Rename DG
Le but est de
passer le diskgroup POCDB à DATARAD et
POCREDO à REDO_RAD.
On a une bd
qui s appuie sur ces diskgroups.
De preference
pratiquer sur de la non prod voir un poc afin d’etre à l aise.
La lecture
des note oracle : Doc ID 948040.1, Doc ID 1335975.1 serait un plus.
exemple
Ci-dessous c’est un exemple bien sur à adapter.
|
#actions
|
commande
|
user
|
|
1
|
# srvctl config database -d POCRAD
|
oracle os
|
|
2
|
select 'ALTER DATABASE RENAME FILE
'''||MEMBER||''' TO '''||REPLACE(MEMBER,'+POCREDO','+REDO_RAD')||'''; ' FROM
v$logfile;
|
oracle sqlplus
|
|
3
|
Select 'ALTER DATABASE RENAME FILE '''||NAME||'''
TO '''||REPLACE(NAME,'+POCDB','+DATARAD')||'''; ' from V$tempfile;
|
oracle sqlplus
|
|
4
|
Select 'ALTER DATABASE RENAME FILE '''||NAME||'''
TO '''||REPLACE(NAME,'+POCDB','+DATARAD')||'''; ' from V$datafile;
|
oracle sqlplus
|
|
5
|
alter database disable block change tracking;
|
oracle sqlplus
|
|
|
Attention au controlfiles
|
|
|
6
|
alter database
flashback off;
|
oracle sqlplus
|
|
7
|
create pfile from
spfile;
modifier le spfile
pour les nouveau diskgroupe
|
oracle sqlplus
|
|
8
|
srvctl stop database -d POCRAD
|
oracle os
|
|
9
|
ASMCMD> lsdg
lsdsk -k -G <diskgroup>
|
grid asmcmd
|
|
10
|
ASMCMD> umount POCDB
|
grid asmcmd
|
|
11
|
ASMCMD> umount
POCREDO
|
grid asmcmd
|
|
12
|
#renamedg phase=both dgname=POCREDO
newdgname=REDO_RAD verbose=true asm_diskstring='/dev/oracleasm/disks/*
|
grid os
|
|
13
|
#renamedg phase=both dgname=POCDB
newdgname=DATARAD verbose=true asm_diskstring='/dev/oracleasm/disks/*
|
grid os
|
|
14
|
ASMCMD> lsdg
|
grid asmcmd
|
|
15
|
ASMCMD> mount DATARAD
|
grid asmcmd
|
|
16
|
ASMCMD>
mount REDO_RAD
|
grid asmcmd
|
|
17
|
crsctl stat res -w
'TYPE = ora.diskgroup.type' –t
|
grid os
|
|
18
|
startup nomount
pfile='/u01/oracle/product/11.2.0/db_1/dbs/initorcl.ora.bkp'
|
oracle sqlplus
|
|
19
|
alter database
mount;
|
oracle sqlplus
|
|
20
|
Rejouer les scripts
issus des etapes #2,#3,#4
|
oracle sqlplus
|
|
21
|
alter database
mount;
|
oracle sqlplus
|
|
22
|
alter database
open;
|
oracle sqlplus
|
|
23
|
create spfile from pfile=’xxxxxx’;
|
oracle sqlplus
|
|
24
|
srvctl modify database -d POCRAD -p
'+REDO_RAD/POCRAD/PARAMETERFILE/spfilePOCRAD.ora'
|
oracle os
|
|
25
|
modifier le fichier
initSID.ora pour ajouter le nouveau spfile (sur chaque noeud)
|
oracle os
|
|
26
|
srvctl modify database -d POCRAD -a
"POCARCH,REDO_RAD,DATARAD"
|
oracle os
|
|
28
|
shutdown immediate
|
oracle sqlplus
|
|
29
|
srvctl start database -d POCRAD
|
oracle os
|
|
30
|
srvctl remove diskgroup -g POCREDO -f
|
oracle os
|
|
31
|
srvctl remove diskgroup -g POCDB -f
|
oracle os
|
Etape #9 :
output du lsdsk
ASMCMD> lsdsk -k
-G datarad
Total_MB Free_MB
