DB Shutdown While dropping/adding ASM disks
Hi all,
I am using oracle database 10gR2 (10.2.0.3) on AIX 5.3.
I have dropped two large disks from my diskgroup and at the same time i added some small disks to the same diskgroup. Now rebalancing and dropping disks process in in progress and EM is showing that it will take more than 10 hours to comple. Although I have set the rebalance power to 11.
MY question is if I shutdown my database at this point then will it effect my database or on next startup ASM will resume the process??.
For safe side I have executed following query:
ALTER DISKGROUP oradata UNDROP DISKS;
So will there be any impact on database if we reboot our database at this stage??
Any idea???
I have taken the risk and have restarted my database and nothing went wrong. My database is up and rebalancing process has restarted.
Similar Messages
-
Hi,
I am consistently getting the following error since yesterday while mounting the ASM disk.
1. I created the ASM disks successfully and created the appropriate ASM DiskGroups using asmca utility.
2. I was able to see the diskgroups as mounted from asmcmd tool.
3. All of a sudden, all the diskgroups were dismounted yesterday and I see the following error in the alert logs for all the
diskgroups.
4. This is using ASM11g
As I exhausted all my avenues to fix, I have turned to you for further assistance. Please do the needful.
SQL> /* ASMCMD */ALTER DISKGROUP DataVSPSI MOUNT
NOTE: cache registered group DATAVSPSI number=3 incarn=0x7934bf5d
NOTE: cache began mount (first) of group DATAVSPSI number=3 incarn=0x7934bf5d
NOTE: Assigning number (3,0) to disk (ORCL:D01_VSP_SI)
Thu Dec 27 17:38:29 2012
NOTE: start heartbeating (grp 3)
kfdp_query(DATAVSPSI): 11
kfdp_queryBg(): 11
NOTE: cache opening disk 0 of grp 3: D01_VSP_SI label:D01_VSP_SI
NOTE: F1X0 found on disk 0 au 2 fcn 0.0
NOTE: cache mounting (first) external redundancy group 3/0x7934BF5D (DATAVSPSI)
NOTE: starting recovery of thread=1 ckpt=2.1 group=3 (DATAVSPSI)
WARNING: IO Failed. group:3 disk(number.incarnation):0.0xeae44fae disk_path:ORCL:D01_VSP_SI
AU:4 disk_offset(bytes):4333568 io_size:122880 operation:Read type:asynchronous
result:I/O error process_id:32162
WARNING: IO Failed. group:3 disk(number.incarnation):0.0xeae44fae disk_path:ORCL:D01_VSP_SI
AU:4 disk_offset(bytes):4202496 io_size:131072 operation:Read type:asynchronous
result:I/O error process_id:32162
ORA-15080: synchronous I/O operation to a disk failed
ERROR: ASM recovery failed to read ACD
NOTE: cache initiating offline of disk 0 group DATAVSPSI
NOTE: process 32162 initiating offline of disk 0.3940831150 (D01_VSP_SI) with mask 0x7e in group 3
WARNING: Disk D01_VSP_SI in mode 0x7f is now being taken offline
NOTE: initiating PST update: grp = 3, dsk = 0/0xeae44fae, mode = 0x15
kfdp_updateDsk(): 12
kfdp_updateDskBg(): 12
ERROR: too many offline disks in PST (grp 3)
WARNING: Disk D01_VSP_SI in mode 0x7f offline aborted
Thu Dec 27 17:38:29 2012
NOTE: halting all I/Os to diskgroup DATAVSPSI
NOTE: crash recovery signalled OER-15130
ERROR: ORA-15130 signalled during mount of diskgroup DATAVSPSI
NOTE: cache dismounting (clean) group 3/0x7934BF5D (DATAVSPSI)
NOTE: lgwr not being msg'd to dismount
NOTE: cache dismounted group 3/0x7934BF5D (DATAVSPSI)
Also, I think the lower level disk is fine as I am able to write to the disk as follows :
[oracle@rmanqa01 trace]$ dd of=/dev/sdp1
Test Test
0+1 records in
0+1 records out
10 bytes (10 B) copied, 4.97435 seconds, 0.0 kB/s
[oracle@rmanqa01 trace]$ id
uid=500(oracle) gid=500(oinstall) groups=6(disk),500(oinstall),501(dba),502(oper),503(asmadmin),504(asmdba),505(asmoper),506(horcm)
[oracle@rmanqa01 trace]$ ls -l /dev/sdp1
brw-rw---- 1 root disk 8, 241 Dec 28 11:59 /dev/sdp1
[oracle@rmanqa01 trace]$
The oracleasm also lists the disk D01_VSP_SI as follows :
[root@rmanqa01 log]# /etc/init.d/oracleasm listdisks
A01_VSP_SI
ADSK01
ARCH_AMS_SI
D01_VSP_SI
DATA_AMS_SI
DDSK01
DEMO_ARCH
DEMO_DATA
L01_VSP_SI
RDSK01
REDO_AMS_SI
You have new mail in /var/spool/mail/root
[root@rmanqa01 log]# /etc/init.d/oracleasm querydisk D01_VSP_SI
Disk "D01_VSP_SI" is a valid ASM disk
[root@rmanqa01 log]#
kfed tool was giving proper data till yesterday.
Today, I am getting the following :
./kfed read /dev/oracleasm/disks/D01_VSP_SI
kfbh.endian: 84 ; 0x000: 0x54
kfbh.hard: 101 ; 0x001: 0x65
kfbh.type: 115 ; 0x002: *** Unknown Enum ***
kfbh.datfmt: 116 ; 0x003: 0x74
kfbh.block.blk: 1936020512 ; 0x004: T=0 NUMB=0x73655420
kfbh.block.obj: 2147486324 ; 0x008: TYPE=0x8 NUMB=0xa74
kfbh.check: 2886846267 ; 0x00c: 0xac11c73b
kfbh.fcn.base: 0 ; 0x010: 0x00000000
kfbh.fcn.wrap: 0 ; 0x014: 0x00000000
kfbh.spare1: 0 ; 0x018: 0x00000000
kfbh.spare2: 0 ; 0x01c: 0x00000000
ERROR!!!, failed to get the oracore error message
[oracle@rmanqa01 bin]$
Please help.
Thanks
V V
Edited by: user13479556 on Dec 28, 2012 12:00 PMThanks Berx for pointing that out.
I deleted the ASM disks and the diskgroups and re-created a fresh and was able to start the ASM instance and mount the DGs.
Now the mount is persistent, but the dbca fails with the following error in the alert log. Under what situations can this error be seen ?
NOTE: Loaded library: System
SUCCESS: diskgroup DATAAMSSI was mounted
SUCCESS: diskgroup ARCHAMSSI was mounted
ERROR: failed to establish dependency between database R3AMSSI and diskgroup resource ora.DATAAMSSI.dg*ERROR: failed to establish dependency between database R3AMSSI and diskgroup resource ora.ARCHAMSSI.dg*Mon Dec 31 20:25:25 2012
SUCCESS: diskgroup LOGAMSSI was mounted
Mon Dec 31 20:25:25 2012
ERROR: failed to establish dependency between database R3AMSSI and diskgroup resource ora.LOGAMSSI.dg
Mon Dec 31 20:25:25 2012
Successful mount of redo thread 1, with mount id 784925673
Completed: Create controlfile reuse set database "R3AMSSI"
MAXINSTANCES 8
MAXLOGHISTORY 1
MAXLOGFILES 16
MAXLOGMEMBERS 3
MAXDATAFILES 100
Datafile
'+DATAAMSSI/R3AMSSI/system01.dbf',
'+DATAAMSSI/R3AMSSI/sysaux01.dbf',
'+DATAAMSSI/R3AMSSI/undotbs01.dbf',
'+DATAAMSSI/R3AMSSI/users01.dbf'
LOGFILE GROUP 1 ('+LOGAMSSI/R3AMSSI/redo01.log') SIZE 51200K,
GROUP 2 ('+LOGAMSSI/R3AMSSI/redo02.log') SIZE 51200K,
GROUP 3 ('+LOGAMSSI/R3AMSSI/redo03.log') SIZE 51200K RESETLOGS
Stopping background process MMNL
Stopping background process MMON
Starting background process MMON
Starting background process MMNL
Mon Dec 31 20:25:28 2012
MMON started with pid=17, OS id=10452
ALTER SYSTEM enable restricted session;
Mon Dec 31 20:25:28 2012
MMNL started with pid=18, OS id=10454
alter database "R3AMSSI" open resetlogs
RESETLOGS after incomplete recovery UNTIL CHANGE 945183
Errors in file /u01/app/oracle/diag/rdbms/r3amssi/R3AMSSI/trace/R3AMSSI_ora_10434.trc:
ORA-00313: open failed for members of log group 1 of thread 1
ORA-00312: online log 1 thread 1: '+LOGAMSSI/r3amssi/redo01.log'
ORA-17503: ksfdopn:2 Failed to open file +LOGAMSSI/r3amssi/redo01.log
ORA-15173: entry 'redo01.log' does not exist in directory 'r3amssi'
Errors in file /u01/app/oracle/diag/rdbms/r3amssi/R3AMSSI/trace/R3AMSSI_ora_10434.trc:
ORA-00313: open failed for members of log group 1 of thread 1
ORA-00312: online log 1 thread 1: '+LOGAMSSI/r3amssi/redo01.log'
ORA-17503: ksfdopn:2 Failed to open file +LOGAMSSI/r3amssi/redo01.log
ORA-15173: entry 'redo01.log' does not exist in directory 'r3amssi'
Mon Dec 31 20:25:29 2012
Checker run found 5 new persistent data failures
Mon Dec 31 20:27:07 2012
I can verify from asmcmd that the redologs mentioned above as missing is very much present in '+LOGAMSSI/R3AMSSI'
Thanks
V V -
ASM disks are not found at node2 on vmware
Hi All,
I'm very new to Oracle RAC. I was tring to install 10g rac in my desktop using vmware workstation. Also I configured the shared storage. Here i'hv rac1 and rac2 two virtual machines. while I created asm disks (ASM1,ASM2,ASM3) on rac1. While listing the asm disks at rac1 display the correct result but rac2 not returning anything.
Can anybody suggest on this. B'coz I tried several times rac installation and but it failed bcoz of mismatch ocr id.
ThanksDoes rac2's operating system see the disks? Assuming Linux for the remainder of this post:
Run "fdisk -l | more" on each node and compare the output. If you know the names of the devices (maybe something like: /dev/sdb /dev/sdc /dev/sdd), you can do an "ls -la /dev/sdb* " on each node and compare the file permissions and ownership. What is the ASM disk discovery string? Chances are rac2 has not picked up the new partitions created on rac1, which could be corrected by running "/sbin/partprobe /dev/sdb". If the device ownership is wrong, simply changing it as the root user won't persist across reboots: you will need to setup udev rules to re-apply permissions and ownership for each boot.
Only when the operating system device files are present, and owned by the correct user ("grid" or "oracle", depending on your installation), then you will be able to scan for the devices in ASM. In the case of RAC, you'll also need to ensure the device files are named identically between nodes. ASMLIB can help, but something like device mapper (/etc/multipath.conf, can be configured even to manage a single path as is the case when using a vmware's vmdk virtual disk) is usually a more robust solution in my opinion.
Some folks incorrectly assume that ASM will eliminate the system administration tasks entirely. The only thing it eliminates is managing devices after they've been successfully configured for use by ASM. LUN creation, zoning, masking, OS provisioning to include HBA bindings et al, and lastly OS permissions are all still very much required.
If still having problems, paste the above mentioned utility outputs and we will try to help you spot the issue. -
Error while creating asm disk group
i am trying to convert my database SRAVAN as an ASM instance.
so do i need to set ORACLE_SID=+ASM???? r else it wil be SRAVAN??
I WAS ENDED UP WITH FOLLOWING ERRORS WHILE CREATING DISK GROUP.
guyz please do help me
SQL> CREATE DISKGROUP dgroup1
2 NORMAL REDUNDANCY
3 FAILGROUP ctlr1
4 DISK '/u04/app/oracle/product/asmdisks/disk1'
5 FAILGROUP ctlr2
6 DISK '/u04/app/oracle/product/asmdisks/disk2';
CREATE DISKGROUP dgroup1
ERROR at line 1:
ORA-15018: diskgroup cannot be created
ORA-15031: disk specification '/u04/app/oracle/product/asmdisks/disk2' matches
no disks
ORA-15025: could not open disk '/u04/app/oracle/product/asmdisks/disk2'
ORA-15059: invalid device type for ASM disk
Linux Error: 32768: Unknown system error
Additional information: 42
Additional information: -1073785968
ORA-15031: disk specification '/u04/app/oracle/product/asmdisks/disk1' matches
no disks
ORA-15025: could not open disk '/u04/app/oracle/product/asmdisks/disk1'
ORA-15059: invalid device type for ASM disk
Linux Error: 32768: Unknown system error
Additional information: 42
Additional information: -1073785968
[oracle@sierra200 dbs]$ cd /u04/app/oracle/product/asmdisks
[oracle@sierra200 asmdisks]$ ls -ltr
total 205008
-rwxrwxrwx 1 oracle oinstall 104857600 Jul 27 11:42 disk1
-rwxrwxrwx 1 oracle oinstall 104857600 Jul 27 11:47 disk2
Thanks & Regards
Sravan DalavaiLooks like you are asking ASM to use plain files. Have you used the Device Loopback (losetup) a d made the emulate raw devices?
http://www.idevelopment.info/data/Oracle/DBA_tips/Automatic_Storage_Management/ASM_20.shtml -
ASM disk added without scan on second node
Hi All ,
Oracle Version:11.2.0.3
I need one help for one issue with ASM disk addition.
It is a two node RAC and one disk group was filled.
One disk was available as UNUSED001 and so we renamed it ran scan disk and added the disk to the diskgroup on node1.
But , as we did not run the scan disk on second node , the name is still showing as UNUSED001 and assigned to diskgroup so showing as MEMBER.
Also, the renamed disk is also showing as MEMBER but not assigned with any diskgroup.
Usually, when this heppens we have to reboot the node to fix the issue , but would like to get idea if this can be fixed without bouncing nodes.Hi ,
+ Probably that disk addition failed with ORA-15075 as same named device is not visible after renaming of disk.
As this validation takes place after writing disk header ,it is showing as MEMBER.
+ Get downtime of cluster on 2nd node and run scandisks on 2nd node.
+ Now renamed disk should be showing up on node 2.
+ if showing up ,then validate .
-- All expected diskgroups were mounted on both nodes or not
sql> select inst_id,name,state from gv$asm_diskgroup;
-- If mounted validate that renamed disk group_number and mount_status
sql> col path for a30
sql> select inst_id,group_number,path,mount_status from gv$asm_disk;
+ If group_number is 0 and mount_status is CLOSED ,then it is not part of any mounted diskgroup.
Add that disk again with force option in same diskgroup.
sql> alter diskgroup <diskgroup_name> add disk 'ORCL:<LABEL_NAME>' force;
And allow rebalance to complete.
Regards,
Aritra -
ASM Disk added .. (showing closed)
I have created ASM instance its working fine.
GROUP_NUMBER FILE_NUMBER COMPOUND_INDEX INCARNATION BLOCK_SIZE BLOCKS BYTES SPACE TYPE
1 267 16777483 752603781 512 102401 52429312 58720256 ONLINELOG
1 268 16777484 752603789 8192 2561 20979712 22020096 TEMPFILE
1 269 16777485 752603809 512 5 2560 1048576 PARAMETERFILE
I have added another disk.
asmtool -create D:\asmdisks\disk4 512
and restarted the ASM Instance and database.
SQL> select * from V$ASM_DISK;
GROUP_NUMBER DISK_NUMBER COMPOUND_INDEX INCARNATION MOUNT_S HEADER_STATU MODE_ST STATE REDUNDA LIBRARY
0 3 3 0 CLOSED CANDIDATE ONLINE NORMAL UNKNOWN System
1 0 16777216 4035805671 CACHED MEMBER ONLINE NORMAL UNKNOWN System
1 1 16777217 4035805672 CACHED MEMBER ONLINE NORMAL UNKNOWN System
1 2 16777218 4035805673 CACHED MEMBER ONLINE NORMAL UNKNOWN System
Can anybody guide me about this..If my instance is down, and I deleted the disk physically then ? I am not asking about database level diskgroup deletion.
please can you tell me how can mirroring is specifed ?
First we dont have diskgroup concept at database level . If instance is down and you delete the disk then nothing will happen because rebalancing will happen internally in your ASM. Before oracle removing the disk it will rebalance the data on that disk to other existing disk before totally removing it.
Mirroring is specified at the time of diskgroup creation like "external redundency or normal redundency or high redundency"
CREATE DISKGROUP disk_group_1 NORMAL REDUNDANCY
FAILGROUP failure_group_1 DISK
'/devices/diska1' NAME diska1,
'/devices/diska2' NAME diska2
FAILGROUP failure_group_2 DISK
'/devices/diskb1' NAME diskb1,
'/devices/diskb2' NAME diskb2; -
Dropping a Disk from ASM disk group
Hi Team,
I was dropping 10 disks from ASM , due to storage migration.
9 Disks I was dropped successfully, but 1 disk (/dev/oracleasm/disks/ST_OCR_VOTE02 ) was not able to recognize , it is part of which group..!!
That is not part of any group but still showing as member..!!! Please help . Not sure this is part of which group.
SQL> select GROUP_NUMBER,DISK_NUMBER,NAME,path,header_status,FAILGROUP from v$asm_disk where path='/dev/oracleasm/disks/ST_OCR_VOTE02';
GROUP_NUMBER DISK_NUMBER NAME PATH HEADER_STATU FAILGROUP
0 0 /dev/oracleasm/disks/ST_OCR_VOTE02 MEMBER
SQL>
0 17 NORMAL FORMER /dev/oracleasm/disks/STASM3_02 0 0
0 9 NORMAL FORMER /dev/oracleasm/disks/STASM1_01 0 0
0 10 NORMAL FORMER /dev/oracleasm/disks/STASM5_01 0 0
0 11 NORMAL FORMER /dev/oracleasm/disks/STASM4_01 0 0
0 12 NORMAL FORMER /dev/oracleasm/disks/STASM3_01 0 0
0 13 NORMAL FORMER /dev/oracleasm/disks/STASM2_01 0 0
0 14 NORMAL FORMER /dev/oracleasm/disks/STASM1_02 0 0
0 15 NORMAL FORMER /dev/oracleasm/disks/STASM2_02 0 0
0 16 NORMAL FORMER /dev/oracleasm/disks/STASM4_02 0 0
0 0 NORMAL MEMBER /dev/oracleasm/disks/ST_OCR_VOTE02 0 0
Thanks ManoharHi ,
As group_number of that disk is 0 ,it is not part of any mounted diskgroup in your system .
So,you cannot drop that disk .
May be sometime back due to some issue that disk was evicted ,but was not dropped properly ,so disk header still holds old diskgroup information.
kfed read <device_name>|egrep "kfbh.type|kfdhdb.hdrsts|kfdhdb.grpname"
above command will tell you the cause of MEMBER
To reuse that disk,
-- Validate same disk having similar status on all nodes of this cluster
-- if So,then to overwrite that disk header ,use FORCE option to create/add it into any existing diskgroup.
Regards,
Aritra -
Why I can't drop ASM disks and stop ASMLib ?
Hi, I wanted to delete disks but this operation failed:
1.
/etc/init.d/oracleasm deletedisk DATA1
Removing ASM disk "DATA1" [FAILED]
I know there is solution to overwrite ASM header information on a disk
dd if=/dev/zero of=<your_raw_device> bs=1024 count=100
and then droping is successful
but I also saw resources that said it should be count=50
Could anyone confirm count=100 or 50 ?
2.
Besides of that I tried to stop ASMLib and it failed too:
/etc/init.d/oracleasm stop
Dropping Oracle ASMLib disks: [ OK ]
Shutting down the Oracle ASMLib driver: [FAILED]
Ok, so using dd will help me to get rid of disks, but I worry about ASMLib in general, why I can't stop it? All ASM instances were removed before. There is no ASM software installed, besides ASMLib.The story is like that:
During Grid Infrastructure 11.2 installation I wanted to create new disk group but all available disks ware marked as 'MEMBER', so that I couldn't use any of them. Then I investigated that there is no database using ASM, and no ASM software installed. But ASMLib lists some disks. So former diskgroups must have been dropped in a bad way. As disks are marked as 'MEMBER' I wanted to drop them and create new ones. But this operation failed.
Yes, I use root account
oracleasm exit failed too:
oracleasm exit
Unmounting ASMlib driver filesystem: failed
Unable to unmount ASMlib driver filesystem
/var/log/oracleasm :
Cleaning any stale ASM disks...
Unmounting ASMlib driver filesystem: umount: /dev/oracleasm: device is busy
umount: /dev/oracleasm: device is busy
failed
Unable to unmount ASMlib driver filesystem
How to deal with this problem? Please help.
Edited by: Przemek P on 2012-02-07 07:00 -
Hi I am using Oracle 10gR2 on Solaris 10. I have some 12 ASM disks in a Disk Group. I want to remove 3 ASM Disks. Can I do this operation online? Will my data be affected? How should I proceed about it? When I will remove the ASM Disks, can I safely detach those disks physically? please help.
regardsahsen.javaid wrote:
Hi I am using Oracle 10gR2 on Solaris 10. I have some 12 ASM disks in a Disk Group. I want to remove 3 ASM Disks. Can I do this operation online? As already commented, yes. I would like to give you a practical example of how utterly neat ASM is in this regard and why it is IMO a major mistake not to use ASM.
We had to swap storage systems. This means that we have to migrate an entire Oracle database (over 1TB in size) from one set of LUNs to a completely different set of LUNs (on a different storage system).
The new LUNs are added to the existing diskgroup. The old LUNs are dropped (the drop is of course not immediate). A rebalance is issued. The next morning the entire database has been moved and restriped on the new set of LUNs and the old LUNs can now be removed from the system. This while the database was up and running and normal processing continued uninterrupted (this is a 24x7 database).
Then I get comments from some customers (like from a major local financial institution recently) that "+No, we do not want to use ASM as we rather use Veritas+". People like that are missing the point of what ASM is. Totally. -
OUI will not detect ASM disks on EqualLogic PS6100
RHEL 5.5 w/ ASMLIB
Oracle 11g w/ Dell EqualLogic PS6100 iSCSI SAN
We've created the ASM disks and can see them before starting the OUI...
[root@dr-dbnode2 logs]# kpartx -l /dev/mapper/eql-8-cb2b76-0fde2e258-3ac70a955884f0dd-oracleasmdisk3
eql-8-cb2b76-0fde2e258-3ac70a955884f0dd-oracleasmdisk3p1 : 0 585954747 /dev/mapper/eql-8-cb2b76-0fde2e258-3ac70a955884f0dd-oracleasmdisk3 63
[root@dr-dbnode2 logs]# kpartx -l /dev/mapper/eql-8-cb2b76-0d8e2e258-55270a955854f0dd-oralceasmdisk2
eql-8-cb2b76-0d8e2e258-55270a955854f0dd-oralceasmdisk2p1 : 0 585954747 /dev/mapper/eql-8-cb2b76-0d8e2e258-55270a955854f0dd-oralceasmdisk2 63
[root@dr-dbnode2 logs]# kpartx -l /dev/mapper/eql-8-cb2b76-074e2e258-08670a9557f4f0dd-oracleasmdisk1
eql-8-cb2b76-074e2e258-08670a9557f4f0dd-oracleasmdisk1p1 : 0 585954747 /dev/mapper/eql-8-cb2b76-074e2e258-08670a9557f4f0dd-oracleasmdisk1 63
[root@dr-dbnode2 etc]# oracleasm listdisks
ASMDATA1
ASMDATA2
ASMDATA3
[root@dr-dbnode2 etc]# oracleasm querydisk -v -d -p ASMDATA1
Disk "ASMDATA1" is a valid ASM disk on device /dev/dm-2[253,2]
/dev/mapper/eql-8-cb2b76-074e2e258-08670a9557f4f0dd-oracleasmdisk1p1: LABEL="ASMDATA1" TYPE="oracleasm"
/dev/sdj1: LABEL="ASMDATA1" TYPE="oracleasm"
[root@dr-dbnode2 etc]# oracleasm querydisk -v -d -p ASMDATA2
Disk "ASMDATA2" is a valid ASM disk on device /dev/dm-8[253,8]
/dev/mapper/eql-8-cb2b76-0d8e2e258-55270a955854f0dd-oralceasmdisk2p1: LABEL="ASMDATA2" TYPE="oracleasm"
/dev/sdf1: LABEL="ASMDATA2" TYPE="oracleasm"
[root@dr-dbnode2 etc]# oracleasm querydisk -v -d -p ASMDATA3
Disk "ASMDATA3" is a valid ASM disk on device /dev/dm-4[253,4]
/dev/mapper/eql-8-cb2b76-0fde2e258-3ac70a955884f0dd-oracleasmdisk3p1: LABEL="ASMDATA3" TYPE="oracleasm"
/dev/sdg1: LABEL="ASMDATA3" TYPE="oracleasm"
[grid@dr-dbnode2 ext]$ kfod asm_diskstring='/dev/oracleasm/disks/*' nohdr=true verbose=true disks=all status=true op=disks
286110 CANDIDATE /dev/oracleasm/disks/ASMDATA1 grid asmadmin
286110 CANDIDATE /dev/oracleasm/disks/ASMDATA2 grid asmadmin
286110 CANDIDATE /dev/oracleasm/disks/ASMDATA3 grid asmadmin
When we run the OUI we get to the 'Add Disks' screen and there are no disks listed at all...
So we tried to run with a -silent install and -force with the disks in our rsp file and the action log shows the following...
[root@dr-dbnode2 logs]# cat installActions2012-01-12_05-06-51PM.log
INFO: Waiting for completion of background operations
INFO: Completed background operations
INFO: Executing action at state CreateASMDiskGroup
INFO: Completed executing action at state <CreateASMDiskGroup>
INFO: Waiting for completion of background operations
INFO: Completed background operations
INFO: Moved to state <CreateASMDiskGroup>
INFO: Waiting for completion of background operations
INFO: Completed background operations
INFO: Validating state <CreateASMDiskGroup>
INFO: inventory location is/opt/oracle/app/oraInventory
INFO: inventory location is/opt/oracle/app/oraInventory
INFO: inventory location is/opt/oracle/app/oraInventory
INFO: Is CRS Installed false
INFO: while getting CSS versionjava.lang.Exception: Unable to find the specified Oracle Home.
WARNING: KFOD not found at
INFO: inventory location is/opt/oracle/app/oraInventory
WARNING: KFOD not found at
FINE: oracle.install.commons.util.Resource:Can't find resource for bundle oracle.install.asm.resource.ASMErrorResID, key oracle.install.asm.bean.ASMInputValidationErrorCode.hint
SEVERE: [FATAL] [INS-30508] Invalid ASM disks.
CAUSE: The disks [dev/oracleasm/disks/ASMDATA1, /dev/oracleasm/disks/ASMDATA2, /dev/oracleasm/disks/ASMDATA3] were not valid.
ACTION: Please choose or enter valid ASM disks.
INFO: Advice is ABORT
INFO: Adding ExitStatus INVALID_USER_INPUT to the exit status set
INFO: Completed validating state <CreateASMDiskGroup>
INFO: Terminating all background operations
INFO: Terminated all background operations
FINE: oracle.install.commons.util.Resource:Can't find resource for bundle oracle.install.commons.util.resource.ErrorCodeResourceBundle, key oracle.install.commons.util.exception.CommonErrorCode.hint
INFO: Finding the most appropriate exit status for the current application
INFO: Exit Status is -2
INFO: Shutdown Oracle Grid Infrastructure
INFO: Unloading Setup Driver
Been stuck here for a couple days...
My current direction is looking at the "WARNING: KFOD not found at" message...
kfod does exist in the /tmp/OraInstall2012-01-12_04-40-31PM/ext/bin/kfod folder while the installation is running...so I'm not sure what to make of it.
Let me know if anyone can shed some light here...
Thanks in advance,
ColbyWe managed to figure it out...
It turns out that a previously failed -silent run of the OUI had left some Oracle files hanging about. The OUI not showing the ASM disks because they were associated to the failed installation. After removing 'ALL' the files from the failed installation and a reboot we were able to get through the full install successfully.
Make sure you fully clean up after a failed install...
cs -
Want to move datafiles, controlfiles, redolog on new ASM Disks (11gR2 RAC)
Hi Guys,
Setup: Two Node 11gR2 (11.2.0.1) RAC on RHEL 5.4
Existing disks are from Old SAN & New Disks are from New SAN.
Can I move all datafiles (+DATA), controlfiles (+CTRL), redolog (+REDO) on new ASM Disks by adding disks in is same Diskgroup & dropping older disks from existing Diskgroup taking advantage of ASM Re-balancing Feature.
1) add required disks in the DATA Diskgroups,
ALTER DISKGROUP DATA ADD DISK
'/dev/oracleasm/disks/NEWDATA3' NAME NEWDATA_0003,
'/dev/oracleasm/disks/NEWDATA4' NAME NEWDATA_0004,
'/dev/oracleasm/disks/NEWDATA5' NAME NEWDATA_0005
REBALANCE POWER 11;
Check rebalance status from v$ASM_OPERATION.
2) When rebalance completes, drop the old disks.
ALTER DISKGROUP DATA DROP DISK
NEWDATA_0000,
NEWDATA_0001
REBALANCE POWER 11;
Check rebalance status from v$ASM_OPERATION.
3) Do it same for Redo log groups & Controlfile Diskgroups.
I hope, I could do this Activity, even if database is Up. is there possibility of Database block Corruption ??? (or is it necessary to perform above steps when database is down)
Would be appreciated, your quick responses on the same.
It's an urgent requirement. Thanks.
Regards,
ManishManish Nashikkar wrote:
Hi Guys,
Setup: Two Node 11gR2 (11.2.0.1) RAC on RHEL 5.4
Existing disks are from Old SAN & New Disks are from New SAN.
Can I move all datafiles (+DATA), controlfiles (+CTRL), redolog (+REDO) on new ASM Disks by adding disks in is same Diskgroup & dropping older disks from existing Diskgroup taking advantage of ASM Re-balancing Feature.
1) add required disks in the DATA Diskgroups,
ALTER DISKGROUP DATA ADD DISK
'/dev/oracleasm/disks/NEWDATA3' NAME NEWDATA_0003,
'/dev/oracleasm/disks/NEWDATA4' NAME NEWDATA_0004,
'/dev/oracleasm/disks/NEWDATA5' NAME NEWDATA_0005
REBALANCE POWER 11;
Check rebalance status from v$ASM_OPERATION.
2) When rebalance completes, drop the old disks.
ALTER DISKGROUP DATA DROP DISK
NEWDATA_0000,
NEWDATA_0001
REBALANCE POWER 11;
Check rebalance status from v$ASM_OPERATION.
3) Do it same for Redo log groups & Controlfile Diskgroups.
I hope, I could do this Activity, even if database is Up. is there possibility of Database block Corruption ??? (or is it necessary to perform above steps when database is down)
Would be appreciated, your quick responses on the same.
It's an urgent requirement. Thanks.
Regards,
Manish
Hi Manish,
Yes you can do that by adding new disk to existing diskgroup and delete old diskgroup. The good thing is this can be done online however you need to make sure the rebalance power is meet your business time, higher rebalance power is faster to rebalance to complete however it also will consume more resources
Cheers -
Correct steps to reload ASM disks into a running ASM instance on Linux RHEL
Hi guys,
i hope this is the correct forum, awfully i did not found any document specifically talking abou this.
first of all let me explain quickly my goal: in my server (it contains the same data of 3 production RACs for reporting purposes) i would like to shutdown a single instance (i have 3) and realod his disks to refresh the data or because for some reason its mount into ASM is failed.
this is a mix of Oracle ASM and Linux stuff...
on this server there is a shell script that perform the following steps:
1) shutdown 3 instances of 11g (11.2.0.2.0 with patch for asm already installed for bug... dont remember)
2) shutdown ASM instance
3) stop ASMLIB
4) stop multipathd
5) remove all disks devices from linux
6) take a snapshot on the 3 production RACs
7) remap the disks to this server (SAN stuff)
8) rescan the disks
9) start ASMLIB
10) start ASM instance
11) mount the DiskGroups
12) start 3 oracle instances
this works fine... my problem is when i try to refresh the disks of just 1 of the 3 instances, i perform the following steps:
1) stop instance ABC
2) dismount ABC diskgroups related to ABC instance
3) wait 5 seconds (dont ask... if idid it without ASM still have the disks locked)
4) delete the ASMLIB disks (oracleasm deletedisk XYZ) relate to such diskgroups
5) remove mpaths related to the disks that need to be refreshed or correctly mounted oneagain
6) remove devices
7) unmap disks on SAN
8) take snapshot (clone disks)
9) map disks from SAN
10) delete bindings file
11) rescan devices
12) execute multipath command (to rebuild bindings file)
13) execute oracleasm scandisks
14) mount DiskGroups
15) start instance ABC
actually i dont understand why the OS (or specifically ASMLIB) do not recognize me anymore one of the ASMLIB disk
do you have any idea ? or im completly crazy doing that ?
best regards,
LucaHi,
well this step
4) delete the ASMLIB disks (oracleasm deletedisk XYZ) relate to such diskgroups
is not good. Because it will clear the ASM Header and as a result ASM will never again be able to detect the disks.
So it is clear that "scnadisks" later will not see the disks, because you wiped them.
Worse: You did the snapshot after deleting them... this gets tricky to get them "repaired" again.
However there is the possibility to relabel a dropped ASM disks. You need to open an SR for this, since the procedure to restore the lost ASM header is not that simple..
Regards
Sebastian -
Hello All,
I have a quick question -
We have an ASM Diskgroup that had 8 Disks in it. I removed 2 Diskss and the Diskgroup has been rebalanced. I have not done anything with the removed Disks.
They both appear in the ASM target as eligble to add back into the Diskgroup.
Do I need to re-format/partition/create disk on them before adding them back in? Would existing data on the Diskl cause an issue in the Diskgroup?
This is the first time I will be adding back Disks that have been previously in the Diskgroup.
Thanks in advance for the advice.
MicheleHi,
Disks eligible to be assigned to a diskgroup must have the status "CANDIDATE" or "FORMER" or "PROVISIONED".
· CANDIDATE - Disk is not part of a disk group and may be added to a disk group with the ALTER DISKGROUP statement
· PROVISIONED - Disk is not part of a disk group and may be added to a disk group with the ALTER DISKGROUP statement. The PROVISIONED header status is different from the
CANDIDATE header status in that PROVISIONED implies that an additional platform-specific action has been taken by an administrator to make the disk available for ASM.
· FORMER - Disk was once part of a disk group but has been dropped cleanly from the group. It may be added to a new disk group with the ALTER DISKGROUP statement.
The ALTER DISKGROUP...DROP DISK without WAIT option statement returns before the drop and rebalance operations are complete. Do not reuse, remove, or disconnect the dropped disk until the HEADER_STATUS column for this disk in the V$ASM_DISK view changes to FORMER.
You don't need change anything on OS Level. Oracle will reuse asmdisk dropped without needs perform any administrative task on OS Level.
Regards,
Levi Pereira -
Rebuild ASM Disk - Copying multiple datafiles from one disk to another
Hi,
I have an environment of four 11GR2 Oracle databases on a Linux server. Each database has its own ASM disk.
DB1 -> ASM_DISK1
DB2 -> ASM_DISK2
DB3 -> ASM_DISK3
DB4 -> ASM_DISK4
I need to rebuild one of the ASM disks (ASM_DISK1), but first I need to copy all of the datafiles to another disk (ASM_DISK2). I tried backing up the database using RMAN, but it was taking too long (nearly two days when I cancelled it). So now I am going to copy the files using ASMCMD CP command.
Basically my task is as follows:
1. Shutdown database.
2. Copy all data from ASM_DISK1 to ASM_DISK2.
3. Drop ASM_DISK1.
4. Re-create ASM_DISK1.
5. Copy all data back to ASM_DISK1.
6. Start database.
Database size is 700GB.
