ASM Instances
In Oracle Database 10g there are two types of instances: database and ASM instances. The ASM instance,
which is generally named +ASM1, is started with the INSTANCE_TYPE=ASM init.ora parameter. This
parameter, when set, signals the Oracle initialization routine to start an ASM instance and not a standard
database instance. Unlike the standard database instance, the ASM instance contains no physical files; such
as logfiles, controlfiles or datafiles, and only requires a few init.ora parameters for startup.
Upon startup, an ASM instance will spawn all the basic background processes, plus some new ones that are
specific to the operation of ASM. The STARTUP clauses for ASM instances are similar to those for
database instances. For example, RESTRICT prevents database instances from connecting to this ASM
instance. NOMOUNT starts up an ASM instance without mounting any disk group. MOUNT option
simply mounts all defined diskgroups2.
The illustration in Figure 1 shows the basic init.ora parameters required to start ASM. Observe that all
ASM processes begin with asm, as opposed to the database instance, whose processes begin with ora.
ASM is the volume manager for all databases that employ ASM on a given node. Therefore, only one
ASM instance is required per node regardless of the number of database instances on the node.
Additionally, ASM seamlessly works with the RAC architecture to support clustered storage environments.
In RAC environments, there will be one ASM instance per clustered node, and the ASM instances
communicate with each other on a peer-to-peer basis using the interconnect.
Figure 1.
Instance type
SQL> select instance_name from v$instance
INSTANCE_NAME
----------------
+ASM
ASM init.ora parameters
*.background_dump_dest='/opt/app/admin/+ASM/bdump'
*.core_dump_dest='/opt/app/admin/+ASM/cdump'
*.instance_type=asm
*.asm_diskgroups=+DATA
*.large_pool_size=12M
*.asm_diskstring='/dev/rdsk/c3t19d*s4'
*.remote_login_passwordfile='SHARED'
*.user_dump_dest='/opt/app/admin/+ASM/udump'
Note that most database parameters are not allowed (or applicable) to be set in ASM instance. If you set an
invalid parameter in the ASM instance, you will get an ORA-15021 error. For example, if you try to set the
control_files parameter for the ASM instance, you’ll receive the following error message:
ORA-15021: parameter "control_files" is not valid in asm instance
1 For RAC configurations, the ASM SID is +ASMx instance, where x represents the instance number.
2 OPEN startup option performs the same function as MOUNT option; i.e., mount all asm_diskgroups.
6
ASM background processes
oracle 2423 1 0 Apr30 ? 00:00:00 asm_pmon_+ASM1
oracle 2425 1 0 Apr30 ? 00:00:00 asm_diag_+ASM1
oracle 2434 1 0 Apr30 ? 00:00:00 asm_mman_+ASM1
oracle 2436 1 0 Apr30 ? 00:00:00 asm_dbw0_+ASM1
oracle 2438 1 0 Apr30 ? 00:00:00 asm_lgwr_+ASM1
oracle 2440 1 0 Apr30 ? 00:00:00 asm_ckpt_+ASM1
oracle 2442 1 0 Apr30 ? 00:00:00 asm_smon_+ASM1
oracle 2444 1 0 Apr30 ? 00:16:03 asm_rbal_+ASM1
ASM installation
In cases where a single ASM instance is managing only one database instance, it may be sufficient to
maintain a single ORACLE_HOME for ASM and the database.
However, for systems that have ASM instance managing the storage for several database instances and
require higher availability, it is recommended that the ASM instance be installed in a separate
ORACLE_HOME (ASM_HOME) than the database ORACLE_HOME3. If installing ASM in a separate
ORACLE_HOME, then the listener should be run from that ASM_HOME.
In Oracle Database 10g Release 2, Oracle Universal Installer (OUI) and Database Configuration Assistant
(DBCA) have been enhanced to allow the user to seamlessly create and install an ASM instance in a
separate ORACLE_HOME. OUI now has options to:
o Install and configure a database that uses ASM for storage management
o Install and configure an ASM instance, without creating a database.
o Install and configure ASM on a system that already has a running database, where subsequently,
the DBA can use the EM Migration Utility to migrate the database to ASM.
Additionally, in Oracle Database 10g Release 2, ASM transparently supports both older and newer
software versions of the database. See ASM – Database Multi-Version Support section for more details.
