sybase启动报错error 945 error3414

双色领带

启动最后报错，数据库被挂起，以前解决过926错误被挂起的问题，重新rebuild_log一下就可以了，但是这个报945错误，rebuild_log之后还是无法解决，初步判定是数据页损坏，请朋友们帮忙解决一下

wfcjz

把你的日志贴上来，错误代码不同，也就意味着问题肯定不一样，926基本上是库挂起，

wfcjz

945多半是设备空间不足，你扩下日志和数据的设备空间试试

双色领带

server  Database 'sybsystemprocs' is now online.
00:00000:00001:2014/08/14 19:25:19.54 server  Recovering database 'melinets'.
00:00000:00001:2014/08/14 19:25:21.79 server  [0] dm_segmap = 19, dm_lstart = 0, dm_vstart = 50331648, dm_size = 1024000
00:00000:00001:2014/08/14 19:25:21.79 server  [1] dm_segmap = 11, dm_lstart = 1024000, dm_vstart = 100663296, dm_size = 512000
00:00000:00001:2014/08/14 19:25:21.79 server  [2] dm_segmap = 35, dm_lstart = 1536000, dm_vstart = 67108864, dm_size = 512000
00:00000:00001:2014/08/14 19:25:21.79 server  [3] dm_segmap = 4, dm_lstart = 2048000, dm_vstart = 83886080, dm_size = 512000
00:00000:00001:2014/08/14 19:25:21.79 server  Error: 945, Severity: 21, State: 1
00:00000:00001:2014/08/14 19:25:21.79 server  Unable to locate page 1768292452 for database id 6 in the in-memory copy of Sysusages (spid=1). This is an internal error. Please contact Sybase Technical Support.
00:00000:00001:2014/08/14 19:25:21.80 server  Error: 3414, Severity: 21, State: 1
00:00000:00001:2014/08/14 19:25:21.80 server  Database 'melinets' (dbid 6): Recovery failed. Check the SQL Server errorlog for further information as to the cause.
00:00000:00001:2014/08/14 19:25:21.80 server  Recovery complete.
00:00000:00001:2014/08/14 19:25:21.80 server  SQL Server's default sort order is:
00:00000:00001:2014/08/14 19:25:21.80 server          'bin_iso_1' (ID = 50)
00:00000:00001:2014/08/14 19:25:21.80 server  on top of default character set:
00:00000:00001:2014/08/14 19:25:21.80 server          'iso_1' (ID = 1).
00:00000:00001:2014/08/14 19:25:21.80 server  Loaded default Unilib conversion handle.
00:00000:00000:2014/08/15 12:41:33.46 kernel  ueshutdown: exiting
00:00000:00000:2014/08/15 12:41:33.46 kernel  Abnormal exit detected - SQL Server process level execution bindings will be cleared on startup.

双色领带

只把报错部分贴出来了，之前的部分都没有报错，物理空间是足够的，要是数据段也没有满

wfcjz

1. (1)赋予sa用户sybase_ts_role的角色
   
2. isql -Usa -P
   
3. 1>sp_role "grant","sybase_ts_role",sa
   
4. 2>go
   
5. 1>quit
   
6. 
   
7. (2)将数据库置为"bypass recovery"状态
   
8. isql -Usa -P
   
9. 1>sp_configure "allow updates",1
   
10. 2>go
    
11. 1>use master
    
12. 2>go
    
13. 1>update sysdatabases set status=-32768
    
14. 2>where name="database_name"
    
15. 3>go
    
16. 1>shutdown with nowait
    
17. 2>go
    
18. 
    
19. (3)rebuild数据库日志
    
20. 重启Server
    
21. isql -Usa -P
    
22. 1>use master
    
23. 2>go
    
24. 1>dbcc rebuild_log(database_name,1,1)
    
25. 2>go
    
26. 1>shutdown with nowait
    
27. 2>go
    
28. 
    
29. (4)重启server
    
30. 1>use master
    
31. 2>go
    
32. 1>update sysdatabases set status=0 where name="database_name"
    
33. 2>go
    
34. 1>sp_configure "allow updates",0
    
35. 2>go
    
36. 1>shutdown with nowait
    
37. 2>go
    
38. 
    
39. (5)在重启server之后,如果数据库恢复正常，rebuild log工作将会成功完成，否则要恢复数据库备份，使用dump database或bcp命令。

复制代码

Good luck!

andkylee

原因在于melinets的sysuages分配信息不对。

wfcjz

回复 7# andkylee


   嗯，是否解决了！