I am using the below script to copy the files.
Copy Script
================
asmcmd ls +ASM_DISK1/DB1/DATAFILE >> asm_list.txt
for FILENAME in `cat asm_list.txt`
do
asmcmd >> asm_LOG.log <<EOF
cp ASM_DISK1/DB1/DATAFILE/$FILENAME ASM_DISK2/DB1_BACKUP/DATAFILE/$FILENAME.dbf
EOF
done
================
I will then rename each file in the database like so:
alter database rename file '+ASM_DISK1/DB1/DATAFILE/filename' to '+ASM_DISK1/DB1/DATAFILE/filename.dbf'
My questions are as follows.
Is this approach a valid solution?
Will renaming the files during copy corrupt the files?
When I copy the files back to the original disk after rebuild, then rename them, will the database be able to start?
Rgs,
Robrgilligan_tnf wrote:
Hi,
I have an environment of four 11GR2 Oracle databases on a Linux server. Each database has its own ASM disk.
DB1 -> ASM_DISK1
DB2 -> ASM_DISK2
DB3 -> ASM_DISK3
DB4 -> ASM_DISK4
I need to rebuild one of the ASM disks (ASM_DISK1), but first I need to copy all of the datafiles to another disk (ASM_DISK2). I tried backing up the database using RMAN, but it was taking too long (nearly two days when I cancelled it). So now I am going to copy the files using ASMCMD CP command.
And how do you propose to update the controlfile to point to the new location?
unless your datafiles are offline and/or the database is down, you will corrupt them and have an unusable database when you finish.
how were you doing this with RMAN? Depending on the size of your database(700G), it very well could take some time. I have restored databases at a rate of >300G/hr from scratch. You will need to shutdown at some point to relocate the controlfiles and system and redo logfiles.
Just curious, what is the problem with diskgroup ASM_DISK1 that you want to rebuild it?
Basically my task is as follows:
1. Shutdown database.
2. Copy all data from ASM_DISK1 to ASM_DISK2.
3. Drop ASM_DISK1.
4. Re-create ASM_DISK1.
5. Copy all data back to ASM_DISK1.
6. Start database.
Database size is 700GB.
I am using the below script to copy the files.
Copy Script
================
asmcmd ls +ASM_DISK1/DB1/DATAFILE >> asm_list.txt
for FILENAME in `cat asm_list.txt`
do
asmcmd >> asm_LOG.log <<EOF
cp ASM_DISK1/DB1/DATAFILE/$FILENAME ASM_DISK2/DB1_BACKUP/DATAFILE/$FILENAME.dbf
EOF
done
================
I will then rename each file in the database like so:
alter database rename file '+ASM_DISK1/DB1/DATAFILE/filename' to '+ASM_DISK1/DB1/DATAFILE/filename.dbf'
My questions are as follows.
Is this approach a valid solution?
Will renaming the files during copy corrupt the files?
When I copy the files back to the original disk after rebuild, then rename them, will the database be able to start?
Rgs,
Rob -
Oracle 12c ASM Disk Problem on AIX
Hi All,
We need to install 12.1.0.2 DB with ASM option on AIX 7.1. When we start to install 12.1.0.2 grid installation, in the 3.step we could not show adding Candidate disks to ASM. Disks right are 660 and oracle:oinstall ownership.
When we start to install 11gR2 installation on same platform we can see the Candidate Disk, and We can install 11gR2 ASM+DB on same platform. But for 12.1.0.2 is not.
How can we add Candidate Disks to 12c ASM İnstallation for below step ? The disks which is not shown for this step are like below. After first error I have changed disks' ownership like oracle:asmadmin, it isn't solved. I need urgent help.Hi Sve,
Thanks for replying me. What you have written are tried but not works. I have solved problem.
The problem is related AIX platform. I have Solved problem with below steps.
1- I have installed grid infrastructure with Software Only option.
2- After install Software Only Standalone Grid, I have tried to install manually ASM instance. But I have taken this error. rtld: 0712-001 Symbol CreateIoCompletionPort was referenced and it is solved with this MOSC ID: Doc ID1949184.1)
3- After Successfull 2.step you can now install ASM manually from asmca.
Maybe you are looking for
-
How can I hide the java script status bar?
I'm trying to hide the status bar using the feature in the advanced section of enable java script. I did a search in the help window of Firefox 5.0 and it shows the area in options to do this. I followed the instructions but the feature in my advance
-
ITunes won't open on updated IOS 7.0.2
Since updated to IOS 7.2 my iphone will not open itunes. It provides no error message. The app simply will not open. Flashes for a second then returns to home screen. I have tried signing in and out of itunes, turning phone on and off, and nothing ha
-
HI all, We are on SRM 5.0 ECS.THe flow is : Ext PR-->created SC in SRM >PO>GR-->INV We are trying to close some PO's but get the foll error: Document item "PR #/Item #" has not yet been carried forward to the current year Posting in fis. yr 2011 is
-
.
-
Direct upload from Photos app to Facebook & iOS Photos Album
When I use the Photos app to upload photos to Facebook, it places all of the pictures in an album called iOSPhotos. There is no option to post them just to my normal Facebook Timeline. In other words, I'm forced to select an Album. This is a problem