Question

我有一个日常进程，可以将数据库从一台服务器恢复到另一台服务器。它恢复了数据库的数据库备份文件，如果找到TLog备份文件，它也会恢复它们。

该过程现在可以使用一段时间，没有任何问题。今天它失败了错误消息：

Message
Executed as user: AOI\sa_SqlAgentSrv_PRD. RegQueryValueEx() returned error 2, 'The system cannot find the file specified.' [SQLSTATE 01000] (Message 22001)  IF EXISTS (select * from sysdatabases where name = 'SiteDbStandby')   ALTER DATABASE SiteDbStandby SET SINGLE_USER WITH ROLLBACK IMMEDIATE  
RESTORE DATABASE [SiteDbStandby] FROM DISK = '\\SqlBackup\Databases\SiteDb.bak' WITH NORECOVERY  , REPLACE , MOVE 'SiteDb_data' to 'G:\MSSQL10.DB2\MSSQL\DATA\SiteDbStandby_SiteDb_data.mdf' , MOVE 'SiteDb_log' to 'J:\MSSQL10.DB2\MSSQL\DATA\SiteDbStandby_SiteDb_log.ndf' [SQLSTATE 01000] (Message 0)  Processed 191216 pages for database 'SiteDbStandby', file 'SiteDb_data' on file 1. [SQLSTATE 01000] (Message 4035)  Processed 4 pages for database 'SiteDbStandby', file 'SiteDb_log' on file 1. [SQLSTATE 01000] (Message 4035)  
RESTORE DATABASE successfully processed 191220 pages in 20.366 seconds (73.352 MB/sec). [SQLSTATE 01000] (Message 3014)  
RESTORE LOG SiteDbStandby FROM DISK = '\\SqlBackup\Logs\SiteDb_20120916_011017.BAK' WITH NORECOVERY 
RESTORE LOG SiteDbStandby FROM DISK = '\\SqlBackup\Logs\SiteDb_20120916_021929.BAK' WITH NORECOVERY 
RESTORE LOG SiteDbStandby FROM DISK = '\\SqlBackup\Logs\SiteDb_20120916_031349.BAK' WITH NORECOVERY 
RESTORE LOG SiteDbStandby FROM DISK = '\\SqlBackup\Logs\SiteDb_20120915_231017.BAK' WITH NORECOVERY 
RESTORE LOG SiteDbStandby FROM DISK = '\\SqlBackup\Logs\SiteDb_20120916_001017.BAK' WITH NORECOVERY [SQLSTATE 01000] (Message 0)  The log in this backup set begins at LSN 42459000001303500001, which is too recent to apply to the database. An earlier log backup that includes LSN 42459000001302800001 can be restored. [SQLSTATE 42000] (Error 4305)  RESTORE LOG is terminating abnormally. [SQLSTATE 42000] (Error 3013).  The step failed.

你可以看到它试图从1AM恢复16.9的日志文件，从2AM恢复16.9的日志文件，从3AM恢复16.9的日志文件，然后是从23PM开始的15.9的日志文件，以及从12AM开始的16.9的日志文件。

