Creating a snapshot database on the XVC server

Thank you for your feedback!

This section describes the steps to mount the XVC snapshots of the PROD database volumes that were created in the previous section and re-create the snapshot copy of the PROD database that is based on these XVC snapshots of volumes. The steps that are provided in this section apply to both XVC databases, although we use XVC DEV as the example database.
Configuring multipath and udev rules
Perform all the steps below as the root user to detect the snapshot volumes, configure the multipath, and create the new udev rules:
1. To detect the snapshot volumes, rescan the SCSI bus on the XVC database server by running the following command:
[root#] rescan-scsi-bus.sh
1. Configure the multipath for the XVC snapshot devices by adding the following entries in the /etc/multipath.conf file:
multipath {
               wwid                    3514f0c5a4c200090
               alias                   C2_DATA11_DEV
        }
       multipath {
               wwid                    3514f0c5a4c200091
               alias                   C2_DATA12_DEV
        }
       multipath {
               wwid                    3514f0c5a4c200092
               alias                   C2_DATA13_DEV
        }
       multipath {
               wwid                    3514f0c5a4c200093
               alias                   C2_DATA14_DEV
        }
       multipath {
               wwid                    3514f0c5a4c200094
               alias                   C2_FRA_DEV
        }
       multipath {
               wwid                    3514f0c5a4c200095
               alias                   C2_REDO1_DEV
        }
       multipath {
               wwid                    3514f0c5a4c200096
               alias                   C2_REDO2_DEV
        }
Note: For each device, the WWID is ‘3<NAA Identifier >’.
In this example, the WWID of C2_DATA11_DEV is ‘3514f0c5a4c200090 and the NAA Identifier is ‘514f0c5a4c200090’.
1. Reload the multipath configuration file:
[root#] service multipathd reload
1. To set the proper Oracle permissions to the snapshot devices, create a new udev rules file /etc/udev/rules.d/60-oracle-asmdevices.rules and add an entry in it for each snapshot device as follows:
KERNEL=="dm-*", ENV{DM_NAME}=="C2_DATA11_DEV?", OWNER:="grid", GROUP:="asmadmin", MODE="0660"

KERNEL=="dm-*", ENV{DM_NAME}=="C2_DATA12_DEV?", OWNER:="grid", GROUP:="asmadmin", MODE="0660"

KERNEL=="dm-*", ENV{DM_NAME}=="C2_DATA13_DEV?", OWNER:="grid", GROUP:="asmadmin", MODE="0660"

KERNEL=="dm-*", ENV{DM_NAME}=="C2_DATA14_DEV?", OWNER:="grid", GROUP:="asmadmin", MODE="0660"

KERNEL=="dm-*", ENV{DM_NAME}=="C2_FRA_DEV?", OWNER:="grid", GROUP:="asmadmin", MODE="0660"

KERNEL=="dm-*", ENV{DM_NAME}=="C2_REDO1_DEV?", OWNER:="grid", GROUP:="asmadmin", MODE="0660"

KERNEL=="dm-*", ENV{DM_NAME}=="C2_REDO2_DEV?", OWNER:="grid", GROUP:="asmadmin", MODE="0660"
1. Reload the udev rules:
[root#] udevadm trigger

Renaming and mounting ASM disk groups and ASM disks
1. Log in to the XVC server as the grid user (the owner of the Oracle ASM instance).
1. Rename the original PROD ASM disk group name for the DEV XVC database:
grid$ renamedg dgname=REDO2 newdgname=REDO2_DEV asm_diskstring=/dev/mapper/* verbose=TRUE
grid$ renamedg dgname=REDO1 newdgname=REDO1_DEV asm_diskstring=/dev/mapper/* verbose=TRUE
grid$ renamedg dgname=DATA newdgname=DATA_DEV asm_diskstring=/dev/mapper/* verbose=TRUE
grid$ renamedg dgname=FRA newdgname=FRA_DEV asm_diskstring=/dev/mapper/* verbose=TRUE
1. Mount the renamed ASM disk groups to the ASM instance in restricted mode.
grid$ asmcmd mount --restrict FRA_DEV
grid$ asmcmd mount --restrict REDO1_DEV
grid$ asmcmd mount --restrict REDO2_DEV
grid$ asmcmd mount --restrict DATA_DEV
1. Rename ASM disks for the snapshot volumes. As grid user, log in to the ASM instance as sysasm and execute the following alter diskgroup SQL commands:
grid$ sqlplus / as sysasm
SQL> alter diskgroup REDO2_DEV rename disks all;
SQL> alter diskgroup REDO1_DEV rename disks all;
SQL> alter diskgroup DATA_DEV rename disks all;
SQL> alter diskgroup FRA_DEV rename disks all;
1. Dismount and remount the ASM disk groups by running the following SQL commands:
grid$ sqlplus / as sysasm
SQL> alter diskgroup REDO2_DEV dismount;
SQL> alter diskgroup REDO1_DEV dismount;
SQL> alter diskgroup DATA_DEV dismount;
SQL> alter diskgroup FRA_DEV dismount;
SQL> alter diskgroup REDO2_DEV mount;
SQL> alter diskgroup REDO1_DEV mount;
SQL> alter diskgroup DATA_DEV mount;
SQL> alter diskgroup FRA_DEV mount;
1. Check the status of ASM diskgroups and ASM disks of the snapshot volumes:
grid$ sqlplus / as sysasm
SQL> SELECT d.path, d.name disk_name, g.name group_name,
                         g.state mount_state
FROM v$asm_disk d, v$asm_diskgroup g                       WHERE d.group_number = g.group_number and d.path like '%_%DEV%';
PATH                         DISK_NAME         GROUP_NAME
--------------------------- ---------------   -----------
MOUNT_STATE
-----------
/dev/mapper/C2_REDO2_DEV1    REDO2_DEV_0000    REDO2_DEV              MOUNTED
/dev/mapper/C2_REDO1_DEV1    REDO1_DEV_0000    REDO1_DEV              MOUNTED
/dev/mapper/C2_FRA_DEV1      FRA_DEV_0000      FRA_DEV                   MOUNTED
/dev/mapper/C2_DATA14_DEV1   DATA_DEV_0003     DATA_DEV                  MOUNTED
/dev/mapper/C2_DATA13_DEV1   DATA_DEV_0002     DATA_DEV                  MOUNTED
/dev/mapper/C2_DATA12_DEV1   DATA_DEV_0001     DATA_DEV                  MOUNTED
/dev/mapper/C2_DATA11_DEV1   DATA_DEV_0000     DATA_DEV                  MOUNTED
/dev/mapper/C2_FRA_DEV1      FRA_DEV_0000      FRA_DEV                      MOUNTED
The four new diskgroups are: DATA_DEV, REDO1_DEV, REDO2_DEV and FRA_DEV.
Mounting the snapshot database
1. Prepare the new snapshot database environments as follows:
  1. Create the directory for the database audit file, which is specified by audit_file_dest='/u01/app/oracle/admin/dbdev/adump'
  2. Log in to the PROD database as sysdba and create a pfile from the PROD database:
SQL> create pfile=’/home/oracle/pfile_dbprod.ora’ from spfile;
1. Use the secure copy command scp to copy /home/oracle/pfile_dbprod.ora to the XVC server.
2. On the XVC database server, edit the pfile_dbprod.ora file by modifying the following database initialization parameters with the new values:
*.cluster_database=false
*db_create_file_dest=’DATA_DEV’
*.thread=1
*.control_files='+DATA_DEV/DBCAP/CONTROLFILE/current.262.986035347','+FRA_DEV/DBCAP/CONTROLFILE/current.296.986035347'
* .db_recovery_file_dest='+FRA_DEV
1. Mount the snapshot database as oracle user:
oracle$ sqlplus / as sysdba
SQL> startup nomount pfile=pfile_dbcap_new.ora
ORACLE instance started.
Total System Global Area 2.1475E+10 bytes
Fixed Size                       22787792 bytes
Variable Size                 7314868528 bytes
Database Buffers         1.4093E+10 bytes
Redo Buffers               44318720 bytes
SQL> alter database mount;
Database altered.
Changing the database file names
1. Log in to the newly mounted snapshot database and run the following queries to find the current data files and redo log file names:
SQL> select name from v$datafile;
SQL> select member from v$logfile;
1. Based on the results of the queries shown in step 1, we can generate the alter database rename database files commands to change the location of the database and redo log files to adapt to the new ASM diskgroup name and location changes. Run these commands in sqlplus . For example:
SQL> alter database rename file '+DATA/DBCAP/DATAFILE/system.258.986035241' to '+DATA_DEV/DBCAP/DATAFILE/system.258.986035241';
SQL> alter database rename file '+DATA/DBCAP/DATAFILE/tpcctab.268.986040989' to '+DATA_DEV/DBCAP/DATAFILE/tpcctab.268.986040989';
SQL> alter database rename file '+DATA/DBCAP/DATAFILE/sysaux.259.986035277' to '+DATA_DEV/DBCAP/DATAFILE/sysaux.259.986035277';
SQL> alter database rename file '+DATA/DBCAP/DATAFILE/undotbs1.260.986035293' to '+DATA_DEV/DBCAP/DATAFILE/undotbs1.260.986035293';
SQL> alter database rename file '+DATA/DBCAP/DATAFILE/undotbs2.266.986035373' to '+DATA_DEV/DBCAP/DATAFILE/undotbs2.266.986035373';
Database altered.
SQL> alter database rename file '+DATA/DBCAP/DATAFILE/users.261.986035293' to '+DATA_DEV/DBCAP/DATAFILE/users.261.986035293';
Rename REDO logs files:
SQL> alter database rename file '+REDO1/DBCAP/ONLINELOG/group_5.262.986037407' to '+REDO1_DEV/DBCAP/ONLINELOG/group_5.262.986037407';
SQL> alter database rename file '+REDO2/DBCAP/ONLINELOG/group_6.263.986037413' to '+REDO2_DEV/DBCAP/ONLINELOG/group_6.263.986037413';
SQL> alter database rename file '+REDO1/DBCAP/ONLINELOG/group_7.263.986037417' to '+REDO1_DEV/DBCAP/ONLINELOG/group_7.263.986037417';
SQL> alter database rename file '+REDO2/DBCAP/ONLINELOG/group_8.261.986037421' to '+REDO2_DEV/DBCAP/ONLINELOG/group_8.261.986037421';
SQL> alter database rename file '+REDO1/DBCAP/ONLINELOG/group_9.258.986037557' to '+REDO1_DEV/DBCAP/ONLINELOG/group_9.258.986037557';
SQL> alter database rename file '+REDO2/DBCAP/ONLINELOG/group_10.262.986037565' to '+REDO2_DEV/DBCAP/ONLINELOG/group_10.262.986037565';
SQL> alter database rename file '+REDO1/DBCAP/ONLINELOG/group_11.259.986037571' to '+REDO1_DEV/DBCAP/ONLINELOG/group_11.259.986037571';
SQL> alter database rename file '+REDO2/DBCAP/ONLINELOG/group_12.258.986037575' to '+REDO2_DEV/DBCAP/ONLINELOG/group_12.258.986037575';
1. As oracle user, log in to the database as sysdba to open the database:
sqlplus / as sysdba
SQL> alter database open;
1. Change the database name and DBID:
  1. Shutdown the database:
SQL> shutdown immediate;
1. Reset the ORACLE_SID to a new database instance name (dbcdev in this example)
oracle$ export ORACLE_SID=dbcdev
1. Run the Oracle DBID utility ‘nid' command to change the existing database name (dbcap in this instance) to a new database name(dbcdev in this instance)
oracle$ which nid
/u01/app/oracle/product/12.2.0/dbhome_2/bin/nid
oracle $ export ORACLE_SID=dbcdev
oracle $ /u01/app/oracle/product/12.2.0/dbhome_2/bin/nid TARGET=/ DBNAME=dbcdev
