failed parse elapsed time过大分析案例

朋友公司做压力测试,awr比较异常,让给看看
1-1


这里显示数据库db time较大,数据库应该比较繁忙,主要等待事件为:library cache: mutex X
1-4

但是Load Profile显示Parses (SQL)和Hard parses (SQL)均不大
1-2

但是发现failed parse elapsed time特别大,也就是说这个库出现该问题,主要可能是由于sql语句执行解析失败导致,而解析失败最大的可能性就是sql语句语法/权限错误.对于这类问题可以通过设置event 10035进行跟踪
1-3

演示设置event 10035进行跟踪的效果

SQL> select * from v$version;

BANNER
--------------------------------------------------------------------------------
Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
PL/SQL Release 11.2.0.4.0 - Production
CORE    11.2.0.4.0      Production
TNS for 64-bit Windows: Version 11.2.0.4.0 - Production
NLSRTL Version 11.2.0.4.0 - Production

SQL> ALTER SYSTEM SET EVENTS '10035 trace name context forever, level 1';

System altered.

SQL> ALTER SESSION SET EVENTS '10035 trace name context forever, level 1';

Session altered.

SQL> select 1;
select 1
       *
ERROR at line 1:
ORA-00923: FROM keyword not found where expected


SQL> select * from xifenfei_t;
select * from xifenfei_t
              *
ERROR at line 1:
ORA-00942: table or view does not exist

查看alert日志,观察错误语句
1-5


发表在 Oracle性能优化 | 标签为 | 留下评论

ERROR: diskgroup XXXX was not mounted

aix平台10.2.0.5 2节点RAC,由于节点2系统盘故障,通过节点1镜像系统,复制到节点2,结果由于节点2磁盘顺序和节点1不匹配,aix工程师进行了相关操作之后,节点1重启之后datadg磁盘组无法mount

SQL> alter diskgroup datadg mount 
Mon Jun 10 23:23:46 CST 2019
NOTE: cache registered group DATADG number=1 incarn=0x8cf61164
Mon Jun 10 23:23:46 CST 2019
NOTE: Hbeat: instance first (grp 1)
Mon Jun 10 23:23:50 CST 2019
NOTE: start heartbeating (grp 1)
Mon Jun 10 23:23:50 CST 2019
NOTE: cache dismounting group 1/0x8CF61164 (DATADG) 
NOTE: dbwr not being msg'd to dismount
ERROR: diskgroup DATADG was not mounted

检查datadg磁盘组相关信息

Tue Jan 29 19:21:45 CST 2019
NOTE: start heartbeating (grp 2)
NOTE: cache opening disk 0 of grp 2: DATADG_0000 path:/dev/rhdisk6
Tue Jan 29 19:21:45 CST 2019
NOTE: F1X0 found on disk 0 fcn 0.0
NOTE: cache opening disk 1 of grp 2: DATADG_0001 path:/dev/rhdisk7
NOTE: cache opening disk 2 of grp 2: DATADG_0002 path:/dev/rhdisk8
NOTE: cache opening disk 3 of grp 2: DATADG_0003 path:/dev/rhdisk9
NOTE: cache mounting (first) group 2/0x60E59155 (DATADG)
* allocate domain 2, invalid = TRUE 
Tue Jan 29 19:21:45 CST 2019
NOTE: attached to recovery domain 2
Tue Jan 29 19:21:45 CST 2019
NOTE: cache recovered group 2 to fcn 0.849668
Tue Jan 29 19:21:45 CST 2019
NOTE: LGWR attempting to mount thread 1 for disk group 2
NOTE: LGWR mounted thread 1 for disk group 2
NOTE: opening chunk 1 at fcn 0.849668 ABA 
NOTE: seq=21 blk=5394 
Tue Jan 29 19:21:46 CST 2019
NOTE: cache mounting group 2/0x60E59155 (DATADG) succeeded
SUCCESS: diskgroup DATADG was mounted

