6. Standby (aix_rac2) :
· Verification de la bonne synchronisation de la standby :
Sur la primaire
Archive log list;
Oldest online log sequence 26
Next log sequence to archive 28
La sequence courante est la 27.
Sur la standby avec la requete de verification on est aussi a la sequence 27.
· Arrêter le recover managed
alter database recover managed standby database cancel;
· Verification de la log db
· Verification du switchover status :
select switchover_status from v$database;
to primary
· le switch :
alter database commit to switchover to primary;
shutdown immediate
startup
select database_role from V$database;
alter system switch logfile;
7. Primaire (aix_rac1) :
alter database commit to switchover to standby;
shutdown immediate
startup mount
select database_role from V$database;
alter database recover managed standby database disconnect from session;
VI. Broker
Avant meme de demarrer voila la configuration par defaut sur la primaire :
SQL> show parameter broker
NAME VALUE
dg_broker_config_file1 /u01/app/oracle/11.2.0/dbs/dr1 MG.dat
dg_broker_config_file2 /u01/app/oracle/11.2.0/dbs/dr2MG.dat
dg_broker_start FALSE
alter system set dg_broker_config_file1=’ +data/MG/DATAGUARDCONFIG/dr1 MG.dat ';
alter system set dg_broker_config_file2='+FRA/MG/DATAGUARDCONFIG/dr2 MG.dat ';
Si on met les fichiers de config sur le meme dg on recoit l erreur ORA-02097
Sur les 2 serveurs je demarre le broker (process dmon)
alter system set dg_broker_start=true;
Dans le broker :
create configuration dguard as primary database is 'MG' connect identifier is 'MG';
add database ‘MGSTDBY’ as connect identifier is ‘MGSTDBY’ maintained as physical;
enable configuration;
· administration du broker :
Show configuration ;
DGMGRL> show configuration;
Configuration - dguard
Protection Mode: MaxPerformance
Databases:
MG - Primary database
Warning: ORA-16789: standby redo logs not configured
MGSTDBY - Physical standby database
Error: ORA-16797: database is not using a server parameter file
Fast-Start Failover: DISABLED
Configuration Status:
ERROR
Correction de ses erreurs
Sur MG@aix_rac1
ALTER DATABASE ADD STANDBY LOGFILE '+data'; quatre fois.
Sur MGSTDBY@aix_rac2
SQL> ALTER DATABASE ADD STANDBY LOGFILE '/u01/app/oracledata/arch/MGSTDBY/stdbylog/stdbylog1' size 104857600;
SQL> ALTER DATABASE ADD STANDBY LOGFILE '/u01/app/oracledata/arch/MGSTDBY/stdbylog/stdbylog2' size 104857600;
SQL> ALTER DATABASE ADD STANDBY LOGFILE '/u01/app/oracledata/arch/MGSTDBY/stdbylog/stdbylog3' size 104857600;
SQL> ALTER DATABASE ADD STANDBY LOGFILE '/u01/app/oracledata/arch/MGSTDBY/stdbylog/stdbylog4' size 104857600;
Create spfile from pfile;
Shutdown immediate
Startup mount
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE DISCONNECT FROM SESSION;
DGMGRL> show configuration;
Configuration - dguard
Protection Mode: MaxPerformance
Databases:
MG - Primary database
MGSTDBY - Physical standby database (disabled)
Fast-Start Failover: DISABLED
Configuration Status:
SUCCESS
DGMGRL> show database verbose 'MG';
Voila les properties :
DGConnectIdentifier | 'MG' |
ObserverConnectIdentifier | '' |
LogXptMode | 'ASYNC' |
DelayMins | '0' |
Binding | 'optional' |
MaxFailure | '0' |
MaxConnections | '1' |
ReopenSecs | '300' |
NetTimeout | '30' |
RedoCompression | 'DISABLE' |
LogShipping | 'ON' |
PreferredApplyInstance | '' |
ApplyInstanceTimeout | '0' |
ApplyParallel | 'AUTO' |
StandbyFileManagement | 'MANUAL' |
ArchiveLagTarget | '0' |
LogArchiveMaxProcesses | '4' |
LogArchiveMinSucceedDest | '1' |
DbFileNameConvert | '/u01/app/oracledata/stdby, +DATA/MG/datafile, /u01/app/oracledata/stdby, +DATA/MG/tempfile' |
LogFileNameConvert | ' /u01/app/oracledata/stdby, +DATA/MG/onelinelog' |
FastStartFailoverTarget | '' |
StatusReport | '(monitor)' |
InconsistentProperties | '(monitor)' |
InconsistentLogXptProps | '(monitor)' |
SendQEntries | '(monitor)' |
LogXptStatus | '(monitor)' |
RecvQEntries | '(monitor)' |
HostName | 'aix_rac1' |
SidName | 'MG' |
StaticConnectIdentifier | (DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=192.168.100.180)(PORT=1521)) (CONNECT_DATA=(SERVICE_NAME=MG_DGMGRL)(INSTANCE_NAME=MG)(SERVER=DEDICATED)))' |
StandbyArchiveLocation | 'USE_DB_RECOVERY_FILE_DEST' |
AlternateLocation | '' |
LogArchiveTrace | '0' |
LogArchiveFormat | '%t_%s_%r.dbf' |
TopWaitEvents | '(monitor)' |
DGMGRL> show database verbose 'MGSTDBY';
· La log du broker est
$diag_dest/rdbms/<$ORACLE_SID>/<$ORACLE_SID>/trace/drc<$ORACLE_SID>.log
8. Bascule dataguard avec le broker :
Elle peut se faire a distance mais je fais ca depuis un des serveurs impliqué dans la configuration.
switchover to ‘MGSTDBY’; ( dep uis le role standby vers le primaire).
DGMGRL> switchover to 'MGSTDBY';
Performing switchover NOW, please wait...
New primary database "MGSTDBY" is opening...
Operation requires shutdown of instance "MG" on database "MG"
Shutting down instance "MG"...
ORA-01109: database not open
Database dismounted.
ORACLE instance shut down.
Operation requires startup of instance "MG" on database "MG"
Starting instance "MG"...
Unable to connect to database
ORA-12514: TNS:listener does not currently know of service requested in connect descriptor
Failed.
Warning: You are no longer connected to ORACLE.
Please complete the following steps to finish switchover:
start up and mount instance "MG" of database "MG"
DGMGRL>
Apres stop/start de la base MG@aix_rac1
DGMGRL> show configuration;
Configuration - dguard
Protection Mode: MaxPerformance
Databases:
MGSTDBY - Primary database
MG - Physical standby database
Fast-Start Failover: DISABLED
Configuration Status:
SUCCESS
Les db sont synchrones .
Retoure sur MG :
switchover to 'MG';
startup de MGSTDBY;
Remarque :
Avec le broker la standby passe de facon automatique en mode recover managed.
9. Fast-start Failover (P438/439):
Moyen de basculer automatiquement selon des cas bien precis sans intervention :
Offline datafile
Corruption du dictionnaire , du controlfile
Logfile innacessible
Stuck archiver
Threshold,….
Commande pour mettre ca en place
Dgmgrl>Enable fast_start failover condition ‘XXXXX’;
Show fast_start failover;
On peut avoir en plus un observer :
L observer est sur un autre serveur .
VII. Clusterware to manage service applicatif
Ca fontionne pas encore .
http://uhesse.wordpress.com/2010/09/06/data-guard-oracle-restart-in-11gr2/
Aucun commentaire:
Enregistrer un commentaire