This new feature provides higher availability as well as the foundation for the Database Storage
Consolidation Feature. For more information on Database Storage Clustering see the following section:
Database consolidation and clustering
Upgrading the ASM instance
There are no special procedures for upgrading ASM instance from 10.1 to 10.2. From an ASM point of
view, upgrading is just a matter of upgrading the software; i.e.; there are no persistent changes to disk
groups as part of upgrade. Simply, install 10.2 as a separate ORACLE_HOME, then all that DBUA should
need to do is move the configuration files from the old ORACLE_HOME to the new ORACLE_HOME
and make any updates to those files that are ORACLE_HOME specific.
The upgrade of the ASM instance from 10.1 to 10.2 can be accomplished through DBUA. First all the
databases that use the ASM instance being upgraded must be shutdown.
1) Start DBUA (Database Upgrade Assistant ) from 10.2 ORACLE_HOME/bin
2) DBUA has two options
a) upgrade ASM instance
b) upgrade database
3) Select "upgrade ASM" to upgrade the ASM instance.
Oracle 10 DBA Interview questions and answers
1. Is the following SQL statement syntactically correct? If not, please rewrite it correctly.
SELECT col1 FROM tableA WHERE NOT IN (SELECT col1 FROM tableB);
Ans. SQL is incorrect.
Correct SQL : SELECT col1 FROM tableA WHERE col1 NOT IN (SELECT col1 FROM tableB);
2. What is a more efficient way to write this query, to archive the same set?
Ans: SELECT col1 from tableA minus SELECT col1 from tableB
3.How would you determine that the new query is more efficient than the original query?
Ans: Run explain plan on both query and see the result .
4.How can we find the location of the database trace files from within the data dictionary?
Ans: Generally trace file on the database server machine is located in one of two locations:
1. If you are using a dedicated server connection, the trace file will be generated in the directory specified by
the USER_DUMP_DEST parameter.
2.If you are using a shared server connection, the trace file will be generated in the directory specified by the
BACKGROUND_DUMP_DEST parameter.
you can run sqlplus>SHOW PARAMETER DUMP_DEST
or
select name, value
from v$parameter
where name like '%dump_dest%'
5. What is the correct syntax for a UNIX endless WHILE loop?
while :
do
commands
done
6. Write the SQL statement that will return the name and size of the largest datafile in the database.
SQL> select name,bytes from v$datafile where bytes=(select max(bytes) from v$datafile);
7. What are the proper steps to changing the Oracle database block size?
cold backup all data files and backup controlfile to trace, recreate your database
with the new block size using the backup control file, and restore. It may be easier
with rman. You can not change datbase block size on fly.
8. Using awk, write a script to print the 3rd field of every line.
Ans:
awk '{print }'
awk '{print $3}
awk '{print $3}
9.Under what conditions, is a nested loop better than a merge join?
Ans:
Optimizer uses nested loop when we are joining tables containing small number of rows with an efficient driving
condition.
It is important to have an index on column of inner join table as this table is probed every time for a new value
from outer table.
Optimizer may not use nested loop in case:
1. No of rows of both the table is quite high
2. Inner query always results in same set of records
3. The access path of inner table is independent of data coming from outer table.
merge join is used to join two independent data sources. They perform better than nested loop when the volume of
data is big in tables
but not as good as hash joins in general.
10.Which database views would you use to ascertain the number of commits a user's session has performed?
Joining V$SESSTAT ,V$STATNAME
select * from V$SESSTAT a ,V$STATNAME b where b.CLASS=a.STATISTIC# and b.NAME='user commits' and a.sid=
11.What does the #!bin/ksh at the beginning of a shell script do? Why should it be there?
Ans: On the first line of an interpreter script, the "#!", is the name of a program which should be used to
interpret the contents of the file.
For instance, if the first line contains "#! /bin/ksh", then the contents of the file are executed as a korn shell
script.
12.What command is used to find the status of Oracle 10g Clusterware (CRS) and the various components it manages
(ONS, VIP, listener, instances, etc.)?
Ans:
$ocrcheck
13.Describe a scenario in which a vendor clusterware is required, in addition to the Oracle 10g Clusterware.
If you choose external redundancy for the OCR and voting disk, then to enable redundancy, the disk subsystem must be configurable for RAID mirroring/vendor clusterware. Otherwise, your system may be vulnerable because the OCR and voting disk are single points of failure.
