First, the description of the problem: based on common data disaster ORACLE database environment
fault performance:
1, ORACLE database can not start or does not work.
2, ORACLE ASM storage damage.
3, ORACLE data files are missing.
4, ORACLE data file is partially damaged.
5, ORACLE DUMP file corruption.
Second, the solution
◆ detector
1, detects whether there is a hardware failure, such as hardware failures, hardware processing switch
2, a read-only detect whether a failure of the user performance described in the same
◆ restore
a backup: read-only memory to make a complete failure mirror (see Appendix)
2 for data analysis in the backup and restore operations.
3, generally, the recovered data will be temporarily stored in the other bank
◆ acceptance
of good recovery data can be verified for correctness. If confirmed, payment -> be transferred to the media and recovered data -> Billing (receipts) and reports.
If you can not recognize the results of data recovery, return the original media, no service charge, free issue a report.
Third, the data recovery possibilities
★ ORACLE database can not start or does not work:
If this fault occurs sudden, often high recoverability. From a technical point of view the bottom, if the table is not damaged SYSTEM, data recovery easier; SYSTEM be time-consuming if the table is corrupt, the data need to manually check table structure, the recovery time.
★ ORACLE ASM storage destruction:
The ASM reset, or the composition of some members of ASM faulty device, after an error without substantial new data is written, data is often preferable to restore.
★ ORACLE data files are missing:
no matter ORACLE data files are deleted, formatted or lost for unknown reasons, as long as no new data is written, no matter what operating system, data can be recovered by the rules of the internal organization of ORACLE data files, but name of the data file may require manual checking.
★ ORACLE data file is partially damaged:
The ORACLE data file portion is damaged (e.g., cover), extraction and recombination complex data, the data recording portion may generally be recovered undamaged, can be added into the new table, but is time-consuming .
★ ORACLE DUMP file corruption:
the ORACLE the DUMP file is damaged, the damaged portion is removed, the rest can be added to normal data table.
Fourth, the time
1TB less storage space (not to recover the data capacity), usually two working days to be completed; increased with more than 1TB storage capacity, generally will increase the recovery period.
If large data tables, data extraction, data collation will spend a lot of time, it needs specific time, according to specific circumstances.
[Tips]
★ against software failure, data loss, should be reduced as much as possible the operation of the store, sometimes even driving machine, doing nothing, it could lead to disaster further intensified. Conditions permitting, the best after damage, do a full backup to disk or storage volumes
★ against hardware failure, after the device is not working properly, you should add as little power to avoid further damage to the equipment.
How to avoid
make a backup program, as far as possible to avoid a single storage backup, such data is very important, consider off-site backup.