怎样修复INTERBASE数据库？

[lovecat_sc]

现有一ITERBASE数据库，查询其中一表时出现如下错误：
General SQL error.
database file appears corrupt()
bad checksum
checksum error on database page 5284
在DELPHI的SQL EXPLORER中能看到该表数据，但到2000多行后就不行了，又出现民上错误。请问有无方法修复？

[jsg_jn]

备份恢复一下

[lqflg]

运行IBConsole中菜单:
database->maintenance->validation

[lqflg]

运行IBConsole中菜单:
database->maintenance->validation就可以修复

[erickleung]

假如这不是突然停电, 而数据库却是在windows server上, 加上把 forecd write 关掉. 一般情形下可以用 interbase上的
gfix 及 gbak 修好.

首先把数据文件备份, 不要在生产文件直接修复!!
$ cp atlas.gdb atlas_test.gdb (Linux)
copy atlas.gdb atlas_test.gdb (windows)

$ gfix -v -f atlas_test.gdb
Mend database corruption. Only do this if the previous step reported database corruption.

$ gfix -m -i atlas_test.gdb
Check to see if corruption was mended.

$ gfix -v -f atlas_test.gdb
If gfix still reports database corruption, we need to take further steps to repair the database.

若 gfix无效, 可以再用 gbak 作备份, 然後修复
The backup/restore procedure will create a new database with no old record versions, no page fragmentation, and newly-formed indices and transaction log.


Back up database.
备份:
$ gbak -g -b -z -v atlas_test.gdb atlas.gbk

If there is corruption in record versions of a limbo transaction, then you may need to include the -l switch to the above gbak command to ignore limbo transactions.
Create new database from backup. This should create a pristine database, with newly-made indices, no old record versions, etc.

$ gbak -c -v -z atlas.gbk atlas_new.gdb
Check to see if new database is corrupt.

$ gfix -v -f atlas_new.gdb