OPS/RAC Vs Raw Device/File System

Hello, firts to all, I want to know what is OPS/RAC and Raw Device/File System.
And then know the diferens between both.
I apologize for my english.
Thanks.

OPS is Oracle parallel server was available up to Oracle 8i and RAC is Real application cluster is available from Oracle 9i.
Raw device is just presenting the disk to server and using it. File system is presenting disk more transparent and managable form using OS or third party software.
You need to refer respective documents to know them in detail.
Ashok

Similar Messages

RAW DEVICE와 FILE SYSTEM 간에 오라클 데이터화일 이동하기

제품 : ORACLE SERVER
작성날짜 : 1999-11-30
RAW DEVICE와 FILE SYSTEM 간에 오라클 데이터화일 이동하기
======================================================
유닉스 명령이 dd를 이용하여 오라클 데이터화일을 Unix File System과 Raw
Device 간에 이동할 수 있으나, 플랫폼 별 Raw Device의 특성으로 인하여 주의할
점이 있다. 만일 잘못된 이동으로 인하여 데이터베이스가 기동하지 못하면,
ORA-7366 (sfifi: invalid file, file does not have valid header block.)
에러가 발생할 수 있다.
예를 들어 Digital unix의 경우는 raw device에 64k OS 헤더가 필요하므로 dd
명령어 옵션 중 iseek와 oseek를 활용하여야 한다.
다음은 예제를 통하여 데이터화일을 Raw Device에서 Unix File System으로
이동하는 절차이다.
(운영 현황)
- 현재의 위치: /dev/rdsk/c0t15d0s7
- 이동할 위치: /oracle/file/system.dbf
- 화일의 크기: 488636416 bytes <--- V$DATAFILE.BYTES column 값!
- DB_BLOCK_SIZE: 2048 bytes
(준비 단계)
1. Oracle 블럭 수 계산:
BYTES / DB_BLOCK_SIZE = 488636416 / 2048 = 238592 (블럭)
2. O/S file header 블럭 추가:
238592 + 1 = 238593 (블럭)
: "ls -l /oracle/file/system.dbf" 명령으로 확인 가능하며, O/S file
header는 1블럭으로 항상 일정함.
3. Raw Device OS header 블럭 계산:
64K / DB_BLOCK_SIZE = 65536 / 2048 = 32 (블럭)
: 사용할 dd 명령어 중 블럭의 크기를 DB_BLOCK_SIZE(2048바이트)로 할
예정이므로 2048로 나누어야 함.
(명령어 형식)
$ dd if=<raw device> of=<UFS file> bs=<oracle blocksize>
iseek=<blocks to skip> count=<total count>
(명령어 수행 절차)
(1) SVRMGR> STARTUP MOUNT
(2) SVRMGR> !dd if=/dev/rdsk/c0t15d0s7 of=/oracle/file/system.dbf
bs=2048 iseek=32 count=238593
(3) SVRMGR> ALTER DATABASE RENAME FILE '/dev/rdsk/c0t15d0s7' TO
'/oracle/file/system.dbf';
(4) SVRMGR> ALTER DATABASE OPEN;
========================================================================
반대로 Unix File System에서 Raw Device로 이동하는 명령어 형식은 아래와 같다.
(명령어 형식)
$ dd if=<UFS file> of=<raw device> bs=<oracle blocksize> \
oseek=<blocks to skip> count=<total count>
유사한 방식으로 데이타 화일 뿐 아니라 리두 로그 화일도 이동할 수 있다.
[주의] Raw Device의 블럭 헤더의 크기는 OS마다 다를 수 있으므로 플랫폼
벤더사에 미리 확인한 후 작업하여야 한다.

RAC on RAW device

1) Hi I am configuring rac on raw device 9i on aix5.Whether i can create a datafile morethan 1 gb in raw device.Can anybody give one example with creating a tablespace with a raw device configuration.The steps have to follow
2) why we cannot store archivelogs on rawdevices
with regards
ramya

1)Whether we can store RMAN backup on RAW devices .
Answer : Yes, you can, just create the raw partition and point out the destination when you use the rman command. or you can take a backup to the local file system first and then move to raw device.
2)whether we can do the export on RAW devices
Answer : Same as No.1
3) Whether we can create a 2gb datafile on RAW devices
Answer : Yes, you can.
I hope it would be helpful to you.

Browsing other devices' file system.

I figured out that a Nokia 9500 does not act as a PDA when connected with other devices. While from a PDA u can see a Nokia 9500 file system, from a Nokia 9500 u can't see the other device file system and browse it. Am I missing something?

It's the PDA that's missing something... security rules and regulations! You have never been able to browse the file system of another device from Nokia phones that I know of.
This is something about other manufacturers that really disturbs me... does nobody else think that potentially this could lead to serious security issues??
PS. If any Nokia people read this, thanks... and please don't adopt this in future either!

Can you please advice on RAW device file permission change to crw-r-----

Currently we have Oracle9i (9.2.0.5) RAC on IBM AIX 5.2 system with 2 node cluster.
We use the RAW device for Oracle data file default file permission for raw device is crw-rw---- can we change to crw-r-----(remove write permission from group).
Can you please advice.
Thanks & Regards

Hi,
What is your intention behind this? If protection, there can ve other ways possible.
Regards,
Ankit

Tablespace creation problem in RAC using raw device.

Hi ,
If i created a tablespace with the .dbf extension i am able to create the tablespace. on other hand if i use .dbf extension it;s failing..
can anyone help me out why it is happening??
here is the code...
Oracle9i Enterprise Edition Release 9.2.0.8.0 - Production
With the Partitioning, Real Application Clusters, OLAP and Oracle Data Mining options
JServer Release 9.2.0.8.0 - Production
SYS at terac >
SYS at terac >CREATE TABLESPACE FLOW_1 DATAFILE
2 '\\.\FLOW_1' SIZE 10304 K REUSE AUTOEXTEND OFF
3 EXTENT MANAGEMENT LOCAL AUTOALLOCATE SEGMENT SPACE MANAGEMENT AUTO
4
SYS at terac >/
Tablespace created.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
CREATE TABLESPACE TEST2 DATAFILE '\\.\FLOW_1.DBF' SIZE 10304 K
REUSE AUTOEXTEND OFF EXTENT MANAGEMENT LOCAL AUTOALLOCATE SEGMENT
SPACE MANAGEMENT AUTO
SYS at TERAC >/
CREATE TABLESPACE TEST2 DATAFILE '\\.\FLOW_1.DBF' SIZE 10304 K
ERROR at line 1:
ORA-01119: error in creating database file '\\.\FLOW_1.DBF'
ORA-27040: skgfrcre: create error, unable to create file
OSD-04002: unable to open file
O/S-Error: (OS 2) The system cannot find the file specified.
thanks
james.

I'm by no means a windows expert so I might be totally wrong here, but here goes.
You have to refer to the raw device by its name in the datafile clause. If you use a different name then the o/s can't find the device. Simple as that.

Solaris 10g RAC Clusterware Raw device dd comand block size, sizing

Hi
We are in the process of installing clusterware for RAc on solaris 10. I have 4 luns 500mb each, we want to use them for the voting disks and ocr,
1. Which one do we use /dev/dsk or /dev/rdsk?
2) Also how do you erase the data to clear them out, I think you use the dd comand correct, but what is the full sytax to cllear the voiting disk and the ocr, do I need to specify a block size ??
3) Somebody was saying I should label the disks, just incase the controllers change the numbering if they ever get reconfigured, can this be done in SUN Solaris, if so can you give me an example, and do I specify the label as the ocr path??
Thanks
For all the info.

