If system log files are duplexed and if either of the system log files can no longer be automatically extended, the applicable file group is placed in unextendable status and is excluded from extension. If this occurs, the size of the system log file that can be used by the applicable file group is the smaller of the two sizes, and the differential area is kept as an area that cannot be used for system log files.

When single operation of system log files is applied, even if an error occurs in a system log file, HiRDB continues to automatically extend the normal system log file. However, if that system log file goes into a state in which it cannot be automatically extended, HiRDB does not recognize that an error has occurred in a system log file. Therefore, in this case, single operation is not carried out. After a file group that can no longer be automatically extended is placed in unextendable status, if an error occurs in a system log file, single operation is carried out. However, if the error has occurred in a system log file that cannot be extended, the file group is returned to extendable status and the normal system log file in which no error has occurred is automatically extended. When single operation of system log files is continued in this way, the file size of the system log file in which the error occurred becomes different from the size of the normal system log file. Therefore, after the cause of the error has been eliminated, when you recover the system log file in which the error occurred, we recommend that you re-create the file in which the error occurred so that it is larger than the normal file.

If automatic opening of system log files is used and if there is no swappable standby file when the HiRDB system (or unit) is restarted, HiRDB opens a reserved file if one is available, making it a swappable target, and then continues processing. However, if no reserved file is available or if all opened files are full, HiRDB extends the current file.

If the automatic log unloading facility for system log files is used and if the system log file is in one of the following statuses when the HiRDB system (or unit) is restarted, HiRDB waits until one generation of logs is finished unloading:

• There is no swappable standby file or reserved file.
• There is a system log file waiting to be unloaded.

When unloading is finished, the system log file becomes current. If that file becomes full and there is no swappable standby file, the current file is extended.