pat98's always & forever

RAC Database AUTO START 

자동관리 -> 수동관리로

1. 확인

[EXA1]oracle@exa1:/home/oracle# srvctl config database -d exa

Database unique name: EXA

Database name: EXA

Oracle home: /u01/app/oracle/product/11.2.0.4/dbhome_1

Oracle user: oracle

Spfile: +DATA/EXA/spfileEXA.ora

Domain: 

Start options: open

Stop options: immediate

Database role: PRIMARY

Management policy: AUTOMATIC

Server pools: EXA

Database instances: EXA1,EXA2

Disk Groups: DATA,RECO

Mount point paths: 

Services: 

Type: RAC

Database is administrator managed

2. 변경

[EXA1]oracle@exa1:/home/oracle# srvctl modify database -d exa -y MANUAL

3. 확인

[EXA1]oracle@exa1:/home/oracle# srvctl config database -d exa

Database unique name: EXA

Database name: EXA

Oracle home: /u01/app/oracle/product/11.2.0.4/dbhome_1

Oracle user: oracle

Spfile: +DATA/EXA/spfileEXA.ora

Domain: 

Start options: open

Stop options: immediate

Database role: PRIMARY

Management policy: MANUAL

Server pools: EXA

Database instances: EXA1,EXA2

Disk Groups: DATA,RECO

Mount point paths: 

Services: 

Type: RAC

Database is administrator managed

Exadata Cell node Start, Stop, Restart

CellCLI> alter cell startup services rs

CellCLI> alter cell startup services ms

CellCLI> alter cell startup services cellsrv 

CellCLI> alter cell shutdown services rs

CellCLI> alter cell shutdown services ms

CellCLI> alter cell shutdown services cellsrv 

CellCLI> alter cell shutdown services all

CellCLI> alter cell shutdown services all force

CellCLI> alter cell restart services rs

CellCLI> alter cell restart services ms

CellCLI> alter cell restart services cellsrv

CellCLI> alter cell restart services all

CellCLI> alter cell restart services all force

ssh 접속후..

maintenance system poweroff 

This will turn off power to the appliance. Are you sure? (Y/N) Y

끝

증상 : DB startup 시 Fail 되고 "Direct NFS: please check that oradism is setuid" 메세지 남음

- alertlog 내용 -

Direct NFS: please check that oradism is setuid

Mon Mar 03 15:45:22 2014

Direct NFS: please check that oradism is setuid

Mon Mar 03 15:46:22 2014

Direct NFS: please check that oradism is setuid

Mon Mar 03 15:47:21 2014

Errors in file /u01/app/oracle/diag/rdbms/red/RED4/trace/RED4_diag_47735.trc  (incident=41009):

ORA-00494: ???? ?? [CF]?(?) ?? ????(900? ??) 'inst 4, osid 306'? ?? ???????.

Incident details in: /u01/app/oracle/diag/rdbms/red/RED4/incident/incdir_41009/RED4_diag_47735_i41009.trc

Mon Mar 03 15:47:22 2014

Direct NFS: please check that oradism is setuid

Mon Mar 03 15:47:23 2014

Sweep [inc][41009]: completed

Sweep [inc2][41009]: completed

Mon Mar 03 15:48:21 2014

Killing enqueue blocker (pid=306) on resource CF-00000000-00000000 by (pid=47752)

 by killing session 1289.3003

Mon Mar 03 15:48:23 2014

Direct NFS: please check that oradism is setuid

확인사항 : 

해당 화일의 owner,permission을 확인해 보면 root -> oracle 로 바뀌어져 있다.

