If the instance cannot write to a datafile in the SYSTEM tablespace, an undo tablespace (if the database is in automatic undo management mode, which is the preferred choice in Release 10 g), or a datafile in a tablespace containing active rollback segments (if in manual undo management mode), then the database issues an error and shuts down the instance. The effect of a datafile write error depends upon which tablespace the datafile is in. At the next checkpoint, a write error will occur when the database attempts to write the file header as part of the standard checkpoint process. The database continues to run, but the error is returned each time an unsuccessful read occurs. In a read error, the instance cannot read a datafile and an operating system error is returned to the application, along with an error indicating that the file cannot be found, cannot be opened, or cannot be read. Media failures are either read errors or write errors. Damage to a nonmultiplexed online redo log causes database operation to halt and may cause permanent loss of data.ĭamage to any control file, whether it is multiplexed or not, halts database operation when the database attempts to read or write to the damaged control file (which happens frequently, for example at every checkpoint and log switch). If a media failure damages a disk containing one copy of a multiplexed online redo log, then the database can usually continue to operate without significant interruption. Multiplexed files should be stored on separate disks. When an online redo log or control file is multiplexed, multiple copies of the file are maintained on the system. Database operation after a media failure of online redo log files or control files depends on whether the online redo log or control file is protected by multiplexing, as recommended.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |