Configuring Disk for ASM in a Standalone environment

Hi,
I am attempting to install ASM for the very first time ! I am installing ASM in a Standalone Linux 6 environment. I am following CH3 of the 11gR2 Database Installation Guide. I have installed the Oracle ASM packages and have configured ASM.
I am now trying to configure the disk to use the ASM Library Driver
In my case, my server has a single 250 Gb SCSI disk. I have used this for the install of Linux 6. However I have left some space unformatted which I had intended to use with ASM.
when I do fdisk -l I get
Disk /dev/sde: 250.1 GB, 250059350016 bytes
255 heads, 63 sectors/track, 30401 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x000b3ade
   Device Boot      Start         End      Blocks   Id System
/dev/sde1   *           1          64      512000   83 Linux
Partition 1 does not end on cylinder boundary.
/dev/sde2              64       30402   243685376   8e Linux LVM
Disk /dev/mapper/vg_lab3-lv_root: 10.5 GB, 10485760000 bytes
255 heads, 63 sectors/track, 1274 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000
Disk /dev/mapper/vg_lab3-lv_swap: 4227 MB, 4227858432 bytes
255 heads, 63 sectors/track, 514 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000
Disk /dev/mapper/vg_lab3-lv_home: 52.4 GB, 52428800000 bytes
255 heads, 63 sectors/track, 6374 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000
when I do df -k I get
Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/mapper/vg_lab3-lv_root
                      10079084   3435876   6131208 36% /
tmpfs                  1551236        76   1551160   1% /dev/shm
/dev/sde1               495844     52850    417394 12% /boot
/dev/mapper/vg_lab3-lv_home
                      50395844   2217864 45617980   5% /home
A couple of things I am not sure of -
Q1. Has the linux install in effect created some virtual disks i.e.
/dev/mapper/vg_lab3-lv_root, /dev/mapper/vg_lab3-lv_swap and /dev/mapper/vg_lab3-lv_home ?
Q2. Are /dev/sde1 and /dev/sde2 virtual disks in their own right or partitions of /dev/sde ?
The 11gR2 install manual states on P3-13 that I should use fdisk or parted to create a single whole-disk partition on the disk devices to use
I am not quite sure what I need to do -
Q3a. I assume I can use part of my physical disk ( i.e. /dev/sbe2 ) with ASM rather than it having to be a complete physical disk ?
Q3b. Do I simply mark /dev/sbe2 as an ASM disk using oracleasm createdisk DISK1 /dev/sbe2 ?
or do I somehow have to create a partition on /dev/sbe2 ? ( I guess this harks back to Q2. in that is sbe2 a disk or a partition ? )
any help greatly appreciated
Jim

Thanks Billy.
I have checked back to my Linux install note and indeed it does seem to create the logical volumes at that stage i.e. it creates Lv_root, Lv_home and Lv_swap. During the nistall I was offered to specify the type of installation. The following options were given -
- Use All Space
- Replace Existing Linux Systems
- Shrink Current System
- Use Free Space
- Create Custom Layout
Q1. I chose the Use All Space. However from what you are saying it looks as if I perhaps should have used the option 'Create Custom Layout'. I am guessing the use all space option is what created the Logical Volumes - is this correct and should I have used Create Custom Layout ?
I reduced lv_home from 183732 to 50000 Mb. This left 173 Gb free - I intended to use this free space with ASM
Q2. Can a single disk only have 4 primary partitions ?
I guess I am where I am now, so I am wondering how I proceed with the disk layout I have presently.
Can you clarify from my output of fdisk -l that was in my original post, what exactly I have. My understanding is as follows -
There are 3 logical volumes created : /dev/mapper/vg_lab3-lv_root, /dev/mapper/vg_lab3-lv_swap and /dev/mapper/vg_lab3-lv_home
There is then the physical volume /dev/sde. On this physical volume is /dev/sde1 and /dev/sde2
Q3. Are these partitions on that volume ?
Q4. /dev/sde1 appears to be a boot partition and /dev/sde2 appears to be the partition of 173 Gb that I did not allocate during the Linux install. Is this correct ?
Q5. Is it possible that I can now take /dev/sde2 and mark that as an ASM Disk ? I wasn't sure if creating an ASM disk is restricted to an entire physical disk or if you can use a partition ?
When I do the following as root user
/etc/init.d/oracleasm createdisk VOL1 /dev/sde2
Unable to open device "/dev/sde2" Device or resource busy
I cannot understand how this is the case - /dev/sde2 is not mounted
lsof /dev/sde2 does not show anything using /dev/sde2
/etc/init.d/oracleasm status
shows ASM is loaded and /dev/oracleasm is mounted
Any help you can offer would be greatly received.
Thanks,
Jim

Similar Messages

Configuring Filesystem Free Disks for ASM ? Can you use free space for ASM

Is there a way to dedicate the available disk space for ASM in order to create a database under ASM?
I was hoping that I would not have to reinstall the Linux OS Enterprise 5 (Red Hat). So I de-installed my 2-node RAC environment, installed the ASM required RPMs for Red Hat linux Enterperise 5 and then proceeded to reinstall 11g Clusterware and set "Configure Automatic Storage Management (ASM)".
Did I miss something? THanks

Thank you.
I have 3 VM-Oracle guests running Oracle 11g. When I've used NFS as share that worked fine, however I was not using ASM. I know wish to get smart about ASM implementation.
Server1(rac1) Linux E5 64bit
Server2(rac2) Linux E5 64bit
Server3(asm) Linux E5 64bit (I created four partitions
./oracleasm createdisk DATDISK /dev/xvda10 OK
./oracleasm createdisk VOTDISK /dev/xvda11 OK
./oracleasm createdisk OCRDISK /dev/xvda12 OK
./oracleasm createdisk OFRDISK /dev/xvda13 OK
On Node1 and Node 2
./oracleasm scandisk
Scanning the system for Oracle ASMLib disks: OK
After this should I or how do you make a partition shared? Could I used NFS? so that I can proceed with the RAC install and then create database .

2 TB Disk For Asm (Sugesstions Please)

Hello,
I have setup a rac install on 11g Aix 6.1 and i have a 2TB disk that will be used for ASM. When i reach the configuration page of ASM to setup disks i see that the raw drive which is supposed to show a 2TB raw drive is showing the size as 4095 TB which is a big mismatch.
We had cleared the headers of the disk with this command "dd if=/dev/zero of=/dev/rhdisk7 bs=1k count=256" and i'm not sure if this caused the problem.
Any suggestions here. I am on 11.1.0.6 or is this a bug.
I was thinking of breaking this disk into 500G each into 4 drives equalling to 2TB.
Can someone please help here.

SQL> create diskgroup 'DWHDATAVOL1' external redundancy disk '/dev/raw/raw3','/dev/raw/raw4';
create diskgroup 'DWHDATAVOL1' external redundancy disk
'/dev/raw/raw3','/dev/raw/raw4'
FEHLER in Zeile 1:
ORA-15018: Plattengruppe kann nicht erstellt werden
ORA-15099: disk '/dev/raw/raw4' is larger than maximum size of 2097152 MBs
ORA-15099: disk '/dev/raw/raw3' is larger than maximum size of 2097152 MBshttp://jonathanlewis.wordpress.com/2009/08/14/tera-bug/
metalink : bug 7243305, 6453944
fixed on 11.2
- use 500G into 4 drives to 2TB.
- use 11gR2

Start and Stop ASM in a Standalone Environment

Hi,
I have just installed 11gR2 Grid Infrastructure and have ASM up and running.
Q1. How do I shut ASM down cleanly ? So far I have
shutdown the +ASM instance
shutdown the listener
however I still appear to have a number of processes running
grid1     6184     1 0 10:53 ?        00:00:19 /home/app/oracle/product/11.2.0/grid/bin/ohasd.bin reboot
grid1     6338     1 0 10:54 ?        00:00:31 /home/app/oracle/product/11.2.0/grid/bin/oraagent.bin
grid1     6351     1 0 10:54 ?        00:00:00 /home/app/oracle/product/11.2.0/grid/bin/evmd.bin
grid1     6387 6351 0 10:54 ?        00:00:00 /home/app/oracle/product/11.2.0/grid/bin/evmlogger.bin -o /home/app/oracle/product/11.2.0/grid/evm/log/evmlogger.info -l /home/app/oracle/product/11.2.0/grid/evm/log/evmlogger.log
grid1     6883     1 0 10:57 ?        00:00:05 /home/app/oracle/product/11.2.0/grid/bin/cssdagent
grid1     6897     1 0 10:57 ?        00:00:01 /home/app/oracle/product/11.2.0/grid/bin/ocssd.bin
when I do =>
grid1[+ASM]>srvctl status ASM
ASM is not running.
Q2. what are the above processes and how do I get them to close cleanly ?
Q3. What is the startup procedure for ASM ?
any help appreciated,
Jim

Look up the command
crsctl stop has

Skip partitioning a disk for ASM use

I think according to documentation, if you use ASMLib, you have to partition the disk first (usually create one single primary partition). If I don't use ASMLib (but still use ASM), is there anything bad if I use the disk without creating a partition first?
Oracle and clusterware 11gR2, if it matters.

Thanks. Indeed Oracle says recommend, not require in 11gR2 documentation:
http://download.oracle.com/docs/cd/E11882_01/install.112/e16763/oraclerestart.htm#CHDIAECD
"Oracle recommends that you create a single whole-disk partition on each disk that you want to use."
It dropped the following sentence that is in 11gR1 documentation:
http://download.oracle.com/docs/cd/B28359_01/install.111/b28263/stor_rac.htm#BABJBJAA
"To include devices in a disk group, you can specify either whole-drive device names or partition device names."
The problem I had with a partition is that, sometimes, after a minor kernel upgrade, partitions are gone under /dev. It happened to my RHEL4 box. After the upgrade, we lost /dev/vpath[cdefg]1 (note the number 1; vpath is IBM's sdd devices), but /dev/vpath[cdefg] remained. Interestingly, fdisk still indicates the partitions exist. Just the device files for partitions in /dev are missing. partprobe, cfgvpath, and other commands made no difference. If I remember right, manually mknod to create the files worked.

Configure Apps for SharePoint 2013 in dev environment without DNS

I have a SP 2013 dev env
http://spitlab/ .I want to configure app store in this environment
will I be able to do it without access to a DNS server
I followed the below two articles
http://www.ashokraja.me/post/Develop-SharePoint-2013-Napa-App-In-Local-Dev-Environment-Configuring-On-Premises-without-DNS.aspx
I am able to install third party apps but when I click on it . it gets redirected to sfs.in
next i try this
http://sharepointconnoisseur.blogspot.com/2013/07/shortcut-to-prepare-sharepoint-2013-app.html
same thing I am able to install 3rd party apps but when i click on it .. it goes to intranet.com
so is it possible to install third party apps on a dev box without DNS and check out third party apps and if what steps am i missing ?

Hi,
If you click the 3rd party apps and it is redirected to sfs.in or intranet.com, this means you configured app domain correctly.
You can read the official document per the following first link to understand what app domain is (with DNS configured), app domain format is as bellow image (borrowed from this
article), and app domain is defined as you want(e.g. ContosoApps.com).
Without DNS, as your above two articles described, the app domain (e.g. apps.com, or apps.sfs.in) is written manually in hosts file directly, you can construct an app domain as your own, then after you install a custom developed app, it should be the following
app url format.
http://technet.microsoft.com/en-us/library/fp161236(v=office.15).aspx
http://www.ashokraja.me/post/Develop-SharePoint-2013-Napa-App-In-Local-Dev-Environment-Configuring-On-Premises-without-DNS.aspx
http://sharepointconnoisseur.blogspot.jp/2013/07/shortcut-to-prepare-sharepoint-2013-app.html
Thanks
Daniel Yang
TechNet Community Support

Configuring EM for ASM

I have installed Oracle 10g database and created default database orcl. Also I configured ASM. Now I want to manage asm through EM.

I found solution. If someone have the same problem, please follow the steps.
1. Add following bold lines to listener.ora (in windows, C:\oracle\product\10.2.0\db_1\NETWORK\ADMIN)
# listener.ora Network Configuration File: C:\oracle\product\10.2.0\db_1\network\admin\listener.ora
# Generated by Oracle configuration tools.
SID_LIST_LISTENER =
(SID_LIST =
(SID_DESC =
(SID_NAME = PLSExtProc)
(ORACLE_HOME = C:\oracle\product\10.2.0\db_1)
(PROGRAM = extproc)
*(SID_DESC =*
*(SID_NAME = +ASM)*
*(ORACLE_HOME = C:\oracle\product\10.2.0\db_1)*
LISTENER =
(DESCRIPTION_LIST =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = 150.2.202.103)(PORT = 1521))
2. Restart your listener (lsnrctl stop, lsnrctl start)
3. Go to Enterprise Manager then choose Migrate to ASM from Administration page. Fill the form and click continue. then click database link on the upper right corner. (It shows home page)
4. Then click the your ip or hostname in General part. Click Targets and then you will see something like '+ASM bla bla'. click on it.
that's all.
Edited by: Munkhbayar on Sep 8, 2009 2:16 AM

Shared disk (SAN EVA 4400) configuration for ASM on linux RHEL5

Hi all,
I want to install RAC database (oracle 10g on RHEL5). Now we are configurind shared storage for the both servers (nodes). In disk configuration (SAN on EVA 4400) we have created the volume group for the disk to use for ASM and present the disk group to the servers. Actually when we check the server we cannot see the disk (shared storage).
[root@cdr-analysis01 ~]# ls /dev/cciss/ -lR
/dev/cciss/:
total 0
brw-r----- 1 root disk 104, 0 Jul 1 2009 c0d0
brw-r----- 1 root disk 104, 1 Jul 1 14:29 c0d0p1
brw-r----- 1 root disk 104, 2 Jul 1 14:29 c0d0p2
brw-r----- 1 root disk 104, 3 Jul 1 14:29 c0d0p3
brw-r----- 1 root disk 104, 4 Jul 1 2009 c0d0p4
brw-r----- 1 root disk 104, 5 Jul 1 14:29 c0d0p5
brw-r----- 1 root disk 104, 6 Jul 1 14:29 c0d0p6
brw-r----- 1 root disk 104, 7 Jul 1 2009 c0d0p7
brw-r----- 1 root disk 104, 8 Jul 1 14:29 c0d0p8
(they are local disks not the shared disk)
My questions are:
- How to mount the disk to be seen in the server
- How to configure the disk for ASM is there any document for this. In my knowledge the disk to use for ASM doesn't need to be formated with a file system and shouldn't see when we run the df command.
Does anybody can help me as it's very urgent for us.
Thank you
Raitsarevo

Hi all,
NOw the disk is presented to server. but my actual problem is how to create shared partition of this disk for the servers.
Our actual status is we have created one volume group with 500 Gb in the sun and this volume group is called vg_oradata. we will use this storage for OCR, voting disk and the databse file (ASM). A si know from documentation i have to create one shared partition for OCR, one shared partition for Voting disk and another for database.
That i want to do is creating for example :
ocr_partition (25 Gb) : for OCR
vote_partition (25 Gb): for Voting disk
oradata_part1 (150 Gb): for database (ASM)
oradata_part2 (150 Gb): for database (ASM)
oradata_part3 (150 Gb): for database (ASM)
My problem is how to create this partition because when i'm trying to create it from Logical Volume Manager of Linux i cannot find them in to servers.
I mean i'm new in system administrator and our SAN vendors also doesn't know how to it in LInux system.
Could you help me please if possible.
My system is Linux Red Hat Enterprise 5 and i'm going to use Oracle 1Og RAC.
Thank you
Lucienot

Database is not created on ASM configured Disks

hi,
I have installed Oracle 11gR2 Grid Infrastructure and Automatic Storage Management for a Standalone Server through "Grid" user. After then I install Oracle Database 11gR1 Software through "Oracle" User. There is no problem occure during the installation of these software. The OS is OEL-5.4. AND
ASM configured Disks are "DISK1, DISK2 AND DISK3" and the Group Name is "+DATA".
The Problem is when i want to create Database and chose ASM configured disk for Database storage, at the database creation process on 27% an error occur that is "ORA-03114 Not connected to Oracle". Sir but when i chose File System for Database Storage, Database created successfully.
Sir i am new in Oracle Grid Infrastructure and ASM please help me.

Hi buddy,
Are there errors in the alert.log of RDBMS istance, and what about traces ? Have it been generated ?
Regards,
Cerreia

ORA-15063: ASM discovered an insufficient number of disks for diskgroup

Hello DBAs,
I have encountered this problem. I am using storage vendor snapshot capability. On my first node, I have 2 disks for ASM diskgroups. ORCL:DATA belongs to DATA diskgroup and ORCL:FLASH belongs to FLASH diskgroup. After making the snapshots of these two disks, I mapped them to my second host which already has ASM instance. I ran /etc/init.d/oracleasm scandisk and listdisks I see two disks. But when I start up ASM instance I get this error:
SQL> startup
ASM instance started
Total System Global Area 130023424 bytes
Fixed Size 2019032 bytes
Variable Size 102838568 bytes
ASM Cache 25165824 bytes
ORA-15032: not all alterations performed
ORA-15063: ASM discovered an insufficient number of disks for diskgroup "FLASH"
ORA-15063: ASM discovered an insufficient number of disks for diskgroup "DATA"
Of course, I have the asm_diskstring and asm_diskgroups parameter set in my init+ASM.ora file. The ASM version is identical across both hosts. I guess my question would be, is it possible to do this? Do I need to re-create the ASM disks using /etc/init.d/oracleasm? Any help is much appreciated.
Thanks,
TD

What is your configuration?
If you have AIX and EMC san, then refer to this metalink note: 467702.1
In my case, it was HP-UX and HP san EVA8200. But our problem was two different disks were presented to the two nodes, with same names by the unix admin/san admin guys.
Before they could figure out the problem, I had done a lot of research for few days to figure this out. Another problem I noticed in some cases was improper permissions. Either the db software owner didn't have permissions to the asm disks, or if the owner of both software was same, then the actual permissions/ownership of the shared raw devices was different on the two nodes.
Hope this helps. And I will appreciate once you fix the problem if you can also update this forum with your solution. That will help the community in future.
Thanks

Can't create disks for oracle ASM

Sorry for duplicating my post, I posted it in linux group, but don't know if I also can get help here.
Hi, all advanced,
I am installing oracle 10gR2 in Fedora core 4, and I prefer to use oracle Automatic Storage Management feature. Since oracle doesn't support FC4, I have to set up rRHEL4 kernel core to install ASM driver and configure it. The things went well until I want to create disks for ASM, for example:
# /etc/init.d/oracleasm createdisk DISK1 /dev/hda5
Marking disk "/dev/hda5" as an ASM disk: asmtool: Unable to clear device "/dev/hda5": Input/Output error
When I install the FC4, /dev/hda5 is only one of the partitions in the physical disk, not a full independent physical disk. Same to other disks that I want to put into ASM disks. I google the net but didn't find any similar problem happened to other players. So I need your help!
wand

Hi Rafael,
What version are you using? I have had similar problems in the past. Worst case I had to recreate the widget and copy the coding across.
Can you send me the project file, I will understand also if you do not want to, I will try it on my machine.
Regards
John

Configuring rman for dataguard environment.

