Why are we using UDEV if ASMLib is available for working with ASM disks? “For RHEL6, Oracle will only provide ASMLib software and updates when configured with a kernel distributed by Oracle. Oracle will not provide ASMLib packages for kernels distributed by Red Hat as part of RHEL6.

1. Setup 4650
2. Setup 1603

Board of Funeral Directors and Embalmers Rules. Be sure to select the correct paper size in your print setup. 0660-05: Funeral Directors and Embalmers: 0660-06. View and Download Honeywell VisionPRO 8000 installer setup and system test pocket manual online. VisionPRO 8000 Thermostat pdf manual download.

ASMLib updates will be delivered via Unbreakable Linux Network(ULN) which is available to customers with Oracle Linux support. ULN works with both Oracle Linux or Red Hat Linux installations, but ASMlib usage will require replacing any Red Hat kernel with a kernel provided by Oracle.” – UDEV rules files will be available in /etc/udev/rules.d directory UDEV Filename Rules:.

Filenames begin with two decimal digits, to sequence the rules into ascending numerical order. After a dash, filenames contain descriptive words. The filename must end with “.rules” to be recognized. Rule clauses using a double equal sign () are predicate matching patterns, to determine if the rest of the line should be processed. Rule clauses are separated by a comma. Rule clauses with only a single equal sign (=) are actions to be performed.

Content after a sharp sign / hash mark / octothorpe (#) is ignored as a comment. Blank lines are OK. Caution About Changing Files Owned By The Distribution Do not change the default 50-udev-default.rules file. This file will be over-written should the UDEV package be updated. The Setup: 1.

I have created a virtual storage of 10GB in virtual box and attached it as shareable device on both of my rac nodes (rac1 & rac2). OS is OEL 5.5 VBoxManage createhd –filename asm.vdi –size 10240 –format VDI –variant Fixed VBoxManage storageattach rac1 –storagectl “SATA” –port 1 –device 0 –type hdd –medium asm.vdi –mtype shareable VBoxManage storageattach rac2 –storagectl “SATA” –port 1 –device 0 –type hdd –medium asm.vdi –mtype shareable VBoxManage modifyhd asm.vdi –type shareable 2. Created 5 partitions of 1GB each on the attached storage using fdisk root@rac1 # fdisk -l Disk /dev/sdb: 10.7 GB, 0 bytes 255 heads, 63 sectors/track, 1305 cylinders Units = cylinders of 16065. 512 = 8225280 bytes Device Boot Start End Blocks Id System /dev/sdb1 1 131 5 Extended /dev/sdb5 1 1+ 83 Linux /dev/sdb6 126 2 83 Linux /dev/sdb7 251 3 83 Linux /dev/sdb8 376 5 83 Linux /dev/sdb9 501 6 83 Linux Setting up UDEV rules 1. Add the following to the “/etc/scsiid.config” file to configure SCSI devices as trusted. Create the file if it doesn’t already exist.

Options=-g 2. Identify the unique SCSI id from scsiid command If linux 5, root@rac1 install# /sbin/scsiid -g -u -s /block/sdb SATAVBOXHARDDISKVB52c7cf45-b84d83a4 If linux 6, root@rac1 install#/sbin/scsiid -g -u -d /dev/sdb SATAVBOXHARDDISKVB52c7cf45-b84d83a4 3. Now, setup rules file: vi /etc/udev/rules.d/99-oracle-asmdevices.rules Add a rule for the above device.

A rule is essentially an if-then-else logic. In the rule, we specify is satisfied for an event, actions will be taken to setup the device. If the event is a scsi device (KERNEL, BUS attributes), then call the /sbin/scsiid -g -u -s program, passing the block device as first argument (PROGRAM attribute and%p in the rule definition). If the RESULT of the program call matches with a value of 259abde4c0000 (RESULT attribute in the rule), then create a device entry as “asmcrs01″, with owner as grid, group owner as oinstall, and permissions as 0660.

Syntax: KERNEL”sd.”, BUS”scsi”, PROGRAM”/sbin/scsiid -g -u -s%p”, RESULT”″, NAME=”″,OWNER=”grid”, GROUP=”oinstall”, MODE=”0660″ 3.1. View Mounted Disks SQL SELECT groupnumber, disknumber, mountstatus, headerstatus, modestatus, state, redundancy, name, failgroup, path FROM v$asmdisk WHERE modestatus = 'ONLINE'; GROUPNUMBER DISKNUMBER MOUNTS HEADERSTATU MODEST STATE REDUNDA NAME FAILGROUP PATH - - - - - - - - - - 1 0 CACHED MEMBER ONLINE NORMAL UNKNOWN DATA1 DATA1 ORCL:DATA1 HEADERSTATUS column indicates the use of the disk and may contain one of the following values:.

MEMBER – This disk is assigned to a disk group. PROVISIONED – The disk has been stamped with ASM data thereby prepared for assigned to a disk group and use by ASM. At present though, the disk is not being used.

CANDIDATE – The disk has been discovered and is potentially usable by ASM, but otherwise no actions have been taken to utilize the device. Create Disk Groups SQL CREATE DISKGROUP sidDG1 NORMAL REDUNDANCY FAILGROUP sidDB1FGa DISK '/dev/data2', '/dev/disk3' FAILGROUP sidDB1FGb DISK '/dev/data2a', '/dev/data3a' Diskgroup created. Drop Disk Group SQL DROP DISKGROUP sidDG1 INCLUDING CONTENTS; Diskgroup dropped. Adding a disk SQL ALTER DISKGROUP sidDG1 ADD FAILGROUP sidDB1FGa DISK '/dev/data4' FAILGROUP sidDB1FGb DISK '/dev/data4a'; Diskgroup altered. Dropping a Disk SQL ALTER DISKGROUP sidDG1 DROP DISK '/dev/data4'; Diskgroup altered. Alter rebalance power Override the default rebalancing power limit defined in the ASMPOWERLIMIT attribute for any disk group SQL ALTER DISKGROUP sidDG1 REBALANCE POWER 5; Diskgroup altered.

Mount/Dismount Disk Groups SQL ALTER DISKGROUP sidDG1 DISMOUNT; Diskgroup altered. SQL ALTER DISKGROUP sidDG1 MOUNT; Diskgroup altered. View basic metadata SQL SELECT name, allocationunitsize, state, type, offlinedisks FROM v$asmdiskgroup; NAME ALLOCATIONUNITSIZE STATE TYPE OFFLINEDISKS - - - - - sidDG1 1048576 MOUNTED EXTERN 0 sidDG2 1048576 CONNECTED EXTERN 0. SQL SELECT groupnumber, disknumber, mountstatus, state, redundancy FROM v$asmdisk; GROUPNUMBER DISKNUMBER MOUNTS STATE REDUNDA - - - - - 1 0 CACHED NORMAL UNKNOWN 1 1 CACHED NORMAL UNKNOWN 2 0 CACHED NORMAL UNKNOWN 2 1 CACHED NORMAL UNKNOWN Internal Metadata Consistency Check There may be inconsistencies between the internal metadata maintained for disk group files and the actual storage of those files. This can be checked through SQL ALTER DISKGROUP data CHECK REPAIR; Diskgroup altered.

Resize Diskgroup The ALTER DISKGROUP RESIZE operation will allow one to modify the size of storage and to perform a rebalance operation as well. SQL ALTER DISKGROUP data RESIZE DISK data1 SIZE 300 G REBALANCE POWER 2; Diskgroup altered. If the new size is greater than disk capacity, the statement will fail. Disk Group Version Compatibility Compatible attribute settings at the disk group level can allow older relational databases to coexist with newer ones in the same ASM environment. This attribute also determines whether or not newer ASM features are enabled. Setting Description COMPATIBLE.ASM Determines the compatibility level of the ASM instance. COMPATIBLE.RDBMS Determines the compatibility level of all RDBMS instances using the ASM disk group.

Suppose that an Oracle 11g R2 ASM instance needed to support Oracle 10g R1 database instances, and that one wanted to ensure that no ASM features beyond Oracle 10g were used within the disk groups. One would set the compatibility attributes for both properties to 10.1.

This is the case in the disk group listed next within the V$ASMDISKGROUP view. SQL SELECT name, compatibility, databasecompatibility FROM v$asmdiskgroup; NAME COMPATIBIL DATABASEC - - - DATA 10.1.0.0.0 10.1.0.0.0. Modifying Disk Group Compatibility If we want to use this same ASM instance to take advantage of the new features within Oracle 11g create a different disk group named “sidDG1” using the same ASM instance.

Next, within this new disk group one might enable the new features by a higher setting for COMPATIBLE.ASM. Nuance pdf pro 8.1. SQL ALTER DISKGROUP sidDG1 SET ATTRIBUTE 'compatible.asm' = '11.2'; Diskgroup altered. If we want to restrict the disk group to be compatible to database instances SQL ALTER DISKGROUP sidDG1 SET ATTRIBUTE 'compatible.rdbms' = '11.2'; Diskgroup altered. Disk Group Access Control We can isolate access to ASM data for one instance from another database instance which may have distinct security authorizations. Otherwise, any user with SYSDBA rights would be able to access any ASM disk group or file for any instance using the ASM installation. With access control, a user with SYSDBA rights is selectively granted SYSASM rights on the disk group or file to which access control has been granted. Note: This feature is available starting with Oracle 11g R2.

Implementing Access Control In order to implement access control, each database installation must be owned by a distinct system account. Access control will be implemented as below. One connects to the database instance using SYSDBA rights. A new disk group is created for that database instance or an existing disk group is modified. Default access control privileges are assigned to the disk group at this same time and all subsequent files created within the disk group will inherit these privilege settings.

Individual files whose default access control privileges were inherited from the disk group may be overridden with file-specific privileges. Access Control Details The specific access control privileges for a disk group are assigned in this manner:.

One enables access control for the disk group. One specifies the rights for the owner, access control user group and other users in the form of a 3-digit user mask, with each digit indicating the rights for the respective user accounts.

One adds system accounts as authorized users for the disk group. One adds users from the list of authorized users to the list of user groups for the disk group. The 3-digit mask may be understood as follows, with each of the 3 positions in the mask indicating the privilege assigned to the owner, user group and other users respectively. Managing Disk Group Access Control Groups With the access control privileges in effect for a disk group, one can then define which system accounts belong to the user group(s) associated with the disk group. Firstwe need to define one or more system accounts as users for the disk group. The owner is always considered to have access to the disk group. SQL ALTER DISKGROUP data ADD USER 'oracle1', 'administrator'; Diskgroup altered.

Next, these users may be associated with an access control group by means of the ALTER DISK GROUP ADD USERGROUP statement. These users will now obtain the user group privileges assigned to the disk group earlier in this discussion. SQL ALTER DISKGROUP data ADD USERGROUP 'test4xstorageadmin' WITH MEMBER 'oracle1', 'administrator'; Diskgroup altered. Now list the system accounts associated with the disk group user groups by querying V$ASMUSER. SQL SELECT groupnumber, usernumber, osname FROM v$asmuser; GROUPNUMBER USERNUMBER OSNAME - - - 1 1 oracle 1 2 administrator 1 3 oracle1 A new member to the user group may be added. SQL ALTER DISKGROUP data MODIFY USERGROUP 'test4xstorageadmin' ADD MEMBER 'oracle2'; Diskgroup altered. Or a member may be dropped from the user group, thereafter obtaining its privileges from the other users position in the mask.

SQL ALTER DISKGROUP data MODIFY USERGROUP 'test4xstorageadmin' DROP MEMBER 'oracle2'; Diskgroup altered. Finally a user group may be dropped altogether. SQL ALTER DISKGROUP data DROP USERGROUP 'test4xstorageadmin'; Diskgroup altered. Additional Disk Group Attributes. DISKREPAIRTME Individual disks within a group for which redundancy is enabled can fail while the others remain operational.

Until such time as the failed device is repaired and brought back online, extents updated on the other online disks of the group become stale and must be eventually copied to the failed disk once it is back online. As long as the failed disk is part of the group and the stale extents are tracked, the disk group may be repaired once the device is online again. The disk group attribute DISKREPAIRTIME indicates the time period for which stale extent tracking should occur. If the offline disk is brought online during this period, then it will be resynchronized using fast mirror resync. If the offline disk is not brought online during this time period, then it will be dropped from the disk group permanently. At a later point in time a device can be added to the group and it may be either resynchronized or rebalanced with the other disk group members as needed.