$ls -l  $ORACLE_HOME/bin/oradism

%rwsr-xr-x oracle dba oradism

해결책 :

#chown root:root oradism 

Set the setuid

#chmod 4755 oradism

1. 전체노드 ASM 내리기

2. 기존 정보확인

CellCLI> list griddisk
         DATAC1_CD_00_gtceladm01         active
         DATAC1_CD_01_gtceladm01         active
         DATAC1_CD_02_gtceladm01         active
         DATAC1_CD_03_gtceladm01         active
         DATAC1_CD_04_gtceladm01         active
         DATAC1_CD_05_gtceladm01         active
         DATAC1_CD_06_gtceladm01         active
         DATAC1_CD_07_gtceladm01         active
         DATAC1_CD_08_gtceladm01         active
         DATAC1_CD_09_gtceladm01         active
         DATAC1_CD_10_gtceladm01         active
         DATAC1_CD_11_gtceladm01         active
         DBFS_DG_CD_02_gtceladm01        active
         DBFS_DG_CD_03_gtceladm01        active
         DBFS_DG_CD_04_gtceladm01        active
         DBFS_DG_CD_05_gtceladm01        active
         DBFS_DG_CD_06_gtceladm01        active
         DBFS_DG_CD_07_gtceladm01        active
         DBFS_DG_CD_08_gtceladm01        active
         DBFS_DG_CD_09_gtceladm01        active
         DBFS_DG_CD_10_gtceladm01        active
         DBFS_DG_CD_11_gtceladm01        active
         RECOC1_CD_00_gtceladm01         active
         RECOC1_CD_01_gtceladm01         active
         RECOC1_CD_02_gtceladm01         active
         RECOC1_CD_03_gtceladm01         active
         RECOC1_CD_04_gtceladm01         active
         RECOC1_CD_05_gtceladm01         active
         RECOC1_CD_06_gtceladm01         active

3. 이름변경 (default 로 붙은 C1을 빼고 싶다)

CellCLI> alter griddisk DATAC1_CD_00_gtceladm01 name='DATA_CD_00_gtceladm01'
GridDisk DATA_CD_00_gtceladm01 successfully altered

CellCLI> alter griddisk DATAC1_CD_01_gtceladm01 name='DATA_CD_01_gtceladm01'
GridDisk DATA_CD_01_gtceladm01 successfully altered

CellCLI> alter griddisk DATAC1_CD_02_gtceladm01 name='DATA_CD_02_gtceladm01'
GridDisk DATA_CD_02_gtceladm01 successfully altered

CellCLI> alter griddisk DATAC1_CD_03_gtceladm01 name='DATA_CD_03_gtceladm01'
GridDisk DATA_CD_03_gtceladm01 successfully altered

CellCLI> alter griddisk DATAC1_CD_04_gtceladm01 name='DATA_CD_04_gtceladm01'
GridDisk DATA_CD_04_gtceladm01 successfully altered

CellCLI> alter griddisk DATAC1_CD_05_gtceladm01 name='DATA_CD_05_gtceladm01'
GridDisk DATA_CD_05_gtceladm01 successfully altered

CellCLI> alter griddisk DATAC1_CD_06_gtceladm01 name='DATA_CD_06_gtceladm01'
GridDisk DATA_CD_06_gtceladm01 successfully altered

CellCLI> alter griddisk DATAC1_CD_07_gtceladm01 name='DATA_CD_07_gtceladm01'
GridDisk DATA_CD_07_gtceladm01 successfully altered

CellCLI> alter griddisk DATAC1_CD_08_gtceladm01 name='DATA_CD_08_gtceladm01'
GridDisk DATA_CD_08_gtceladm01 successfully altered

CellCLI> alter griddisk DATAC1_CD_09_gtceladm01 name='DATA_CD_09_gtceladm01'
GridDisk DATA_CD_09_gtceladm01 successfully altered

CellCLI> alter griddisk DATAC1_CD_10_gtceladm01 name='DATA_CD_10_gtceladm01'
GridDisk DATA_CD_10_gtceladm01 successfully altered

CellCLI> alter griddisk DATAC1_CD_11_gtceladm01 name='DATA_CD_11_gtceladm01'
GridDisk DATA_CD_11_gtceladm01 successfully altered

