服务器掉电引起用户数据库出现605错误！如何修复？

jackyy

服务器掉电引起用户数据库出现605错误

solaris 9 ,sybase11.9.2
阵列掉电，导致user_data库出现605错误，SYBASE.log日志如下：

00:00000:00001:2009/10/28 17:17:36.53 server  Recovering database 'user_data'.
00:00000:00012:2009/10/28 17:17:37.03 server  Error: 1608, Severity: 18, State: 4
00:00000:00012:2009/10/28 17:17:37.03 server  A client process exited abnormally, or a network error was encountered. Unless other e
rrors occurred, continue processing normally.
00:00000:00011:2009/10/28 17:17:37.03 server  Error: 1608, Severity: 18, State: 4
00:00000:00011:2009/10/28 17:17:37.03 server  A client process exited abnormally, or a network error was encountered. Unless other e
rrors occurred, continue processing normally.
00:00000:00007:2009/10/28 17:17:37.03 server  Error: 1608, Severity: 18, State: 4
00:00000:00007:2009/10/28 17:17:37.03 server  A client process exited abnormally, or a network error was encountered. Unless other e
rrors occurred, continue processing normally.
00:00000:00017:2009/10/28 17:17:37.03 server  Error: 1608, Severity: 18, State: 4
00:00000:00017:2009/10/28 17:17:37.03 server  A client process exited abnormally, or a network error was encountered. Unless other e
rrors occurred, continue processing normally.
00:00000:00009:2009/10/28 17:17:37.03 server  Error: 1608, Severity: 18, State: 4
00:00000:00009:2009/10/28 17:17:37.03 server  A client process exited abnormally, or a network error was encountered. Unless other e
rrors occurred, continue processing normally.
00:00000:00010:2009/10/28 17:17:37.03 server  Error: 1608, Severity: 18, State: 4
00:00000:00010:2009/10/28 17:17:37.03 server  A client process exited abnormally, or a network error was encountered. Unless other e
rrors occurred, continue processing normally.
00:00000:00001:2009/10/28 17:21:06.92 server  The transaction log in the database 'user_data' will use I/O size of 2 Kb.
00:00000:00001:2009/10/28 17:21:06.94 server  Database 'user_data' is now online.
00:00000:00001:2009/10/28 17:21:06.94 server  Recovery complete.
00:00000:00001:2009/10/28 17:21:06.94 server  SQL Server's default sort order is:
00:00000:00001:2009/10/28 17:21:06.94 server    'bin_iso_1' (ID = 50)
00:00000:00001:2009/10/28 17:21:06.94 server  on top of default character set:
00:00000:00001:2009/10/28 17:21:06.94 server    'iso_1' (ID = 1).
00:00000:00001:2009/10/28 17:21:06.94 server  Loaded default Unilib conversion handle.
00:00000:00019:2009/10/28 17:21:12.14 server  Starting diagnostics for read failure:
00:00000:00019:2009/10/28 17:21:12.14 server  Device Information:
                                 Device number = 13
                                 Phyname = /cwums/sybdata/user_data2.dat
00:00000:00019:2009/10/28 17:21:12.14 server  Buffer Information:
                                 Buf addr = 0xfea3df74, Mass addr = 0xfea3df74
                                 Buf pageno = 2549312, Mass pageno = 2549312, dbid = 7
00:00000:00019:2009/10/28 17:21:12.14 server  
                                 Buf virtpg = 218605120, Mass virtpg = 218605120
                                 Buf stat = 0x1, Mass stat = 0x1008
                                 Mass keep = 1, Mass awaited = 0