日志的恢复顺序搞砸了。

我回顾了这份工作的历史，我们有大约3周的日常执行历史。在每个执行中，序列是正确的。例如：

Executed as user: AOI\sa_SqlAgentSrv_PRD. RegQueryValueEx() returned error 2, 'The system cannot find the file specified.' [SQLSTATE 01000] (Message 22001)  IF EXISTS (select * from sysdatabases where name = 'SiteDbStandby')   ALTER DATABASE SiteDbStandby SET SINGLE_USER WITH ROLLBACK IMMEDIATE  
RESTORE DATABASE [SiteDbStandby] FROM DISK = '\\SqlBackup\Databases\SiteDb.bak' WITH NORECOVERY  , REPLACE , MOVE 'SiteDb_data' to 'G:\MSSQL10.DB2\MSSQL\DATA\SiteDbStandby_SiteDb_data.mdf' , MOVE 'SiteDb_log' to 'J:\MSSQL10.DB2\MSSQL\DATA\SiteDbStandby_SiteDb_log.ndf' [SQLSTATE 01000] (Message 0)  Processed 191032 pages for database 'SiteDbStandby', file 'SiteDb_data' on file 1. [SQLSTATE 01000] (Message 4035)  Processed 6 pages for database 'SiteDbStandby', file 'SiteDb_log' on file 1. [SQLSTATE 01000] (Message 4035)  
RESTORE DATABASE successfully processed 191038 pages in 21.703 seconds (68.768 MB/sec). [SQLSTATE 01000] (Message 3014)  
RESTORE LOG SiteDbStandby FROM DISK = '\\SqlBackup\Logs\SiteDb_20120914_231026.BAK' WITH NORECOVERY 
RESTORE LOG SiteDbStandby FROM DISK = '\\SqlBackup\Logs\SiteDb_20120915_001019.BAK' WITH NORECOVERY 
RESTORE LOG SiteDbStandby FROM DISK = '\\SqlBackup\Logs\SiteDb_20120915_011032.BAK' WITH NORECOVERY 
RESTORE LOG SiteDbStandby FROM DISK = '\\SqlBackup\Logs\SiteDb_20120915_021140.BAK' WITH NORECOVERY 
RESTORE LOG SiteDbStandby FROM DISK = '\\SqlBackup\Logs\SiteDb_20120915_031053.BAK' WITH NORECOVERY [SQLSTATE 01000] (Message 0)  Processed 0 pages for database 'SiteDbStandby', file 'SiteDb_data' on file 1. [SQLSTATE 01000] (Message 4035)  Processed 7 pages for database 'SiteDbStandby', file 'SiteDb_log' on file 1. [SQLSTATE 01000] (Message 4035)  
RESTORE LOG successfully processed 7 pages in 0.009 seconds (5.588 MB/sec). [SQLSTATE 01000] (Message 3014)  Processed 0 pages for database 'SiteDbStandby', file 'SiteDb_data' on file 1. [SQLSTATE 01000] (Message 4035)  Processed 0 pages for database 'SiteDbStandby', file 'SiteDb_log' on file 1. [SQLSTATE 01000] (Message 4035)  
RESTORE LOG successfully processed 0 pages in 0.009 seconds (0.000 MB/sec). [SQLSTATE 01000] (Message 3014)  Processed 0 pages for database 'SiteDbStandby', file 'SiteDb_data' on file 1. [SQLSTATE 01000] (Message 4035)  Processed 1229 pages for database 'SiteDbStandby', file 'SiteDb_log' on file 1. [SQLSTATE 01000] (Message 4035)  
RESTORE LOG successfully processed 1229 pages in 0.196 seconds (48.965 MB/sec). [SQLSTATE 01000] (Message 3014)  Processed 0 pages for database 'SiteDbStandby', file 'SiteDb_data' on file 1. [SQLSTATE 01000] (Message 4035)  Processed 3 pages for database 'SiteDbStandby', file 'SiteDb_log' on file 1. [SQLSTATE 01000] (Message 4035)  
RESTORE LOG successfully processed 3 pages in 0.013 seconds (1.277 MB/sec). [SQLSTATE 01000] (Message 3014)  Processed 0 pages for database 'SiteDbStandby', file 'SiteDb_data' on file 1. [SQLSTATE 01000] (Message 4035)  Processed 0 pages for database 'SiteDbStandby', file 'SiteDb_log' on file 1. [SQLSTATE 01000] (Message 4035)  
RESTORE LOG successfully processed 0 pages in 0.009 seconds (0.000 MB/sec). [SQLSTATE 01000] (Message 3014)  
use master restore database SiteDbStandby with recovery [SQLSTATE 01000] (Message 0)  
RESTORE DATABASE successfully processed 0 pages in 4.360 seconds (0.000 MB/sec). [SQLSTATE 01000] (Message 3014)  use SiteDbStandby ALTER AUTHORIZATION ON DATABASE::SiteDbStandby TO sa ALTER DATABASE SiteDbStandby SET COMPATIBILITY_LEVEL = 100 [SQLSTATE 01000] (Message 0).  The step succeeded.

所以，正确的顺序是前一天晚上11点，当天上午12点，当天凌晨1点，凌晨2点和凌晨3点。

所有日志都存在于其共享文件夹目录中。一切看起来都很正常。

你是否碰巧想到了为什么这份工作失败了？

提前致谢，

罗尼。

Answer 1

我们在RESTORE LOG动态查询中添加了ORDER BY子句。

这确保将使用正确的恢复日志顺序。

数据库和日志恢复作业失败

1 个答案: