ASM migration

I have a LINUX server with an ASM instance up and running and a 10.2 database instance up and running that is presently not using the ASM, each in different ORACLE_HOME.
I also have a 10.2 database on a different (Windows) server that I want to migrate from the Windows server to the LINUX server and create the tablespaces (datafiles) from the Windows server within the ASM diskgroup on the LINUX server .
I am thinking all I need to do is change the init.ora parameters on the LINUX ORCL database to reference the ASM diskgroup (+DATA) for datafile creation, then run a script on ORCL to create the tablespaces in ASM on the LINUX server, then take an export from the Windows server and import it into the LINUX server.
Any comments? WIll this work?

and create the tablespaces (datafiles) from the Windows server within the ASM diskgroup on the LINUX server . You mean you want to migrate COMPLETELY the instance on windows to LINUX right? If yes, that should be fine.
I am thinking all I need to do is change the init.ora parameters on the LINUX ORCL database to reference the ASM diskgroup (+DATA) for datafile creation, then run a script on ORCL to create the tablespaces in ASM on the LINUX server, then take an export from the Windows server and import it into the LINUX server.Are you referring to - db_create_file_dest* init parameters? If yes, that should be fine. As such this is optional, you can create tablespaces on the database (where the ASM instance is running) without these parameters by just indicated datafile as '+ASM/....'. BTW, with this you will have a mix of ASM and non-ASM datafiles, since you are planning on creating ASM datafiles in a database which already has non-asm files. You can convert these datafiles into ASM using RMAN....
The overall/high-level plan looks fine...Good Luck.
Chandra

Similar Messages

Filesystem to ASM migration in streams setup

Hi All,
We have a Bi Directional replication setup between databases A and B.
Database A [ 10.2.0.4 ] is already in ASM
Database B [ 10.2.0.4 ] is in Filesystem and we migrated it to ASM now..
Steps taken...
1. Stopped Bi Directional replication
2. Migrated Database B to ASM [ this includes Temp and redo log files - we dropped and recreated the INACTIVE redo log ] All datafiles, temp and redo, archive log dest was changes to ASM disk group...
3. Started the Bi directional streams
4. Updates from Database A coming to B without any issues and apply process is updating the database B tables.
5. But Database A apply process is not applying the changes made in Database B.
All the capture, propogation and apply process are running on both sides.
Please advise.
Thanks & Regards,
Rakesh
Edited by: Oracleappsdba on Jan 14, 2011 9:28 AM

Hi All,
We found that the propagation is ending in error with
ORA-00604: error occurred at recursive SQL level 4
ORA-12154: TNS:could not resolve the connect identifier specified
ORA-06512: at "SYS.DBMS_AQADM_SYS", line 1087
ORA-06512: at "SYS.DBMS_AQADM_SYS", line 7639
ORA-06512: at "SYS.DBMS_AQADM", line 631
ORA-06512: at line 1
This is happening only after the ASM migration. We can able to connect to the target database with the db link from the user directly without any issues. Its erroring only through propagation process.
Please advise.
Thanks & regards,
Rakesh

Single 10g Instance to RAC - without ASM migration

Is it possible to comment the following steps to convert existing 10.1.0.5 single instance database to RAC? We're using OCFS2 and it is not possible to perform ASM migration.
1. Copy the existing $ORACLE_HOME/dbs/init<SID1>.ora to
$ORACLE_HOME/dbs/init<db_name>.ora. Add the following parameters to
$ORACLE_HOME/dbs/init<db_name>.ora:
*.cluster_database = TRUE
*.cluster_database_instances = 2
     *.undo_management=AUTO (Add if you don't have it )
<SID1>.undo_tablespace=undotbs (undo tablespace which already exists)
     <SID1>.instance_name=RAC1
     <SID1>.instance_number=1
     <SID1>.thread=1
<SID1>.local_listener=LISTENER_RAC1
where LISTENER_RAC1 is an entry in the tnsnames.ora file like:
LISTENER_RAC1 =
(ADDRESS = (PROTOCOL = TCP)(HOST = <node1>)(PORT = 1521))
Keep only one line in $ORACLE_HOME/dbs/init<SID1>.ora:
ifile=$ORACLE_HOME/dbs/init<db_name>.ora
You could also create a common spfile from this pfile and add a line
like spfile=$ORACLE_HOME/dbs/spfile<db_name>.ora in each init<SIDn>.ora
file. For more details about how to do this, please refer to Note 136327.1.
2. Open your database and run $ORACLE_HOME/rdbms/admin/catclust.sql to create
cluster database specific views within the existing instance.
3. Recreate control file if you defined maxinstances to be 1 when you created
the single instance database.
To check your current setting of maxinstances, run the following command
while the database is mounted or open and connected as a user with DBA
privileges:
% sqlplus /nolog
SQL> connect / as sysdba
SQL> startup mount
SQL> alter database backup controlfile to trace;
The trace file is located in udump directory. Check the maxinstance value
in the CREATE CONTROLFILE statement. Please refer to Note 118931.1
Recreating the Controlfile in RAC and OPS
4. Add instance specific parameters in the init<db_name>.ora for the second
instance on the second node and set appropriate values for it:
     *** Names may need to be modified
     <SID2>.instance_name=RAC2
     <SID2>.instance_number=2
     <SID2>.local_listener=LISTENER_RAC2
     <SID2>.thread=2
     <SID2>.undo_tablespace=UNDOTBS2
<SID2>.cluster_database = TRUE
<SID2>.cluster_database_instances = 2
where LISTENER_RAC2 is an entry in the tnsnames.ora file like:
LISTENER_RAC2 =
(ADDRESS = (PROTOCOL = TCP)(HOST = <node2>)(PORT = 1521))
5. From the first instance, mount the database and run the following command:
     *** Path names, file names, and sizes will need to be modified
alter database
add logfile thread 2
group 3 ('/dev/RAC/redo2_01_100.dbf') size 100M,
group 4 ('/dev/RAC/redo2_02_100.dbf') size 100M;
alter database enable public thread 2;
6. Create a second Undo Tablespace from the existing instance:
     *** Path names, file names, and sizes will need to be modified
CREATE UNDO TABLESPACE UNDOTBS2 DATAFILE
     '/dev/RAC/undotbs_02_210.dbf' SIZE 200M ;
7. Set ORACLE_SID and ORACLE_HOME environment variables on the second node.
8. Start the second Instance.

In order not to affect the current home on node1, another user oracle2 is created which belong to racinstall group. On node2, which doesn’t have any oracle home installed, CRS will be installed so that node2 will become the primary node.
Been looking into this CRS for sometime as the installation has failed due to OCFS2 kernel panic upon running root.sh on the second node. Performed the cleanup as described Doc ID: Note:239998.1 (10g RAC: How to Clean Up After a Failed CRS Install), but now, the installation is failing and giving the error “ File not found /ClusterCRS/install/rootadd”. This message didn’t appear the first attempt which could indicate that the cleanup described in the note was not sufficient.

Single node file system to 3 node rac and asm migration

hi,
we have several utl_file and external table applications running on 10.2 single node veritas file system. and we want to migrate to 3 node RAC ASM environment. what is the best practices in order to succeed this migration during this migration. thanks.

1. Patch to 10.2.0.3 or 10.2.0.4 if not already there.
2. Dump Veritas from any future consideration.
3. Build and validate the new RAC environment and then plug in your data using transportable tablespaces.
Do not expect the first part of step 3 to work perfectly the first time if you do not have experience building RAC clusters.
This means have appropriate hardware in place for perfecting your skills.
Be sure, too, that you are not trying to do this with blade or 1U servers. You need a minimum of 2U servers to be able
to plug in sufficient hardware to have redundant paths to storage and for cache fusion and public access (a minimum of 6 ports).
And don't let any network admin try to convince you that they can virtualize the network paths: They can not do so successfully
for RAC.

Filesystem to asm migration

I am running 10.2.0.3 on Oracle linux "OEL4" with ext3 filesystem.
I just receive a brand new server and i want to move my actual database into it.
I want to run my 10.2.0.3 database on asm 11g into my new server.
Once asm testing is over, il migrate my 10.2.0.3 to 11g database.
I saw documentation on migrating filesystem to asm, but it seems that the example given is where a database with the filesystem resides on the same machine.
Is there documentation somewhere on how to migrate my database into asm, but not on the same machine?
Or do you know the steps i need to perform to do this?

What makes it complex is the following:
1- Lack of space on my new server to run both filesystem and migrate to asm.
2- 32 bits to 64 bits migration.
3-The database is 1.1T in size, a little to much for an export/import
Here is what I am going to do. I will try that today.
I have 3 servers.
Server A: I have my actual database on OEL4 32 bits with oracle 10g on filesystem
Server B: I will install OEL5 64 bits running 10g filesystem
Server C: I will install OEL5 64 bits running 10g database with ASM
Steps:
1) Restore a rman backup of my database on server B
2) migrate the database on server B to 64 bits
3) NFS mount server B from Server C
Mount the database
Perform a Rman backup with location = ASM
Switch database to copy;

ASM Migrate Archive Redo Logs

Hi all,
Is there any way to migrate existing archived redo logs to the ASM diskgroup?
Regards,
Vijayaraghavan K

Thanks..
I have copied the spfile from the last backup to the asmdiskgroup by using below command
RESTORE SPFILE TO '+DATA/spfilesid.ora';
When i try to start the database it is picking the spfile from file system alone. It is possible to rmap the spfile to the diskgroup?
Regards,
Vijayarghavan K

ASM migration issues-

Hello All,
I am new to ASM & RAC. I have a 10g Database. I have installed 11gR2 grid infrastructure and configured ASM on the same oracle home.
My 10g database Oracle home owner is - oracle
My 11gR2 grid Oracle home owner is - oragird
But both share common groups. When i am trying to migrate Oracle 10g database into 11gR2 ASM. I am getting below error. Any ideas/help is very much appreciated.
RMAN> RESTORE CONTROLFILE FROM '/qsynergy1/oradata/cntrl01.dbf';
Starting restore at 19-APR-11
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=534 devtype=DISK
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of restore command at 04/19/2011 11:00:46
ORA-19504: failed to create file "+EBS_DATA"
ORA-17502: ksfdcre:4 Failed to create file +EBS_DATA
ORA-15001: diskgroup "EBS_DATA" does not exist or is not mounted
ORA-15055: unable to connect to ASM instance
ORA-15055: unable to connect to ASM instance
ORA-19600: input file is control file (/qsynergy1/oradata/cntrl01.dbf)
ORA-19601: output file is control file (+EBS_DATA)
thank you
David

Hi,
have you changed the groupid of the 10g oracle executable?
You have to run "setasmgidwrap" on the 10g oracle executable, that the ownership information is set correctly to work with 11gR2 GI with different users.
See here MoS Note:
Pre 11.2 Database Issues in 11gR2 Grid Infrastructure Environment (Doc ID 948456.1)
Regards
Sebastian

File system to ASM Migration

Currently we have unix file system for storage. We are planning to move to ASM. Based on documentation RMAN and physical standby could be the best way to migrate.
1. Is expdp/impdp an option?
2. Given an ASM env. is RMAN the only backup solution?
Thank you.

1. Yep. Doing it like that also lets you reorganize your tablespaces if you so desire. -
Let's say if I have 100 tablespaces.
a. create new tablespaces with diskgroups
b. Use impdp to import the data
2. RMAN main method; but could do something re breaking a mirror of a LUN
a. What do you mean by main method?
b. Can I not have online backups like snap clone (and/or offline the tblspace and move the datafiles?)
Appreciate your help.