通过这里可以看出来datadg磁盘组是由rhdisk6-9 四块磁盘组成,查询相关磁盘信息发现
asm-foreign


这里确定rhdisk7磁盘异常,通过kfed分析磁盘情况

D:\BaiduNetdiskDownload\xifenfei>kfed read rhdisk7.dd
kfbh.endian:                          0 ; 0x000: 0x00
kfbh.hard:                           34 ; 0x001: 0x22
kfbh.type:                            0 ; 0x002: KFBTYP_INVALID
kfbh.datfmt:                          0 ; 0x003: 0x00
kfbh.block.blk:                   49407 ; 0x004: blk=49407
kfbh.block.obj:                       0 ; 0x008: file=0
kfbh.check:                           0 ; 0x00c: 0x00000000
kfbh.fcn.base:                    58396 ; 0x010: 0x0000e41c
kfbh.fcn.wrap:                   131072 ; 0x014: 0x00020000
kfbh.spare1:                 4294967064 ; 0x018: 0xffffff18
kfbh.spare2:                 2105310074 ; 0x01c: 0x7d7c7b7a
005918A00 00002200 0000C0FF 00000000 00000000  [."..............]
005918A10 0000E41C 00020000 FFFFFF18 7D7C7B7A  [............z{|}]
005918A20 00000000 00000000 00000000 00000000  [................]
  Repeat 253 times
KFED-00322: Invalid content encountered during block traversal: [kfbtTraverseBlock][Invalid OSM block type][][0]

D:\BaiduNetdiskDownload\xifenfei>kfed read rhdisk7.dd blkn=1
kfbh.endian:                          0 ; 0x000: 0x00
kfbh.hard:                            0 ; 0x001: 0x00
kfbh.type:                            0 ; 0x002: KFBTYP_INVALID
kfbh.datfmt:                          0 ; 0x003: 0x00
kfbh.block.blk:                       0 ; 0x004: blk=0
kfbh.block.obj:                       0 ; 0x008: file=0
kfbh.check:                           0 ; 0x00c: 0x00000000
kfbh.fcn.base:                        0 ; 0x010: 0x00000000
kfbh.fcn.wrap:                        0 ; 0x014: 0x00000000
kfbh.spare1:                          0 ; 0x018: 0x00000000
kfbh.spare2:                          0 ; 0x01c: 0x00000000
006EF8A00 00000000 00000000 00000000 00000000  [................]
  Repeat 255 times
KFED-00322: Invalid content encountered during block traversal: [kfbtTraverseBlock][Invalid OSM block type][][0]

D:\BaiduNetdiskDownload\xifenfei>kfed read rhdisk7.dd blkn=2|more
kfbh.endian:                          0 ; 0x000: 0x00
kfbh.hard:                          130 ; 0x001: 0x82
kfbh.type:                            3 ; 0x002: KFBTYP_ALLOCTBL
kfbh.datfmt:                          1 ; 0x003: 0x01
kfbh.block.blk:                33554432 ; 0x004: blk=33554432
kfbh.block.obj:                16777344 ; 0x008: file=128
kfbh.check:                  3844041089 ; 0x00c: 0xe51f6981
kfbh.fcn.base:               1297484544 ; 0x010: 0x4d560b00
kfbh.fcn.wrap:                        0 ; 0x014: 0x00000000
kfbh.spare1:                          0 ; 0x018: 0x00000000
kfbh.spare2:                          0 ; 0x01c: 0x00000000
kfdatb10.aunum:                       0 ; 0x000: 0x00000000
kfdatb10.shrink:                  49153 ; 0x004: 0xc001
kfdatb10.ub2pad:                  20555 ; 0x006: 0x504b
kfdatb10.auinfo[0].link.next:      2048 ; 0x008: 0x0800
kfdatb10.auinfo[0].link.prev:      2048 ; 0x00a: 0x0800
kfdatb10.auinfo[0].free:              0 ; 0x00c: 0x0000
kfdatb10.auinfo[0].total:         49153 ; 0x00e: 0xc001
kfdatb10.auinfo[1].link.next:      4096 ; 0x010: 0x1000
kfdatb10.auinfo[1].link.prev:      4096 ; 0x012: 0x1000
kfdatb10.auinfo[1].free:              0 ; 0x014: 0x0000
kfdatb10.auinfo[1].total:             0 ; 0x016: 0x0000
kfdatb10.auinfo[2].link.next:      6144 ; 0x018: 0x1800
kfdatb10.auinfo[2].link.prev:      6144 ; 0x01a: 0x1800
kfdatb10.auinfo[2].free:              0 ; 0x01c: 0x0000
kfdatb10.auinfo[2].total:             0 ; 0x01e: 0x0000
kfdatb10.auinfo[3].link.next:      8192 ; 0x020: 0x2000
kfdatb10.auinfo[3].link.prev:      8192 ; 0x022: 0x2000
kfdatb10.auinfo[3].free:              0 ; 0x024: 0x0000