The default value for DISKREPAIRTIME is 3.6 hours, but it may be modified using the following command. SQL ALTER DISKGROUP sidDG1 SET ATTRIBUTE 'diskrepairtime' = '4h'; Diskgroup altered. Fast Mirror Resync We can quickly bring a disk within a disk group back online after a crash by means of the fast mirror resync capability. Consider the following scenario: an individual disk within a disk group crashes and for a period of two hours is offline from its disk group. Each update to a different extent creates a stale extent, although these are relatively few since updates during this period are concentrated on a few specific extents. Once the disk comes online, re-synchronizing it with the other members of the disk group can be very time consuming. For this reason one can bring the disk online individually with the following command.

SQL ALTER DISKGROUP sidDG1 ONLINE DISK d10001; Diskgroup altered. This immediately allows new writes to be included on this disk, avoiding any further backlog of stale extents.

Then, as processing permits, extents which became stale during the disk outage are copied to the device. This enhancement provides a dramatic reduction in disk group failure repair time. AUSIZE By default an allocation unit is 1 MB in size. However, if one knows in advance that very large files will be used in a disk group, then one can modify the allocation unit size at the time the disk group is created using the AUSIZE attribute.

The larger sizes permitted are 2, 4, 8, 16, 32 and 64 MB. With variable size extents the same formula in obtaining allocation units is used, but the size of the extents is much larger based upon the larger size allocation unit.

In this example a 2 MB allocation unit is specified for the disk group. SQL CREATE DISKGROUP sidDG1 NORMAL REDUNDANCY FAILGROUP sidDB1FGa DISK '/dev/data2', '/dev/disk3' FAILGROUP sidDB1FGb DISK '/dev/data2a', '/dev/data3a' ATTRIBUTE 'ausize' = '2M'; Diskgroup created.

Using a Template By default if you want to place all data files within the hot region of the disk for a particular disk group modify the attributes of the system-supplied template named DATAFILE in this manner. SQL ALTER DISKGROUP data ALTER TEMPLATE 'DATAFILE' ATTRIBUTES (UNPROTECTED COARSE HOT); Diskgroup altered. Thereafter, if one wanted to create a new data file using this template reference both the disk group and the template using the syntax ‘+DiskGroup(TemplateName)’. SQL ALTER TABLESPACE SiderisTS2 ADD DATAFILE '+DATA(DATAFILE)' SIZE 100M; Tablespace altered.

ASM Disk Group Sector Size Disk groups include an attribute known as SECTORSIZE. A substantial degradation in performance will occur if the SECTORSIZE setting is 512 while 4K sector disk devices are used. So we can use this setting to ensure that the operation of the disk group is consistent with the disk devices contained within the group. SQL CREATE DISKGROUP sidDG1 NORMAL REDUNDANCY FAILGROUP sidDB1FGa DISK '/dev/data2', '/dev/disk3' FAILGROUP sidDB1FGb DISK '/dev/data2a', '/dev/data3a' ATTRIBUTE 'sectorsize' = '4096'; Diskgroup created. SQL SELECT groupnumber, sectorsize FROM v$asmdiskgroup; GROUPNUMBER SECTORSIZE - - 1 4096 All devices within a disk group must have the same SECTORSIZE setting. ASM Intelligent Data Placement One would implement appropriate data placement of the data files using a few basic steps. First, one must ensure that the compatible.rdbms disk group attribute is set to the appropriate release.

SQL ALTER DISKGROUP data SET ATTRIBUTE 'compatible.rdbms' = '11.2'; Diskgroup altered. Next, using the ALTER DISKGROUPMODIFY FILE command within the ASM instance, one can move the data files to the appropriate region of the disk. SQL ALTER DISKGROUP data MODIFY FILE '+DATA/test4g/datafile/siderists1.277' ATTRIBUTE (COLD); Diskgroup altered. SQL ALTER DISKGROUP data MODIFY FILE '+DATA/test4g/datafile/siderisidx1.265' ATTRIBUTE (HOT); Diskgroup altered. Note: Initially this operation will only affect the placement of new file extensions. When rebalance operation happens, then all file extents will be placed according to the instructions given here. In addition to the primary region of a file, one could also specify the placement of mirror regions using the MIRRORHOT and MIRRORCOLD attributes as well.

Setup 4650

Setup 1603

Of course, the disk group must support mirroring in order for this option to succeed. SQL ALTER DISKGROUP data MODIFY FILE '+DATA/test4g/datafile/siderisidx1.265' ATTRIBUTE (HOT MIRRORCOLD); Diskgroup altered.