14.How would you find the interconnect IP address from any node within an Oracle 10g RAC configuration?
using oifcfg command.
se the oifcfg -help command to display online help for OIFCFG. The elements of OIFCFG commands, some of which are
optional depending on the command, are:
*nodename—Name of the Oracle Clusterware node as listed in the output from the olsnodes command
*if_name—Name by which the interface is configured in the system
*subnet—Subnet address of the interface
*if_type—Type of interface: public or cluster_interconnect
You can use OIFCFG to list the interface names and the subnets of all of the interfaces available on the local node
by executing the iflist keyword as shown in this example:
oifcfg iflist
hme0 139.185.141.0
qfe0 204.152.65.16
You can also retrieve specific OIFCFG information with a getif command using the following syntax:
oifcfg getif [ [-global | -node nodename] [-if if_name[/subnet]] [-type if_type] ]
To store a new interface use the setif keyword. For example, to store the interface hme0, with the subnet
139.185.141.0, as a global interface (to be used as an interconnect for all of the RAC instances in your cluster),
you would use the command:
oifcfg setif -global hme0/139.185.141.0:cluster_interconnect
For a cluster interconnect that exists between only two nodes, for example rac1 and rac2, you could create the cms0
interface with the following commands, assuming 139.185.142.0 is the subnet addresses for the interconnect on rac1
and rac2 respectively:
oifcfg setif -global cms0/139.185.142.0:cluster_interconnect
Use the OIFCFG delif command to delete the stored configuration for global or node-specific interfaces. A specific
node-specific or global interface can be deleted by supplying the interface name, with an optional subnet, on the
command line. Without the -node or -global options, the delif keyword deletes either the given interface or all of
the global and node-specific interfaces on all of the nodes in the cluster. For example, the following command
deletes the global interface named qfe0 for the subnet 204.152.65.0:
oifcfg delif -global qfe0/204.152.65.0
On the other hand, the next command deletes all of the global interfaces stored with OIFCFG:
oifcfg delif -global
15.What is the Purpose of the voting disk in Oracle 10g Clusterware?
Voting disk record node membership information. Oracle Clusterware uses the voting disk to determine which instances are members of a cluster. The voting disk must reside on a shared disk. For high availability, Oracle recommends that you have a minimum of three voting disks. If you configure a single voting disk, then you should use external mirroring to provide redundancy. You can have up to 32 voting disks in your cluster.
16.What is the purpose of the OCR in Oracle 10g Clusterware?
Ans: Oracle Cluster Registry (OCR) is a component in 10g RAC used to store the cluster configuration information. It is a shared disk component, typically located in a shared raw volume that must be accessible to all nodes in the cluster.
The daemon OCSSd manages the configuration info in OCR and maintains the changes to cluster in the registry.
17. In Oracle Streams archived log downstream capture, which database view can be used to determine which archived
logs are no longer needed by the capture process?
Ans: V$ARCHIVE_DEST_STATUS
SELECT col1 FROM tableA WHERE NOT IN (SELECT col1 FROM tableB);
Ans. SQL is incorrect.
Correct SQL : SELECT col1 FROM tableA WHERE col1 NOT IN (SELECT col1 FROM tableB);
2. What is a more efficient way to write this query, to archive the same set?
Ans: SELECT col1 from tableA minus SELECT col1 from tableB
3.How would you determine that the new query is more efficient than the original query?
Ans: Run explain plan on both query and see the result .
4.How can we find the location of the database trace files from within the data dictionary?
Ans: Generally trace file on the database server machine is located in one of two locations:
1. If you are using a dedicated server connection, the trace file will be generated in the directory specified by
the USER_DUMP_DEST parameter.
2.If you are using a shared server connection, the trace file will be generated in the directory specified by the
BACKGROUND_DUMP_DEST parameter.
you can run sqlplus>SHOW PARAMETER DUMP_DEST
or
select name, value
from v$parameter
where name like '%dump_dest%'
5. What is the correct syntax for a UNIX endless WHILE loop?
while :
do
commands
done
6. Write the SQL statement that will return the name and size of the largest datafile in the database.
SQL> select name,bytes from v$datafile where bytes=(select max(bytes) from v$datafile);
7. What are the proper steps to changing the Oracle database block size?
cold backup all data files and backup controlfile to trace, recreate your database
with the new block size using the backup control file, and restore. It may be easier
with rman. You can not change datbase block size on fly.
8. Using awk, write a script to print the 3rd field of every line.
Ans:
awk
awk '{print $3}
awk '{print $3}
9.Under what conditions, is a nested loop better than a merge join?
Ans:
Optimizer uses nested loop when we are joining tables containing small number of rows with an efficient driving
condition.
It is important to have an index on column of inner join table as this table is probed every time for a new value
from outer table.