1. Which one do we use /dev/dsk or /dev/rdsk?
$$
Please use Character Device ,This is general practice. However from 10.2.0.1 Block Device can also work.
# chown oracle:dba /dev/rdsk/cxtydzs6
# chmod 660 /dev/rdsk/cxtydzs6
2) Also how do you erase the data to clear them out, I think you use the dd comand correct, but what is the full sytax to cllear the voiting disk and the ocr, do I need to specify a block size ??
$$
do
dd if=/dev/zero of=/dev/ocr_disk$i bs=8192 count=25000 &
done
3) Somebody was saying I should label the disks, just incase the controllers change the numbering if they ever get reconfigured, can this be done in SUN Solaris, if so can you give me an example, and do I specify the label as the ocr path??
$$
Oracle recommends that you use a file name similar to dbname_raw.conf for this file.
The following example shows a sample mapping file for a two-instance RAC cluster. Some of the partitions use alternative symbolic link names. Make sure that the partition device file name that you specify identifies the same partition on all nodes.
system=/dev/rdsk/c2t1d1s3
sysaux=/dev/rdsk/c2t1d1s4
example=/dev/rdsk/c2t1d1s5
users=/dev/rdsk/c2t1d1s6
temp=/dev/rdsk/c2t1d2s3
undotbs1=/dev/rdsk/c2t1d2s4
undotbs2=/dev/rdsk/c2t1d2s5
redo1_1=/dev/rdsk/c2t1d2s6
redo1_2=/dev/rdsk/c2t1d3s3
redo2_1=/dev/rdsk/c2t1d3s4
redo2_2=/dev/rdsk/c2t1d3s5
control1=/dev/rdsk/c2t1d4s3
control2=/dev/rdsk/c2t1d4s3
spfile=/dev/rdsk/dbname_spfile_raw_5m
pwdfile=/dev/rdsk/dbname_pwdfile_raw_5m
Hope this will help..
Bharat
NS

Raw devices files

Hi,
I already created raw devices on database 10g R2. How can make those raw devices available to oracle user?
Best Regards
Yassin
Message was edited by:
Yassin

Yes, the os is Sun Solaris and my question how can
create tablespaces on raw devices.create tablespace abcd datafile '/dev/rdsk/c0t0d0s4/abcd.dbf' size ...
This is an example only, adapt it to your needs! Never fill up the rawdevice completely, leave some space for emergency situations. A different question: why are you using raw devices? What's the output of uname -a? Beginning with Solaris 8 (don't know the maintenance level), you can mount filesystems with the direct io option which gives you almost raw device performance with all advantages of a file system.
Thank you.
Yassin
Message was edited by:
Yassin

Is this a bug in APEX 3.0.1 Installation RAC database with RAW device???

Hello,
I am getting the following error when I try to complete approval process.
ORA-20001: Unable to create tablespace. ORA-01
119: error in creating database file '//./FLOW_1.dbf' ORA-27040: skgfrcre: creat
e error, unable to create file OSD-04002: unable to open file O/S-Error: (OS 2)
The system cannot find the file specified.
I checked the alert log for error here is the error in the alert log file.
ORA-1119 signalled during: CREATE TABLESPACE FLOW_1 DATAFILE '//./FLOW_1.dbf'..
and i found the syntax for create tablespace was wrong.
To create a tablespace in RAC with RAW device environment under windows.
it should have forward black slash instead of backward slash "/" without .dbf extension.
so, I changed the syntax i was able to create tablespace through sqlplus.
When APEX tool try to create this tablespace it throws this error..
Old: CREATE TABLESPACE FLOW_1 DATAFILE
'\\.\FLOW_1.dbf' SIZE 10304 K REUSE AUTOEXTEND OFF
EXTENT MANAGEMENT LOCAL AUTOALLOCATE SEGMENT SPACE MANAGEMENT AUTO
error : ORA-01119: error in creating database file '\\.\FLOW_1.dbf'
New:
CREATE TABLESPACE FLOW_1 DATAFILE
'\\.\FLOW_1' SIZE 10304 K REUSE AUTOEXTEND OFF
EXTENT MANAGEMENT LOCAL AUTOALLOCATE SEGMENT SPACE MANAGEMENT AUTO
Tablespace created.
Is it a bug in APEX 3.0.1.??
Is anyone installed APEX in RAC ??
James