4. 신규 정보확인

CellCLI> list griddisk
         DATA_CD_00_gtceladm01           active
         DATA_CD_01_gtceladm01           active
         DATA_CD_02_gtceladm01           active
         DATA_CD_03_gtceladm01           active
         DATA_CD_04_gtceladm01           active
         DATA_CD_05_gtceladm01           active
         DATA_CD_06_gtceladm01           active
         DATA_CD_07_gtceladm01           active
         DATA_CD_08_gtceladm01           active
         DATA_CD_09_gtceladm01           active
         DATA_CD_10_gtceladm01           active
         DATA_CD_11_gtceladm01           active
         DBFS_DG_CD_02_gtceladm01        active
         DBFS_DG_CD_03_gtceladm01        active
         DBFS_DG_CD_04_gtceladm01        active
         DBFS_DG_CD_05_gtceladm01        active
         DBFS_DG_CD_06_gtceladm01        active
         DBFS_DG_CD_07_gtceladm01        active
         DBFS_DG_CD_08_gtceladm01        active
         DBFS_DG_CD_09_gtceladm01        active
         DBFS_DG_CD_10_gtceladm01        active
         DBFS_DG_CD_11_gtceladm01        active
         RECO_CD_00_gtceladm01           active
         RECO_CD_01_gtceladm01           active
         RECO_CD_02_gtceladm01           active
         RECO_CD_03_gtceladm01           active
         RECO_CD_04_gtceladm01           active
         RECO_CD_05_gtceladm01           active
         RECO_CD_06_gtceladm01           active
         RECO_CD_07_gtceladm01           active
         RECO_CD_08_gtceladm01           active
         RECO_CD_09_gtceladm01           active
         RECO_CD_10_gtceladm01           active
         RECO_CD_11_gtceladm01           active

버전 : 11.2.0.4

OS : HP-UX 11.31

Filesystem : RAW device

증상 : RAW device 이용 DB 생성시 에러가지 권한에러와 함께 파일 생성이 안됨 (dd 해도 안됨, 권한 및 permission 정상)

원인 : RAW header에 corrupt 또는 이상한 data가 기록되어 있을시 또는 이전에 한번 사용을 했을시 header 손상으로 판단하고 에러뿌림

       이전 버전에서는 header가 이상해도 그냥 써버렸으나 11.2.0.4 에서는 무시하지 않고 에러를 뿌린다고 함

Fix :  12.1.0.2  or 12.2

해결책 : Datafile 생성 또는 추가시 REUSE 옵션을 사용하여 생성하면 에러없음.

            or raw device 사용하지 말라.

ex) CREATE TABLESPACE UNDOTBS1 DATAFILE '/dev/vg01/rundotbs01' SIZE 4000M REUSE AUTOEXTEND OFF ;

참고문서 :

Bug 18508861 - Create datafile on RAW can produce a corrupt file header / causes ORA-17515 and crash in 11.2.0.4 (문서 ID 18508861.8)

패스워드 복잡성 때문에 유저 생성시 에러가 난다면..

ORA-28003: password verification for the specified password failed

ORA-20006: Password too simple

1.Disable password validation

SQL> ALTER PROFILE DEFAULT LIMIT PASSWORD_VERIFY_FUNCTION NULL;

2. Enable the password validation:

SQL> ALTER PROFILE DEFAULT LIMIT PASSWORD_VERIFY_FUNCTION verify_function_11G;

1. ILOM 이용 remote powering

# ipmitool -H dm01cel01-ilom -U root chassis power on

2. 해당 시스템의 파워 중단/작업

start /System

stop /System

stop -force /System

reset /System

3. ssh 로 콘솔 원격접속

start /SP/console

4. ssh 로 이벤트 로그확인

cd /SP/logs/event

show

show list

5. ILOM 이용 SP reset

reset /SP

6. ssh 로 ILOM 에 접속한후..바꾸고자 하는 ip로..

set /SP/network pendingipdiscovery=static
set /SP/network pendingipaddress=10.10.10.10
set /SP/network pendingipgateway=10.10.10.1
set /SP/network pendingipnetmask=255.255.255.0
set /SP/network commitpending=true

확인

-> show /SP/network

오라클 데이터베이스에서 제공하는 관리 사용자 계정에 대한 요약 

11g RAC 2node -> 3node 추가작업..

직접해 보았습니다. 쉽네요..

11g RAC 3node add.pdf