oracle 10g installation on raw file system and backup

[Shivkumar]

Hi,

Do we need to install control, data and redo logs of Oracle 10g on raw file system ?
How do we take backup of oracle using backup software like Dataprotector because oracle data resides on raw file system ?

I mean how come backup Software like Dataprotector work with raw file system for data or files ?

Thanks,
Shiv

[Steven E. Protter · Accepted solution]

Shalom Shiv,

First question: No. They can be on cooked filesystems.

Second Question: rman and or third party tools like Veritas Netbackup and Data Protector that use rman. Cold OS backups are out of the question on raw.

Third question: Answered by second answer. The third party utils use rman. They can use rman on raw or cooked filesystems.

SEP

[Indira Aramandla]

Hi Shivkumar,

One can backup oracle database (control files, data files redologs) all being on cooked file system or raw devices. It is not necessary that they should be defined on raw devices.

You can use Data protector to perform cold (off-line) hot (on-line) backups for the database on file-systems.

Dataprotector facilitates the configuration and integration of databases like (DB2, Informix, Lotus, MSSQL, MSVSSW, Mailbox, Oracle, SAP, SAPDB, Sybase).
You can use RMAN. Refer to metalink Note:77552.1 to see Configuring Dataprotector with RMAN.

You could also use Dataprotector alone using omnib (man omnib for details) without integration with RMAN, and backup the database (HOT or COLD). You will need to define the backup specification for the database, select the device and media pool and then schedule the backups through a script via cron.


Indira A

[Shivkumar]

Hi Indira,

I was told by some DBAs that control files, data files, and redologs should be installed on raw file systems for better performance and excellent i/o.

Now i am confused which is best option.

Regards,
Shiv

[A. Clay Stephenson]

1) Performance is not everything; I will take reliability over performance everytime as lonf as the performance is acceptable.
2) Backups are very important. With raw i/o you really have two options 1) dd 2) rman.
3) Performance isn't necessarily worse with cooked files. It's certainly acceptable.

[Indira Aramandla]

Hi Shivkumar,

The earlier question was about Oracle database backup "datafiles on raw devices" when using Dataprotector.

It was suggested that one can backup oracle database (control files, data files redologs) all being on cooked file system or raw devices. It is not necessary that they should be defined on raw devices.

Now we are talking about performance.......

Well as A. Clay mentioned, performance is good at the same time reliability is an necessary element. No matter how fast the database executes queries, if one cannot have a good backup and restore tool and procedures, then we end up in problems.

You can use raw devices but the backup option will be limited, Unless it is required (eg:- Parallel Processing) you can have the datafiles, redologs, control files on a cooked file system.


Indira A

[Indira Aramandla]

Hi Shivkumar,

Refer to Note:148462.1 in metalink "CHECKLIST-What else can influence the Performance of the Database" for details of what Other Factors Can Influence Database performance. This document lists several situations outside the database not commonly took in consideration when a database performance problem surfaces.

Also refer to the Note 37914.1 in metalink "Raw Devices and Oracle - 20 Common Questions and Answers" and Note 23037.1 "Using Unix Raw Partitions as Oracle Data File"


Indira A

[Faizer Jameel]

Hi
In the event you are using the online jfs then by setting mincache=direct you could bypass the system buffers and here too you gain performance on cooked file systems.
rgds
Faizer

[sysadm_1]

Hi Shiv,

We are using veritas netbackup to backup oracle 9i RAC raw volumes.In raw volume backups,you have to backup entire volume regardless of data size.
If you use oracle RMAN,the size of backup will be related to data size regardless of size of raw volumes/


-sysadm-

[Faizer Jameel]

Hi
In the event you are using the online jfs then by setting mincache=direct you could bypass the system buffers and here too you gain performance on cooked file systems.

Cold backups could not be done on raw file systems.

Dataprotector/Veritas could be used for backups of Oracle systems online [ on raw or cooked file systems] as rman is the undeline[primary] utilty that is being used for the backup purposes.
rgds
Faizer

[A. Clay Stephenson]

I'll also add one of the major problems with raw i/o --- especially if LVM or VxVM is bypassed but even if not. It is trivially easy to overwrite "unused" space. Documentation is extremely important when raw devices are in play.

The only time I would consider using raw i/o for databases is with Oracle RAC (or OPS) but now that the Cluster File System (CFS) is available for 11.23 then even Oracle RAC can use fully-cooked files.

[Chan 007]

Hi Shiv,

I totally agree with Clay's comments, also I am pretty sure that Oracle 9 and upwards will provide equal performance on RAW or cooked file system.

Data Protector will take a complete rlvol for backing up if you choose. e.g you have a /dev/vg01/rlv_oracle of 1GB, the it will take a complete snapshot of 1GB.

Whereas if you have a File system backed up, if your FS is only 700 MB /1GB then only 700MB will be back-uped.

Chan

Chan

[Hein van den Heuvel]

>> I was told by some DBAs that control files, data files, and redologs should be installed on raw file systems for better performance and excellent i/o.


At best it is a minor effect positive effect, at worst raw devices give a major negative effect (poor sga tuning).
I would not worry about raw untill you see the application is exhausting CPU and IO resources. It really needs to be a very IO intense applications to 'see' the potential benefit from raw. Don't start there unless you must (RAC, and then it is a moot question).

two remarks though.

1) "Cold OS backups are out of the question on raw." and similar comments...
I beg to differ. Of course you can use OS commands (notably dd) to make backups from raw devices. It's just more tricky to get it right. I used to use a (perl :-) script which would query Oracle for all the db file names and sizes. It would parse the sqlplust output and fork multiple (dd if=xxx bs=yyy count=zzz | gzip > backupfile) commands which would run concurrently. Real easy, once the hard work is done :-). Oh yeah... restore worked fine also, and it had an option to restore into a file based structure.

2) if performance becomes a problem and your application is IO intense, then you may find that the bulk of the critical IO is to REDO, UNDO and TEMP. Those are all easy re-created as RAW at any point in time.
Further more, they could be flipped to small & fiebased just before a backup, and recreated large and raw just after a backup.

Cheers,
Hein.