00:00000:00019:2009/10/28 17:21:12.14 server  Page Information from first read attempt:
                                 Page read from disk
                                 ppageno = 2549312, pobjid = 1964534032, pindid = 0
                                 pnextpg = 2549313, pprevpg = 2549167
                                 plevel = 0, pstat = 0x101
                                 pts_hi = 16385, pts_lo = 1513049149
00:00000:00019:2009/10/28 17:21:12.14 server  Page Information from second read attempt:
                                 Page read from disk
                                 ppageno = 2549312, pobjid = 1964534032, pindid = 0
                                 pnextpg = 2549313, pprevpg = 2549167
                                 plevel = 0, pstat = 0x101
                                 pts_hi = 16385, pts_lo = 1513049149
00:00000:00019:2009/10/28 17:21:12.14 server  SDES Information:
                                 dbid = 7, objid = 1452532208, scur.pageid = 2549312
                                 sstat = 0x2100041, sstat2 = 0x20
                                 suid = 1, cacheid = 0
00:00000:00019:2009/10/28 17:21:12.14 server  PSS Information:
                                 pstat = 0x10000, pcurdb = 7, pspid = 19
                                 p2stat = 0x1000, p3stat = 0x81a
                                 plasterror = 0, preverror = 0, pattention = 0
00:00000:00019:2009/10/28 17:21:12.14 server  End diagnostics for read failure:
00:00000:00019:2009/10/28 17:21:12.14 server  Run the DBCC command 'dbcc tablealloc' for both the objects 1452532208, 1964534032
00:00000:00019:2009/10/28 17:21:12.14 server  Error: 605, Severity: 21, State: 1
00:00000:00019:2009/10/28 17:21:12.14 server  An attempt was made to fetch logical page '2549312' in database 'user_data' from cach
e 'default data cache'. Page belongs to object with id '1964534032', not to object 'DIAL_DETAIL_110'.
00:00000:00019:2009/10/28 17:21:12.14 kernel  ************************************
00:00000:00019:2009/10/28 17:21:12.14 kernel  SQL causing error : select C_USER_ID,C_CALLING_NUMBER,FRAME_IP,convert(char(10),D_STAR
T_TIME,102)+' '+convert(char(,D_START_TIME,10,convert(char(10),D_STOP_TIME,102)+
00:00000:00019:2009/10/28 17:21:12.14 kernel  ************************************
00:00000:00019:2009/10/28 17:21:12.14 server  SQL Text: select C_USER_ID,C_CALLING_NUMBER,FRAME_IP,convert(char(10),D_START_TIME,102
)+' '+convert(char(,D_START_TIME,10,convert(char(10),D_STOP_TIME,102)+' '+convert(char(,D_STOP_TIME,10,I_CONNECT_DUR,I_NORMA
L_TIME,N_SPECIAL_SECONDS from DIAL_DETAIL_110 where C_USER_ID='lyq' order by D_START_TIME
00:00000:00019:2009/10/28 17:21:12.14 kernel  curdb = 7 pstat = 0x10000 lasterror = 605
00:00000:00019:2009/10/28 17:21:12.14 kernel  preverror = 0 transtate = 1
00:00000:00019:2009/10/28 17:21:12.14 kernel  curcmd = 193 program =                              
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x48348c pcstkwalk+0x20(0xfc5dbfd8, 0x0, 0x270f, 0x2, 0x0)
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x4833a0 ucstkgentrace+0x178(0x9c009c, 0x2, 0x270f, 0x0, 0x0)