DBNEWID: Release 12.2.0.1.0 - Production on Tue Sep 25 16:02:41 2018
Copyright (c) 1982, 2017, Oracle and/or its affiliates. All rights reserved.
Connected to database DBCAP (DBID=4233973522)
Connected to server version 12.2.0
Control Files in database:
    +DATA_DEV/DBCAP/CONTROLFILE/current.262.986035347
    +FRA_DEV/DBCAP/CONTROLFILE/current.296.986035347
Change database ID and database name DBCAP to DBCDEV? (Y/[N]) => Y
Proceeding with operation
Changing database ID from 4233973522 to 3638836129
Changing database name from DBCAP to DBCDEV
    Control File +DATA_DEV/DBCAP/CONTROLFILE/current.262.986035347 - modified
    Control File +FRA_DEV/DBCAP/CONTROLFILE/current.296.986035347 - modified
    Datafile +DATA_DEV/DBCAP/DATAFILE/system.258.98603524 - dbid changed, wrote new name
    Datafile +DATA_DEV/DBCAP/DATAFILE/tpcctab.268.98604098 - dbid changed, wrote new name
    Datafile +DATA_DEV/DBCAP/DATAFILE/sysaux.259.98603527 - dbid changed, wrote new name
    Datafile +DATA_DEV/DBCAP/DATAFILE/undotbs1.260.98603529 - dbid changed, wrote new name
    Datafile +DATA_DEV/DBCAP/DATAFILE/undotbs2.266.98603537 - dbid changed, wrote new name
    Datafile +DATA_DEV/DBCAP/DATAFILE/users.261.98603529 - dbid changed, wrote new name
    Datafile +DATA_DEV/DBCAP/TEMPFILE/temp.267.98772752 - dbid changed, wrote new name
    Control File +DATA_DEV/DBCAP/CONTROLFILE/current.262.986035347 - dbid changed, wrote new name
    Control File +FRA_DEV/DBCAP/CONTROLFILE/current.296.986035347 - dbid changed, wrote new name
    Instance shut down
Database name changed to DBCDEV.
Modify parameter file and generate a new password file before restarting.
Database ID for database DBCDEV changed to 3638836129.
All previous backups and archived redo logs for this database are unusable.
Database is not aware of previous backups and archived logs in Recovery Area.
Database has been shutdown, open database with RESETLOGS option.
Successfully changed database name and ID.
DBNEWID - Completed successfully.
1. Update the following parameter in the pfile pfile_dbcap_new.ora to reflect the new database name:
*.db_name='dbcdev'
1. Open the database with resetlogs as shown in the following figure:
1. Drop the old redo logs. Recreate the new redo logs by running these commands as sysdba:
SQL> ALTER DATABASE ADD LOGFILE GROUP 13 ('+REDO1_DEV') size 1G;
SQL> ALTER DATABASE ADD LOGFILE GROUP 14 ('+REDO2_DEV') size 1G;
SQL> ALTER DATABASE ADD LOGFILE GROUP 15 ('+REDO1_DEV') size 1G;
SQL> ALTER DATABASE ADD LOGFILE GROUP 16 ('+REDO2_DEV') size 1G;
Recreate spfile, redos, and temp file
1. To create an spfile from a pfile, run this command as sysdba:
SQL> create spfile='+DATA_DEV/dbcdev/dbcdevspfile.ora'
from pfile='/home/oracle/pfile_dbcap_new.ora’
1. Edit the $ORACLE_HOME/dbs/initdbcdev.ora file and update the spfile location in it:
spfile='+DATA_DEV/dbcdev/dbcdevspfile.ora'
1. Recreate the TEMP tablespace:
SQL> CREATE TEMPORARY TABLESPACE TEMP TEMPFILE ‘+TEMP' size 50G;
1. Restart the snapshot database:
oracle$ export ORACLE_SID=dbcdev
oracle$ sqlplus / as sysdba
SQL> shutdown immediate;
SQL> startup;
The XVC snapshot database is created and running. Change other database initialization parameters, such as SGA_TARGET, as needed.

Your Browser is Out of Date

Creating a snapshot database on the XVC server

Configuring multipath and udev rules

Renaming and mounting ASM disk groups and ASM disks

Mounting the snapshot database

Changing the database file names

Recreate spfile, redos, and temp file