对比磁盘可能的损坏情况,由于在aix 平台asm disk的block有一个特征一般0082开头,通过工具打开磁盘,检索该标记对比
正常磁盘
0082-good


异常磁盘
0082-had

通过上述分析,大概评估rhdisk7 元数据部分损坏的不光是block 0和1,人工修复继续使用的可能性不太大,而且基于客户的数据库不大,采取方案是直接拷贝数据文件、redo、控制文件到文件系统,然后在本地文件系统open库
open_db

运气不错,实现完美恢复数据0丢失

发表在 Oracle ASM, 非常规恢复 | 标签为 , | 留下评论

WARNING: Read Failed.导致asm磁盘组异常

有客户对asm dg进行扩容,一段时间之后,asm data 磁盘组直接dismount

Wed May 29 18:37:25 2019
SUCCESS: ALTER DISKGROUP DATA ADD  DISK '/dev/oracleasm/disks/DATA_0028' SIZE 511993M ,
'/dev/oracleasm/disks/DATA_0027' SIZE 511993M ,
'/dev/oracleasm/disks/DATA_0026' SIZE 511993M ,
'/dev/oracleasm/disks/DATA_0025' SIZE 511993M /* ASMCA */
NOTE: starting rebalance of group 1/0x9e18e2f1 (DATA) at power 1
Wed May 29 18:37:26 2019
Starting background process ARB0
Wed May 29 18:37:26 2019
ARB0 started with pid=34, OS id=96638 
NOTE: assigning ARB0 to group 1/0x9e18e2f1 (DATA) with 1 parallel I/O
NOTE: Attempting voting file refresh on diskgroup DATA
NOTE: Refresh completed on diskgroup DATA. No voting file found.
cellip.ora not found.
Wed May 29 19:21:43 2019
WARNING: Read Failed. group:1 disk:27 AU:0 offset:360448 size:4096
WARNING: cache failed reading from group=1(DATA) dsk=27 blk=88 count=1 from disk= 27 
(DATA_0027) kfkist=0x20 status=0x02 osderr=0x0 file=kfc.c line=11596
ERROR: cache failed to read group=1(DATA) dsk=27 blk=88 from disk(s): 27(DATA_0027)
ORA-15080: synchronous I/O operation to a disk failed
ORA-27072: File I/O error
Linux-x86_64 Error: 5: Input/output error
Additional information: 4
Additional information: 704
Additional information: -1
NOTE: cache initiating offline of disk 27 group DATA
NOTE: process _user31879_+asm1 (31879) initiating offline of disk 27.3915911747 (DATA_0027) with mask 0x7e in group 1
NOTE: initiating PST update: grp = 1, dsk = 27/0xe9681243, mask = 0x6a, op = clear
Wed May 29 19:21:43 2019
GMON updating disk modes for group 1 at 10 for pid 35, osid 31879
ERROR: Disk 27 cannot be offlined, since diskgroup has external redundancy.
ERROR: too many offline disks in PST (grp 1)
Wed May 29 19:21:43 2019
NOTE: cache dismounting (not clean) group 1/0x9E18E2F1 (DATA) 
NOTE: messaging CKPT to quiesce pins Unix process pid: 90256, image: oracle@ftz-db-o1 (B000)
Wed May 29 19:21:43 2019
NOTE: halting all I/Os to diskgroup 1 (DATA)
WARNING: Offline for disk DATA_0027 in mode 0x7f failed.
Wed May 29 19:21:43 2019
NOTE: LGWR doing non-clean dismount of group 1 (DATA)
NOTE: LGWR sync ABA=27.3207 last written ABA 27.3207
Wed May 29 19:21:43 2019
ERROR: ORA-15130 thrown in ARB0 for group number 1
Errors in file /oracle/grid_base/diag/asm/+asm/+ASM1/trace/+ASM1_arb0_96638.trc:
ORA-15130: diskgroup "" is being dismounted
ORA-15130: diskgroup "DATA" is being dismounted
Wed May 29 19:21:43 2019
NOTE: stopping process ARB0

