If automatic extension cannot be carried out because an input/output error has caused system log file extension to fail (reason code 209 in message KFPS01390-I), take the following steps:
  1. Use the pdlogls -d sys command to check whether an error shutdown occurred in the applicable system log file.
  2. Refer to the messages that were issued before and after message KFPS01390-I to identify the cause of the input/output error, and then take the necessary corrective action.
  3. To make a system log file extendable again, re-create the applicable system log file.
    Reference note
    The procedure for re-creating a system log file is as follows:
    1. Use the pdlogcls command to close the applicable system log file.
    2. Use the pdlogrm command to delete the applicable system log file.
    3. Use the pdloginit command to re-create a system log file.
    4. Use the pdlogopen command to open the system log file.

If a memory shortage has caused system log file extension to fail
If automatic extension cannot be carried out because a memory shortage has caused system log file extension to fail (reason code 101 in message KFPS01390-I), take the following steps:
  1. Use an OS command to check whether a memory shortage has occurred.
  2. Reassess the memory requirement. If unnecessary programs are running, stop them to reduce memory usage.
  3. To make a system log file extendable again, close the applicable system log file first, and then reopen it.

If a lock segment shortage has caused system log file extension to fail
If automatic extension cannot be carried out because a lock segment shortage has caused system log file extension to fail (reason code 221 in message KFPS01390-I), take the following steps:
  1. Use an OS command to check whether a lock segment shortage has occurred.
  2. Reassess the number of OS lock segments.
    To change the number of lock segments, normally terminate HiRDB, modify the OS parameters, and then restart HiRDB.