Hi,
One observation, Apex is switching the Session ID after one got killed ? I was working on Apex page with browse Item to test open cursor count,
after killing the SID (227) on which the open cursor count was getting increase, it APEX automatically switches to new SID(149) for that session.
Now the problem is even if I have two SID's and one hits the maximum open cursor count, It is not switching to other SID instead the whole application becomes unavailable.
STATNAME SID VALUE USER
opened cursors current 20 14 APEX_PUBLIC_USER
opened cursors current 149 74 APEX_PUBLIC_USER
opened cursors current 194 71 APEX_PUBLIC_USER
opened cursors current 211 5 APEX_PUBLIC_USER
opened cursors current 227 325 APEX_PUBLIC_USER Killed
opened cursors current 244 15 APEX_PUBLIC_USER
opened cursors current 20 14 APEX_PUBLIC_USER
opened cursors current 149 76 APEX_PUBLIC_USER
opened cursors current 194 71 APEX_PUBLIC_USER
opened cursors current 211 5 APEX_PUBLIC_USER
opened cursors current 244 15 APEX_PUBLIC_USER
Please kindly help in this.
Thanks in Advance
Thanks & Regards
Sanjay
Edited by: user11204334 on Dec 8, 2010 1:02 AM

RAC backup with RMAN...put backup on diff file system

hello all,
I have not work a lot on SAN storages. One of my client has implementated 9i RAC. Now he wants to Add two more disk in SAN storage (Implemented RAID). So sun engineers will do this but before this i have to take full database backup(80GB database) throught RMAN. The problem or confussion from my side is that the database is on Sun SAN storage and i have to put the Full database backup taken by RMAN on local hard disk on the node (node 1 of rac). Is the possible since the SAN storage is RAW file system (as i guess) and i am putting the backup on local system.
Please help me out ...i have do this in couple of days..
Please tell me prosedure too to how to change the path of backup in RMAN if above is possible..
Its urgent
Thanks and Regards!!
Pankaj Rawat

Two things:
1) You will not have any problems taking RMAN backup for RAC and raw devices. None of them make your backups any different.
2) Based on your post, you are not very confident in your RMAN skills and this is your real problem. What is a must for you - take the backup, copy it on another machine and try to restore from it. Note, that you should NOT look at your original database during restore or take any files from there (even init.ora or spfile). If you don't have this done and don't have exact procedure - consider your backup as useless. This is a conservative approach but believe me - it's wort it when you SAN engineers screw up your storage. And they warned you. ;-)

Raw devices for 11g rac

hi,
whether 11gr2 rac supports raw devices in ibm aix 6.1 with power ha?
regards
sridhar

Hi;
Pelase see:
Raw Devices and Cluster Filesystems With Real Application Clusters (RAC) [ID 183408.1]
For your future RAC related issue I suggest close your issue here as answered than move your issue Forum Home » High Availability » RAC, ASM & Clusterware Installation which is RAC dedicated forum site.
Regard
Helios

Raw device support in 11gr2

We are currently running oracle 10.2.0.4 RAC using raw devices on the hpux itanium 64bit platform. We want to upgrade to 11.2.0.3 and I am wanting to know what parts of of database environment will still be able to utilize raw devices?
Can my datafiles remain on raw devices or will we need to somehow convert over to using ASM? What about the ocr and voting device? Can they remain on raw devices or do we need to configure ASM for the ocr and voting disk?
Thanks.

923395 wrote:
We are currently running oracle 10.2.0.4 RAC using raw devices on the hpux itanium 64bit platform. We want to upgrade to 11.2.0.3 and I am wanting to know what parts of of database environment will still be able to utilize raw devices?
Can my datafiles remain on raw devices or will we need to somehow convert over to using ASM? What about the ocr and voting device? Can they remain on raw devices or do we need to configure ASM for the ocr and voting disk?
Thanks.when all else fails, Read The Fine Manual
http://docs.oracle.com/cd/E11882_01/install.112/e24169/storage.htm#sthref558