后续继续mount data 磁盘组成功,但是立马又dismount

Wed May 29 18:37:25 2019
SUCCESS: ALTER DISKGROUP DATA ADD  DISK '/dev/oracleasm/disks/DATA_0028' SIZE 511993M ,
'/dev/oracleasm/disks/DATA_0027' SIZE 511993M ,
'/dev/oracleasm/disks/DATA_0026' SIZE 511993M ,
'/dev/oracleasm/disks/DATA_0025' SIZE 511993M /* ASMCA */
NOTE: starting rebalance of group 1/0x9e18e2f1 (DATA) at power 1
Wed May 29 18:37:26 2019
Starting background process ARB0
Wed May 29 18:37:26 2019
ARB0 started with pid=34, OS id=96638 
NOTE: assigning ARB0 to group 1/0x9e18e2f1 (DATA) with 1 parallel I/O
NOTE: Attempting voting file refresh on diskgroup DATA
NOTE: Refresh completed on diskgroup DATA. No voting file found.
cellip.ora not found.
Wed May 29 19:21:43 2019
WARNING: Read Failed. group:1 disk:27 AU:0 offset:360448 size:4096
WARNING: cache failed reading from group=1(DATA) dsk=27 blk=88 count=1 from disk= 27 
(DATA_0027) kfkist=0x20 status=0x02 osderr=0x0 file=kfc.c line=11596
ERROR: cache failed to read group=1(DATA) dsk=27 blk=88 from disk(s): 27(DATA_0027)
ORA-15080: synchronous I/O operation to a disk failed
ORA-27072: File I/O error
Linux-x86_64 Error: 5: Input/output error
Additional information: 4
Additional information: 704
Additional information: -1
NOTE: cache initiating offline of disk 27 group DATA
NOTE: process _user31879_+asm1 (31879) initiating offline of disk 27.3915911747 (DATA_0027) with mask 0x7e in group 1
NOTE: initiating PST update: grp = 1, dsk = 27/0xe9681243, mask = 0x6a, op = clear
Wed May 29 19:21:43 2019
GMON updating disk modes for group 1 at 10 for pid 35, osid 31879
ERROR: Disk 27 cannot be offlined, since diskgroup has external redundancy.
ERROR: too many offline disks in PST (grp 1)
Wed May 29 19:21:43 2019
NOTE: cache dismounting (not clean) group 1/0x9E18E2F1 (DATA) 
NOTE: messaging CKPT to quiesce pins Unix process pid: 90256, image: oracle@ftz-db-o1 (B000)
Wed May 29 19:21:43 2019
NOTE: halting all I/Os to diskgroup 1 (DATA)
WARNING: Offline for disk DATA_0027 in mode 0x7f failed.
Wed May 29 19:21:43 2019
NOTE: LGWR doing non-clean dismount of group 1 (DATA)
NOTE: LGWR sync ABA=27.3207 last written ABA 27.3207
Wed May 29 19:21:43 2019
ERROR: ORA-15130 thrown in ARB0 for group number 1
Errors in file /oracle/grid_base/diag/asm/+asm/+ASM1/trace/+ASM1_arb0_96638.trc:
ORA-15130: diskgroup "" is being dismounted
ORA-15130: diskgroup "DATA" is being dismounted
Wed May 29 19:21:43 2019
NOTE: stopping process ARB0

对于上述的故障现象,本质原因是由于asm 磁盘组增加新磁盘之后,开始做rebalance,但是由于遭遇到 27号盘上有IO读错误,使得asm磁盘组无法正常完成rebalance,因而data磁盘组无法稳定的mount。解决该问题思路,通过patch asm磁盘组,禁止rebalance,从而使得data磁盘组不再dismount,再进行后续恢复

发表在 Oracle ASM, 非常规恢复 | 标签为 , , | 2 条评论

回收站中有大量wri$_rcs表

在对一套Oracle 12.1.0.2的数据库巡检之时发现大量WRI$_RCS_数字_1的表在回收站中,从命名中看该表应该是Oracle某个自动任务处理后,表未被正常处理干净,遗留在回收站中.
wri$_rcs


查询mos确认是Bug 20114306 – Objects left in recyclebin after upgrade to 12.1.0.2 or with fix for bug 16851194 present – superseded (文档 ID 20114306.8)
20114306

可以尝试打上补丁21498770或者23100700然后设置_fix_control

alter system set "_fix_control"='16851194:off' ;