Optimizer may not use nested loop in case:
1. No of rows of both the table is quite high
2. Inner query always results in same set of records
3. The access path of inner table is independent of data coming from outer table.
merge join is used to join two independent data sources. They perform better than nested loop when the volume of
data is big in tables
but not as good as hash joins in general.
10.Which database views would you use to ascertain the number of commits a user's session has performed?
Joining V$SESSTAT ,V$STATNAME
select * from V$SESSTAT a ,V$STATNAME b where b.CLASS=a.STATISTIC# and b.NAME='user commits' and a.sid=
11.What does the #!bin/ksh at the beginning of a shell script do? Why should it be there?
Ans: On the first line of an interpreter script, the "#!", is the name of a program which should be used to
interpret the contents of the file.
For instance, if the first line contains "#! /bin/ksh", then the contents of the file are executed as a korn shell
script.
12.What command is used to find the status of Oracle 10g Clusterware (CRS) and the various components it manages
(ONS, VIP, listener, instances, etc.)?
Ans:
$ocrcheck
13.Describe a scenario in which a vendor clusterware is required, in addition to the Oracle 10g Clusterware.
If you choose external redundancy for the OCR and voting disk, then to enable redundancy, the disk subsystem must be configurable for RAID mirroring/vendor clusterware. Otherwise, your system may be vulnerable because the OCR and voting disk are single points of failure.
14.How would you find the interconnect IP address from any node within an Oracle 10g RAC configuration?
using oifcfg command.
se the oifcfg -help command to display online help for OIFCFG. The elements of OIFCFG commands, some of which are
optional depending on the command, are:
*nodename—Name of the Oracle Clusterware node as listed in the output from the olsnodes command
*if_name—Name by which the interface is configured in the system
*subnet—Subnet address of the interface
*if_type—Type of interface: public or cluster_interconnect
You can use OIFCFG to list the interface names and the subnets of all of the interfaces available on the local node
by executing the iflist keyword as shown in this example:
oifcfg iflist
hme0 139.185.141.0
qfe0 204.152.65.16
You can also retrieve specific OIFCFG information with a getif command using the following syntax:
oifcfg getif [ [-global | -node nodename] [-if if_name[/subnet]] [-type if_type] ]
To store a new interface use the setif keyword. For example, to store the interface hme0, with the subnet
139.185.141.0, as a global interface (to be used as an interconnect for all of the RAC instances in your cluster),
you would use the command:
oifcfg setif -global hme0/139.185.141.0:cluster_interconnect
For a cluster interconnect that exists between only two nodes, for example rac1 and rac2, you could create the cms0
interface with the following commands, assuming 139.185.142.0 is the subnet addresses for the interconnect on rac1
and rac2 respectively:
oifcfg setif -global cms0/139.185.142.0:cluster_interconnect
Use the OIFCFG delif command to delete the stored configuration for global or node-specific interfaces. A specific
node-specific or global interface can be deleted by supplying the interface name, with an optional subnet, on the
command line. Without the -node or -global options, the delif keyword deletes either the given interface or all of
the global and node-specific interfaces on all of the nodes in the cluster. For example, the following command
deletes the global interface named qfe0 for the subnet 204.152.65.0:
oifcfg delif -global qfe0/204.152.65.0
On the other hand, the next command deletes all of the global interfaces stored with OIFCFG:
oifcfg delif -global
15.What is the Purpose of the voting disk in Oracle 10g Clusterware?
Voting disk record node membership information. Oracle Clusterware uses the voting disk to determine which instances are members of a cluster. The voting disk must reside on a shared disk. For high availability, Oracle recommends that you have a minimum of three voting disks. If you configure a single voting disk, then you should use external mirroring to provide redundancy. You can have up to 32 voting disks in your cluster.
16.What is the purpose of the OCR in Oracle 10g Clusterware?
Ans: Oracle Cluster Registry (OCR) is a component in 10g RAC used to store the cluster configuration information. It is a shared disk component, typically located in a shared raw volume that must be accessible to all nodes in the cluster.
The daemon OCSSd manages the configuration info in OCR and maintains the changes to cluster in the registry.
17. In Oracle Streams archived log downstream capture, which database view can be used to determine which archived
logs are no longer needed by the capture process?
Ans: V$ARCHIVE_DEST_STATUS
Subscribe to:
Post Comments (Atom)
Blog Archive
-
▼
2008
(277)
-
▼
September
(261)
- Setting oracle database in archive mode from non-a...
- Oracle Version Information - How to CheckYour Curr...