OS_MB Name Failgroup Failgroup_Type Library
Label UDID Product
Redund Path
262064
23752 262064 DATARAD_0006
DATARAD_0006 REGULAR System UNKNOWN /dev/rdsk/c1d14s0
262064
23752 262064 DATARAD_0001
DATARAD_0001 REGULAR System UNKNOWN /dev/rdsk/c1d18s0
262064
23752 262064 DATARAD_0007
DATARAD_0007 REGULAR System UNKNOWN /dev/rdsk/c1d19s0
ASMCMD> lsdsk -k -G
redo_rad
Total_MB Free_MB
OS_MB Name Failgroup Failgroup_Type Library
Label UDID Product
Redund Path
65448
56920 65448 REDO_RAD_0001
REDO_RAD_0001 REGULAR System UNKNOWN /dev/rdsk/c1d16s0
Etape #12 :
outout renamdg
[dev-mtlunt119:grid::/home/grid/mg]#renamedg phase=both dgname=REDO_RAD
newdgname=REDORA verbose=true asm_diskstring='/dev/rdsk/*'
Parsing parameters..
Parameters in effect:
Old DG name : REDO_RAD
New DG name : REDORA
Phases :
Phase 1
Phase 2
Discovery str : /dev/rdsk/*
Clean : TRUE
Raw only : TRUE
renamedg operation: phase=both dgname=REDO_RAD newdgname=REDORA
verbose=true asm_diskstring=/dev/rdsk/*
Executing phase 1
Discovering the group
Performing discovery with string:/dev/rdsk/*
Identified disk UFS:/dev/rdsk/c1d16s0 with disk number:1 and timestamp
(33036270 -1617638400)
Checking for hearbeat...
Re-discovering the group
Performing discovery with string:/dev/rdsk/*
Identified disk UFS:/dev/rdsk/c1d16s0
with disk number:1 and timestamp (33036270 -1617638400)
Checking if the diskgroup is mounted or used by CSS
Checking disk number:1
Generating configuration file..
Completed phase 1
Executing phase 2
Looking for /dev/rdsk/c1d16s0
Modifying the header
Completed phase 2
Terminating kgfd context 1001bb480
Output du lsdsk
post rename:
ASMCMD>
lsdsk -k
-G datara
Total_MB Free_MB
OS_MB Name Failgroup Failgroup_Type Library
Label UDID Product
Redund Path
262064
23752 262064 DATARAD_0006
DATARAD_0006 REGULAR System UNKNOWN /dev/rdsk/c1d14s0
262064
23752 262064 DATARAD_0001
DATARAD_0001 REGULAR System UNKNOWN /dev/rdsk/c1d18s0
262064
23752 262064 DATARAD_0007
DATARAD_0007 REGULAR System UNKNOWN /dev/rdsk/c1d19s0
ASMCMD>
lsdsk -k -g redora
ASMCMD>
lsdsk -k
-G redora
Total_MB Free_MB
OS_MB Name Failgroup Failgroup_Type Library
Label UDID Product
Redund Path
65448
56920 65448 REDO_RAD_0001
REDO_RAD_0001 REGULAR System UNKNOWN /dev/rdsk/c1d16s0
Erreur
Si on a cette
erreur dans les logs cluster c est lie a golden gate :
2016-05-26 13:54:17.296: [ CRSPE][1177860416] {2:48629:58091} Register
Resource Opora.DATARAD.dg] has failed:CRS-2566:
User 'grid'
does not have sufficient permissions to operate on resource 'pocrac-gg', which
is part of the dependency specification.
Faire
executer par un root user cette
commande :
crsctl setperm resource pocrac-gg -u user:grid:rwx
Aucun commentaire:
Enregistrer un commentaire