确认该_fix_control是否可以设置,可以查询 v$system_fix_control视图

发表在 ORACLE 12C, Oracle Bug | 标签为 | 留下评论

ORA-600 kokasgi1故障恢复

数据库启动报ORA-600 kokasgi1错误

SMON: enabling tx recovery
Database Characterset is WE8ISO8859P1
Errors in file /u01/app/oracle/diag/rdbms/xifenfei/xifenfei1/trace/xifenfei1_ora_10056.trc  (incident=269259):
ORA-00600: internal error code, arguments: [kokasgi1], [], [], [], [], [], [], [], [], [], [], []
Incident details in: /u01/app/oracle/diag/rdbms/xifenfei/xifenfei1/incident/incdir_269259/xifenfei1_ora_10056_i269259.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Errors in file /u01/app/oracle/diag/rdbms/xifenfei/xifenfei1/trace/xifenfei1_ora_10056.trc:
ORA-00600: internal error code, arguments: [kokasgi1], [], [], [], [], [], [], [], [], [], [], []
Errors in file /u01/app/oracle/diag/rdbms/xifenfei/xifenfei1/trace/xifenfei1_ora_10056.trc:
ORA-00600: internal error code, arguments: [kokasgi1], [], [], [], [], [], [], [], [], [], [], []
Error 600 happened during db open, shutting down database
USER (ospid: 10056): terminating the instance due to error 600
Instance terminated by USER, pid = 10056
ORA-1092 signalled during: alter database open resetlogs...
opiodr aborting process unknown ospid (10056) as a result of ORA-1092
Sat May 25 09:40:21 2019
ORA-1092 : opitsk aborting process

该错误在mos上没有查询出来明确的解决方案,但是在google中有人删除user$模拟出该故障
ora-600-kokasgi1


数据库启动10046跟踪

PARSING IN CURSOR #140185422046848 len=189 dep=1 uid=0 oct=3 lid=0 tim=1558756188092143 hv=186852205 
ad='390983730' sqlid='2tkw12w5k68vd'
select user#,password,datats#,tempts#,type#,defrole,resource$, ptime,decode(defschclass,NULL,
'DEFAULT_CONSUMER_GROUP',defschclass),spare1,spare4,ext_username,spare2 from user$ where name=:1
END OF STMT
PARSE #140185422046848:c=0,e=784,p=0,cr=0,cu=0,mis=1,r=0,dep=1,og=4,plh=0,tim=1558756188092141
BINDS #140185422046848:
 Bind#0
  oacdty=01 mxl=32(03) mxlc=00 mal=00 scl=00 pre=00
  oacflg=18 fl2=0001 frm=01 csi=31 siz=32 off=0
  kxsbbbfp=7f7f7648a230  bln=32  avl=03  flg=05
  value="SYS"
EXEC #140185422046848:c=1000,e=1432,p=0,cr=0,cu=0,mis=1,r=0,dep=1,og=4,plh=1457651150,tim=1558756188093835
WAIT #140185422046848: nam='db file sequential read' ela= 5226 file#=1 block#=417 blocks=1 obj#=46 tim=1558756188099198
FETCH #140185422046848:c=1000,e=5465,p=1,cr=1,cu=0,mis=0,r=0,dep=1,og=4,plh=1457651150,tim=1558756188099349
STAT #140185422046848 id=1 cnt=0 pid=0 pos=1 obj=22 op='TABLE ACCESS BY INDEX ROWID USER$ (cr=1 pr=1 pw=0 time=5463 us)'
STAT #140185422046848 id=2 cnt=0 pid=1 pos=1 obj=46 op='INDEX UNIQUE SCAN I_USER1 (cr=1 pr=1 pw=0 time=5461 us)'
CLOSE #140185422046848:c=0,e=10,dep=1,type=0,tim=1558756188099578
ORA-00600: internal error code, arguments: [kokasgi1], [], [], [], [], [], [], [], [], [], [], []

这里比较明显数据库在查询user$中的SYS用户的时候,无法查询数据从而出现ORA-00600: internal error code, arguments: [kokasgi1]错误.通过进一步对USER$表进行分析发现,sys和system被人重命名

SQL> select name from user$ WHERE NAME LIKE 'SYS%';

NAME
------------------------------
SYSDW
SYSMAN
SYSTEMDW

定位到具体问题,解决比较简单,在oracle的open过程中,通过对user$表进行修复,实现数据库完美恢复.

发表在 非常规恢复 | 标签为 , , | 留下评论