Datapump TTS from filesystem to ASM migration

Hi Experts,
I have to move the 1.5 TB of data which is on filesystem running on Solaris SPARC (single instance) to Linux x86 RAC (10.2.0.4) with ASM. I am considering the Transportable tablespaces option using datapump. But I have done the filesystem to filesystem datapump for TTS. Is there any way that I can use datapump TTS for filesystem to ASM directly ? Any suggestion and comments please.
Thanks
Shashi

I'm not sure what your question is. Data Pump works with ASM and supports transportable. You would have to copy your data files from your file system to your asm file system, then do the export on your source, copy the dumpfile to your asm target and then run data pump. If you have more specific questions, please list them.
Hope this helps.
Thanks
Dean

Prerequisites required ASM Migration

Hi ,
Could u please let me know "The prerequisites required before going ahead with the ASM implementation".
Any best practices for disk group confirguration ( I mean how many groups for good performance)
Details of our database:
1) Oracle version : 11.2.0.3
2) OS : Sun solaris 10
3) DB size : 1 TB
I know ASM is part of the Grid Infrastructure. But the current setup was handover from different company and we planned to implement ASM concept here.
Could u please help me out.
Regards,
Steven

It depends. I'm sure that's what they would tell you in the Oracle training class. :)
For starters Oracle recommends two diskgroups named DATA and FRA. If you are using RAC then you will probably create a third diskgroup named CRS.
The redundancy level depends on your storage. If the storage array is providing RAID protection, then you can configure the ASM diskgroups with External Redundancy. But, this is not always the case. You might want to mirror between an external SAN and in internal ioDrive using ASM Normal Redundancy and then set the parameter asm_preferred_read_failure_groups to ensure all reads come from the flash drive. So you see there are countless possibilities that make us say "it depends".
In larger organizations with Enterprise Storage it is rarely so simple. Recall that a diskgroup is a collection of like storage. If you have LUNs of different sizes and performance levels, then you should group the LUNs according to those attributes. Each grouping becomes a diskgroup. For example, never mix RAID 5 LUNs from a Clariion with RAID 5 LUNs from a Symmetrix, and never mix any of those with RAID 10 LUNs. Also, within a diskgroup you want all storage of the same size due to the impact on striping and mirroring. This is all discussed in the Oracle on-line documentation.
If your storage team carves up the array into different classes of storage, then you will probably want to create separate ASM diskgroups. The I/O profiles are completely different for log files versus datafiles, and a well trained storage admin knows how to address each case. So now, if the storage admin gives you high performance LUNs designed for redo logs, lower performance LUNs marked for archivelogs, and a third set of LUNs marked for datafiles, then you will want to create three separate ASM diskgroups named DATA, REDO, and FRA to isolate the storage based on performance characteristics.
Often times with Enterprise Storage you cannot avoid I/O conflicts, so there's no point in separating Oracle's files into their own diskgroups (data, index, redo, etc.) Back on the array you will find the stripes overlay each other. And the array internally ties the storage devices together with either copper or fiber loops, so the I/O's are getting mixed and competing for resources. All I am saying is talk to your storage team and see if they think it will make any difference to try separating tables and indexes into different files given the way they setup the array.
For a small 1 TB database like yours I generally use two ioDrive 1.2 TB PCIe flash memory cards and create one ASM diskgroup named DATA with Normal Redundancy. Bang, done. RAID 10 protection at 5x the performance of a well tuned SAN, and I don't have to wait two weeks for the Enterprise Storage team to carve up LUNs. This takes a great load off the SAN and helps the other enterprise users with their performance issues, and frees up capacity thereby extending the life of the enterprise storage.
The above "answers" are a very generic introduction to planning ASM storage. If you have more specific questions, let us know.
-R

Migrating a database to ASM.

Hi all,
I have an environment for test purposes running 10.2.0.1.0 database on a Linux CentOS 4 box.
Now what I´m trying to accomplish is migrate this database to ASM.
Some adittional info:
$ ps -ef | grep pmon
oracle 4063 1 0 07:27 ? 00:00:00 ora_pmon_APPS102
502 6659 1 0 07:34 ? 00:00:00 asm_pmon_+ASM
$ lsnrctl status
Services Summary...
Service "+ASM" has 1 instance(s).
Instance "+ASM", status BLOCKED, has 1 handler(s) for this service...
Service "APPS102" has 1 instance(s).
Instance "APPS102", status READY, has 1 handler(s) for this service...
What I´m doing is to log into database through EM 10g console as sys user, then I go to the Administration page from where I can run the ASM migration wizard.
Then, after inform my SYS password and my Host account/Password, I´m facing below error:
Listener refused the connection with the following error: ORA-12528, TNS:listener: all appropriate instances are blocking new connections The Connection descriptor used by the client was: (description=(address=(host=VM_PEGASUS.localdomain)(protocol=tcp)(port=1521))(connect_data=(sid=+ASM)(server=dedicated)))
I´ve already stopped and started the listener and restarted both instances. Did not work, same situation.
Well, any help will be very appreciated.
Thanks,
Marcello M.