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x461e20 ucbacktrace+0xa4(0xfc885920, 0x1, 0x9c009c, 0xfda81e90, 0x0)
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x772f0 terminate_process+0xa60(0x0, 0x79b63c, 0x6c00, 0x5400, 0x54c4)
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0xb0074 hdl_default+0x48(0x6, 0x5, 0x15, 0x1, 0xfc5dc32
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x289b70 s_handle+0xda8(0xfc5dc328, 0x6400, 0x15, 0x5360, 0xfd3335c0)
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0xafd40 ex_raise+0x1e0(0x6, 0x0, 0xfd3335c0, 0xfd338adc, 0xfd338ba
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x1a84ac wrongpage+0x4b4(0xfea3df74, 0x0, 0x0, 0x75186910, 0xffffffff)
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x1a6b44 getpage_with_validation+0x1154(0xfccb7c00, 0x6620, 0xfccb7c94, 0x1, 0x26e
640)
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x1a0d90 apl_getnext+0x328(0xfccb7c00, 0x0, 0xfccb7c78, 0xfccb7c94, 0xfccb7ccc)
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x1a063c getnext+0x6c(0xfccb7c00, 0x40000, 0x78c400, 0x78e800, 0x1)
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x31aa84 exec_eop+0x508(0x4, 0xfddfa000, 0x6c00, 0xfccb7c00, 0x4)
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x31acf8 exec_eop+0x77c(0x1, 0xfde12178, 0x6c00, 0x0, 0x1)
00:00000:00019:2009/10/28 17:21:12.14 kernel  [Handler pc: 0x31e5ac execerr installed by the following function:-]
00:00000:00019:2009/10/28 17:21:12.14 kernel  [Handler pc: 0x2b3eac sortmerr installed by the following function:-]
00:00000:00019:2009/10/28 17:21:12.14 kernel  [Handler pc: 0x31e5ac execerr installed by the following function:-]
00:00000:00019:2009/10/28 17:21:12.14 kernel  [Handler pc: 0x31e5ac execerr installed by the following function:-]
00:00000:00019:2009/10/28 17:21:12.14 kernel  [Handler pc: 0x25f300 aritherr installed by the following function:-]
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x319f20 execute+0x9c8(0xfde04ba8, 0x6dec, 0xfd3335c0, 0x5000, 0xfde04da
00:00000:00019:2009/10/28 17:21:12.14 kernel  pc: 0x275d0c s_execute+0x6e0(0xc1, 0x647390, 0xffffffff, 0x0, 0x3)
00:00000:00019:2009/10/28 17:21:12.14 kernel  [Handler pc: 0x288dc8 s_handle installed by the following function:-]
00:00000:00019:2009/10/28 17:21:12.16 kernel  pc: 0x286864 sequencer+0xee0(0xfc00f300, 0x5368, 0xfd3335c0, 0xc, 0x79b400)
00:00000:00019:2009/10/28 17:21:12.16 kernel  [Handler pc: 0xaff20 hdl_backout installed by the following function:-]
00:00000:00019:2009/10/28 17:21:12.16 kernel  [Handler pc: 0x24db58 ut_handle installed by the following function:-]
00:00000:00019:2009/10/28 17:21:12.16 kernel  [Handler pc: 0x24db58 ut_handle installed by the following function:-]
00:00000:00019:2009/10/28 17:21:12.16 kernel  pc: 0x98440 conn_hdlr+0x1b7c(0x6400, 0x1, 0x78e800, 0x79b400, 0x54c0)
00:00000:00019:2009/10/28 17:21:12.16 kernel  pc: 0x493ed8 _coldstart+0x4(0x11, 0x968c4, 0x0, 0x0, 0x0)
00:00000:00019:2009/10/28 17:21:12.16 kernel  end of stack trace, spid 19, kpid 10223772, suid 3
00:00000:00001:2009/10/28 17:22:48.99 server  Starting diagnostics for read failure:

select DIAL_DETAIL_110这张表是，出现如下错误提示，感觉显示的不全，就出错了，再运行别的isql命令，就自动退出isql了。
1> select C_USER_ID from DIAL_DETAIL_110 where C_USER_ID='lyq'
2> go
C_USER_ID        
----------------
lyq              
lyq              
lyq              
lyq              
lyq
Msg 21, Level 21, State 1:
Line 1:
WARNING - Fatal Error 605 occurred at Oct 29 2009  2:31PM.  Please note the
error and time, and contact a user with System Administrator (SA)
authorization.
The SQL Server is terminating this process.
1>


dbcc checkdb过user_data库，错误信息大概如下：

An attempt was made to fetch logical page '2549312' in database 'user_data' from cache 'default data cache'. Page belongs to object with id '1964534032', not to object 'DIAL_DETAIL_110'.
An attempt was made to fetch logical page '2549231' in database 'user_data' from cache 'default data cache'. Page belongs to object with id '1932533918', not to object 'DIAL_DETAIL_111'.  
An attempt was made to fetch logical page '2549239' in database 'user_data' from cache 'default data cache'. Page belongs to object with id '2028534260', not to object 'DIAL_DETAIL_112'.  

按照sybase.log的报告说，我是否可以试一下dbcc tablealloc(DIAL_DETAIL_110,full,fix)，这个命令修复一下，是否有什么危险呢？而且想知道，现在的错误提示都是什么意思？请指教！
另外这条命令的话，是否要在单用户模式下运行？如何把握的sybase切换到单用户模式下？做完dbcc后，又如何恢复到正常模式下呢？
请高手指教！！！！！！！！！！！！！！！！！！！！！！！！！！！！！！！！！！！！！！不胜感激！！！！！！！！！！！！！！！！！！！！！！！！！

hobbylu

重新建表吧

jackyy

阿?没有别的办法了?俺有最近的dump，但是不知道如何恢复呀？看了好多dump/load的资料，还是搞不懂！是直接就能load呢？还是要建设备，初始化设备什么的？教教我呗？实在是搞不定呀！

hobbylu

单用户方式，直接load就可以了

jackyy

请问一下，Phyname = /cwums/sybdata/user_data2.dat，这个user_data2.dat这个文件如果有问题的话（看sybase.log日志，好像是有问题，但判断不清），在单用户下直接load，可以吗？会恢复吗？好担心呀！

WFCJZ

你可以按以下2种方式采取任意一种：

1。用SYBASE CENTEL工具调出坏表的DLL。BCP OUT这个表的数据，然后CREATE TABLE ，在BCP IN

2。直接在单用户模式下LOAD恢复！


怎么感觉楼主用的是浪潮？

jackyy

不是浪潮，是sun.
我这没有图形化界面的，调出坏的DLL是什么意思？还有别的方法调出DLL吗？

[ 本帖最后由 jackyy 于 2009-10-30 00:34 编辑 ]

WFCJZ

打错字了，是把这个建表的DLL调出来，重新建表！

jackyy

严重感谢hobbylu 和 WFCJZ ，真的很感谢！那天那么晚了还给我技术支持！再次感谢！谢谢！！！
dbcc修复无济于事，最后load database了，而且过程曲折，在第一次load的时候，阵列再次掉电！吓了一身冷汗！还好，加电后数据库正常启动了！第二次load后，最后出现如下提示，不知是否对数据库有影响，看了相关文章，好像有的说是什么data和log混在一起了，不知道是什么意思？呵呵，也不知道我的日志是否是这个意思，大家帮看看，谢谢！如下：

Backup Server: 3.42.1.1: LOAD is complete (database user_data).
Caution:  You have set up this database to include space on disk 16 for both
data and the transaction log.  This can make recovery impossible if that disk
fails.
Caution:  You have set up this database to include space on disk 17 for both
data and the transaction log.  This can make recovery impossible if that disk
fails.
Redo pass: 3774 records done (23%); 12000 records left.
Redo pass: 9774 records done (61%); 6000 records left.
Redo pass of recovery has processed 3520 committed and 0 aborted transactions.
Use the ONLINE DATABASE command to bring this database online; SQL Server will
not bring it online automatically.

另外，online database后，所dbcc checkdb检查，其中的syslogs出现如下提示，不知何意？如下
Checking syslogs
The total number of data pages in this table is 49493.
*** NOTICE:  Space used on the log segment is 96.67 Mbytes, 0.79%.
*** NOTICE:  Space free on the log segment is 12091.33 Mbytes, 99.21%.
Table has 881831 data rows.
再次感谢各位的大力支持，谢谢！

wfcjz

提示你的日志空间不足，想办法扩一下！