Hi all,
Here we have a dataguard environment with db1 as primary and db2 as physical standby database. We configured dataguard in 10g environments and we are not using grid control. Can anyone provide the steps for configuring rman for incremental backups in this environment? We want to schedule the backups to be taken in standby db, (that is applicable in case of swithover too). In the case of swithover to db2 what all do we need to do in rman manually to resume backups in db1?
Thanks

Hello;
A Standby database is a copy of the Primary so if you try to register it RMAN will probably tell you its already registered. I would start with this :
Standby based backup
1. Enable auto backup of the control file and server parameter file. CONFIGURE CONTROLFILE AUTOBACKUP ON.
2. Skip backups on datafiles already backup. CONFIGURE BACKUP OPTIMIZATION.
3. Set the archive deletion policy. CONFIGURE DELETION POLICY TO NONE.
You have to decide on the exact RMAN script you want to use, but since the database is registered you can perform backup on either the Primary or the Standby as the need arises.
See "Using RMAN to Back Up and Restore Files" in "Data Guard Concepts and Administration 10g Release 2 (10.2) B14239-05"
http://docs.oracle.com/cd/B19306_01/server.102/b14239/rman.htm
Example in this post
How to manage primary and standby database backups using rman catalog
How to use standby backup to restore primary db
http://alexzeng.wordpress.com/2008/09/13/how-to-use-standby-backup-to-restore-primary-db/
Best Regards
mseberg

How Prepare Disks for RAC installation

How configure disks for RAC instalation for Windows Enterprise Server 2003

Best resource for RAC on Windows:
www.jobcestbon.com/oracle/RacOnWindows.pdf
HTH
Thanks
-Chandra PabbaI read it, but i unable to configure Shared Memory accourding to this link. Please guide me how i can configure shared Memory for RAC in Windows Server Enterprise 2003.
Thanks

Looking for disk I/O test for ASM?

We are running oracle 10.2.0.3 on RHEL4 itanium 64 bit and we are using ASM for out database storage. IN our production environemnt the disk subsystem is an EMC clarion array but in our test environment our disk subsystem is an HP MSA storage.
I am looking for a good way to test the disk I/O throughput between the 2 storage arrays when the disks are configured as oracle ASM disks?
Can ayone make any suggestions on this testing.
Thx.

Perhaps not understanding your question, but we run 10g2 on Linux RHEL with ASM on SAN and use iostat to monitor IO performance.
We implement multipathing and so run /sbin/multipath -l to discover the names of the devices and then run iostat like so:
iostat -x dm-0 dm-1 15
which will print stats every 15 seconds for the 2 named devices dm-0 and dm-1
HTH,
Peter

Creation of ASM disk for OUI

I need to install a Oracle Database in order to install Enterprise Manager Cloud Control 12c.
Need the database to use a ASM disk.
I used the following command to create the disk, per the Oracle Database Installation Guide.
#/usr/sbin/oracleasm createdisk DISK1 /dev/sdd1
#oracleasm listdisks
DISK1
However, when running the OUI for Oracle Database 12c (understand 11.2.0.3 is certified for Cloud Control), step 7
errors INS-30517 when attempting to select "Oracle Automatic Storage Management" for "Storage type".
Researched the error at this location but no cause or action was provided.
http://docs.oracle.com/cd/E16655_01/server.121/e26079/common_errormessages.htm
INS-30517: Automatic Storage Management software is not configured on this system.
The database install guide states that I need to ensure the "disk discovery string" is set the "ORCL:*" or is left empy ("") so the installer discovers these disks.
It doesn't show how to confirm or change the settings.
At this point I'm at a stopping point.

All ASMLib installations require the oracleasmlib and oracleasm-support packages. The oracleasm kernel driver is included in the Oracle UEK kernel. Perhaps you are missing the oracleasmlib package. You can download it from:
Oracle Linux: Oracle ASMLib | Oracle Technology Network
Oracleasmlib is not necessary for ASM to work, but it contains software necessary for Linux oracleasm, including the /usr/sbin/oracleasm-discover utility, which the Oracle installer used in the previous 11g version to detect available ASM volumes.

Configuring Disk for ASM in a Standalone environment

Similar Messages

Maybe you are looking for