Hi,
Please see (How To Use Automatic Shared Memory Management (ASMM) In Oracle10g) Doc ID: Note:295626.1 on Oracle Metalink.
The main goal of this document is to show you how to put in place the ASMM facility introduced with Oracle10g, how to configure it, and also some examples on how to monitor its activity.
Adith

Database migration filesystem to asm

Hi
Let say I have a file system database. (10g on linux)
In other server, I installed fresh oracle 11.2.0.2 with asm on linux
What is the best way to migrate this database to this new server?

In the asm migration documents "rman BACKUP AS COPY" command is recommended.
However, ASM will be on different server, it will not be on local server.
What should be the correct steps to migrate the database to different server with ASM ?
Also, as I mentioned in my first post, I am going higher version of oracle.

Oracle RAC Storage Migration - HP EVA to HP EVA

Hi,
We are in the process of migrating from an old HP EVA4000 to a new HP EVA6400. We have migrated most of the systems that had storage presented from the EVA4000 with the exception of our Oracle RAC databases and wanted to check to see if anyone knew of any gotchas before we proceeded.
We are using CA to replicate the LUNs that are presented to the Oracle cluster. My plan is to shut down the database and all associated RAC services, disable CRS then halt the servers. I will then present the CA replicated LUNs to the servers and initiate a failover. I will then restart the servers and check that the LUNs have appeared from the 6400 before manually restarting CRS and the databases. I will then re-enable CRS before rebooting again to check that everything comes up automatically.
Does this sound correct? Has anyone done this before?
Thanks in advance.
Phil

Seems reasonable, but can't be certain unless you test. Make sure you backup OCR, VOTE, Database before you change the storage allocation.
How big is the database ? How long will your downtime be ?
It may be cleaner to just reinstall CRS and restore the database from backup ?
Are you using ASM ? If so then you could just add new devices to ASM, migrate the data, then remove the old devices.

Create a DR for a ASM instance

Hi All,
We are planning for the ASM migration.
Database - 10.2.0.4
Currently we are using the SRDF process to replicate the data in to the DR site....
After migrating the database to ASM instance, is Data Guard the only option for DR or we have any other methods where we can copy the raw device directly...
Please advice...
Thanks & Regards,
Rakesh

Yes,Dataguard is the best option..
thanks,
baskar.l

Grid control mgmt aganet question...

I installed a new 2 node cluster 10.2.0.3 on linux Itanium and I setup the db as a standby database so that we can test our dtata gurad switchover which we are using to facilitate our ASM migration from our production database.
We successfully switched roles and the new cluster came up as a primary. While the database was up and running as the primary database I installed the grid control mgmt aganets and that all went well and I could loging to the grid control mgmt server and administer the clustered database.
I since have dropped the database and have recreated it as a standby which is now in sync from our prod database. The grid control mgmt agents have been stopped since they could not connect whcile the database is running in a syandby role.
We plan top switch roles of the databases and the new cluster will be coming online as our primary over the weekend.
My quesion is? Will I need to reinstall the grid control agents or can I simply just start them up once the database is up and running as primary again? Is there data in the sysman repository that will be missing that will warrant a reinstall of the grid agaents since this database is not going to be the same database that was in existence at the time when the agents had been installed?
Thanks very much.

It is something that works. It will be best for you to give it a go on your own. If you really want do a lot of practice, then pickup the related Step-by-steps documents and start doing it. You cannot start if you spend time contemplating or having a lot of doubt about it.
Oracle has a lot of pages dedicated to VM and Grid Control. I would suggest that you read up on it.
<br>
Oracle Database FAQs
</br>

ASM migration

Similar Messages

Maybe you are looking for