- Database and Instance Shutdown
- Database and Instance Shutdown
- What Happens When You Open a Database
- How to Identify Your Oracle Database Software Release
- What is block change tracking in RMAN Oracle 10g
- What is Incremental Merge in RMAN 10g?
- Moving from DBMS_JOB to DBMS_SCHEDULER
- Moving from DBMS_JOB to DBMS_SCHEDULER
- Moving from DBMS_JOB to DBMS_SCHEDULER
- Moving from DBMS_JOB to DBMS_SCHEDULER
- Oracle Database 10g Scheduler - Associating Jobs w...
- Oracle Database 10g Scheduler - Creating Jobs With...
- Oracle Database 10g Scheduler
- DBMS_JOB Package
- Data Guard Database Synchronization Options
- Data Guard Role Management
- Data Guard Broker
- Data Guard Protection Modes
- Redo Apply and SQL Apply
- Data Guard Benefits
- Creating a Nested Materialized View: Example
- Creating a Fast Refreshable Materialized View: Exa...
- Automatic Refresh Times for Materialized Views: Ex...
- Periodic Refresh of Materialized Views: Example
- Creating Rowid Materialized Views: Example
- Creating Primary Key Materialized Views: Example
- Creating Subquery Materialized Views: Example
- Creating Prebuilt Materialized Views: Example
- Creating Materialized Join Views: Example
- Creating Materialized Aggregate Views: Example
- Privilages to create a refresh-on-commit materiali...
- Privilages to create a materialized view in anothe...
- What Privilages required to create a materialized ...
- What is materialized view?
- What is Optimizer Plan Stability?
- DBMS_ADVISOR Package
- What is Bigfile tablespace? why do we create it? w...
- Reverting a Table to its Previous Statistics
- Manipulating Statistics Using DBMS_STATS
- Copying Statistics Using DBMS_STATS
- Gathering statistics with DBMS_STAT Package
- Handling Errors During Backups: Example
- Backup Validation with RMAN
- Detection of Logical Block Corruption
- Detecting Physical and Logical Block Corruption
- Tests and Integrity Checks for Backups?
- When RMAN Performs Control File Autobackups
- How RMAN Performs Control File Autobackups?
- what command do you execute to ensure that the con...
- Which background processes coordinates the rebalan...
- Which background processes coordinates the rebalan...
- What is oracle ASM in Oracle 10g? How do we create...
- Load Balancing In Oracle 10g Real Application Clus...
- Oracle 10gR2 load balancing - Server side load bal...
- Oracle 10gR2 load balancing- Client-Side Connectio...
- Oracle 10gR2 load balancing - Part2 - Client-Side ...
- Oracle 10gR2 load balancing - Part1
- Oracle database tuning without any tools - using p...
- Oracle database tuning without any tools - using p...
- Oracle database tuning without any tools - using p...
- Oracle database tuning without any tools - using p...
- Quick DBA Test - Oracle 10g
- what are the major new featurs in Oracle 10g data ...
- Flashback version query in oracle 10g
- Optimizing Performance of Direct Path Loads
- Data Conversion During Direct Path Loads
- Data Loading Methods - SQL Loader
- How Does Data Pump Access Data?
- Oracle data pump 10g impdp help
- Oracle data pump 10g expdp help
- Exporting and Importing External tables using Data...
- Oracle 10g Data Pump APIs
- Difference between old exp/imp and new data pump e...
- Oracle Data pump 10g getting started
- Oracle Data Pump in Oracle Database 10g
- What is FLASHBACK TABLE ? Explain with examples?
- What is Oracle Secure Backup (OSB)?
- What is auxiliary instance? when it is created?
- Automatic Instance Creation for RMAN TSPITR
- Oracle 10 RMAN Views
- How to setup RMAN in Oracle 9i
- What is block pinging and why is it so bad?
- Can I test if a database is running in RAC mode?
- How does one stop and start RAC instances?
- How does one convert a single instance database to...
- How many OCR and voting disks should one have?
- Do you need special hardware to run RAC?
- Can any application be deployed on RAC?
- Oracle Real Application Clusters (RAC) FAQ - previ...
- Backup Scripts When Few Data Blocks Change
- Improving Incremental Backup Performance: Change T...
- Incrementally Updated Backups: A One Week Example
- Basic Incremental Backup Strategy
- Cumulative Incremental Backups
- Differential Incremental Backups
- Level 0 and Level 1 Incremental Backups
- RMAN Incremental Backups
- Using Compressed Backupsets
-
▼
September
(261)
Posts
No comments:
Post a Comment