Globaldevices file system

Hi
Solaris Cluster 3.3 u2 on Solaris 10 x86.
I have dedicated slice for global device file system. During the cluster creation, i have not chosen the default method which is lofi method.
For node "node1",
    Is it okay to use this default (yes/no) [yes]? no
    Testing for "/globaldevices" on "node1" ... done
For node "node2",
    Is it okay to use this default (yes/no) [yes]? no
    Testing for "/globaldevices" on "node2" ... done
After cluster creation, i am seeing lofi device for /global, i am not seeing the slice which i used for /globaldevices is used.
# mount | grep lofi
/global/.devices/node@1 on /dev/lofi/127 read/write/setuid/devices/rstchown/intr/largefiles/logging/noquota/global/xattr/nodfratime/onerror=panic/dev=240007f on Wed Oct 9 07:30:23 2013
/global/.devices/node@2 on /dev/lofi/126 read/write/setuid/devices/rstchown/intr/largefiles/logging/noquota/global/xattr/nodfratime/onerror=panic/dev=240007e on Wed Oct 9 07:30:23 2013
# mount | grep globaldevices
/globaldevices on /dev/dsk/c0t0d0s3 read/write/setuid/devices/rstchown/intr/largefiles/logging/xattr/onerror=panic/dev=840083 on Wed Oct 9 07:30:07 2013
# lofiadm
Block Device             File
/dev/lofi/126            /.globaldevices
# du -h /.globaldevices
1.0G   /.globaldevices
Someone clarify me why the cluster used lofi? Since i have a dedicated slice for /globaldevice, why the cluster created lofi again ?
Thanks & Regards,
Shashi Kanth.

Hi.
According
Planning the Oracle Solaris OS - Oracle Solaris Cluster Software Installation Guide
In case you want have use /globaldevice as separated slice, you should:
1. Create UFS file systems on required device. You require at least 512 MB.
2. Add record for this device to /etc/vfstab
3.Mount /globaldevice.
In this case cluster installation script will not create /globaldevice on lofi.
Additional requirements:
Planning the Global Devices, Device Groups, and Cluster File Systems - Oracle Solaris Cluster Software Installation Guid…
You can use this procedure:
Administering Device Groups - Oracle Solaris Cluster System Administration Guide
for Migrate the Global-Devices Namespace From a lofi Device to a Dedicated Partition.
Regards.

Raw devices versus Cluster File Systems in RAC 10gR2

Hi,
Does anyone using cluster file systems in a RAC 10gR2 installation, specifically IBMs GPFS?
Ive visited a company that is running RAC 10gR2 in AIX over raw devices. Why someone would choose to use raw devices , with all the problems to administer , when all the modern file systems are so powerful? Is there any issues when using cluster file systems + RAC? Is there considerable performance benefits when using raw devices with RAC ?
I´ve always used Oracle stand alone instances over file systems (since version 7) , and performance was always very good. I´ve tested raw devices almost 10 years ago , and even in that time (the hardware today is much better - SAN , 15K rpm disks , huge caches - and the file systems software today is much better) the cost to administer it does not compensate the benefits (only 5% more faster than file systems in Oracle 7).
So , besides any limitations imposed by RAC , why use raw devices nowadays ?
Regards,
Antonio Belloni

Hi,
spontaneously, my question would be: How did you eliminate the influence of the Linux File System Cache on ext3? OCFS2 is accessed with the o_direct flag - there will be no caching. The same holds true for RAW devices. This could have an influence on your test and I did not see a configuration step to avoid it.
What I saw, though, is "counter test": "I have tried comparing simple file copies from an OS level and the speed differences are not apparent - so the issue only manifests itself when the data is read via an oracle db." and I have no good answer to that one.
Maybe this paper has: http://www.oracle.com/technology/tech/linux/pdf/Linux-FS-Performance-Comparison.pdf - it's a bit older, but explains some of the interdependencies.
Last question: While you spent a lot of effort on proving that this one query is slower on OCFS2 or RAW than on ext3 for the initial read (that's why you flushed the buffer cache before each run), how realistic is this scenario when this system goes into production? I mean, how many times will this query be read completely from disk as opposed to use some block from the buffer? If you consider that, what impact does the "IO read time from disk" have on the overall performance of the system? If you do not isolate the test to just a read, how do writes compare?
Just some questions. Thanks.

Raw Device Backup to file system(OPS 8i)

Hi
Our currently setup is
Oracle database 8.1.6 (Oracle Parallel Server) Two Node
Noarchive Mode
Solaris 2.6
all database file ,redo logfiles,controlfiles under raw device.
database size 16 G.B
oracle block size 8192
currently we are using only export backup of oracle.
But now i want to take cold backup of oracle database to disk.
cold backup Raw --> Disk
How we can take cold backup with dd command and skip parameter ?
Is anybody have practical idea of dd command with skip parameter.
Thanks and regards
Kuljeet pal singh

you can use ufsdump instead of dd

OPS/RAC Vs Raw Device/File System

Similar Messages

Maybe you are looking for