Kill all the OS process under particular user

Below command will generate command with list of available process for user_name.

$ ps -aef  | grep -i user_name | awk '{print $2}' | sed 's/^/kill -9 /' > kill.txt

Now execute command to kill the process.


Exadata Documentation

Exadata and Storage cell documentation will be available by default on the below path on each storage cell server.

[root@cellserver1 doc]# pwd

[celladmin@cellserver doc]#ls -ltr
total 29
drwxr-xr-x 3 root root 4096 Jan 23  2014 Japanese
drwxr-xr-x 7 root root 4096 Jan 23  2014 doc
-rwxr-xr-x 1 root root 5419 Jan 23  2014 welcome.html

Exadata DMA Roles and Responsibility

Exadata administration can be done by single Exadata DMA (Database Machine Administration). There is no requirement of team for each domain.Only one specialist (DMA) is required to manage the Exadata.

Based on our experience we have given following list of tasks which need to be performed by DMA. We can add many tasks in this list but we have just covered most common tasks specially from Exadata perspective.

Roles and Responsibility of DMA

Hardware installation and maintenance: A database machine required periodic maintenance like firmware upgrade or replace any failure component of machine.

Network Administration: Exadata comes with CISCO ethernet switch for Management and ILOM network, infiniband swithces for cell and DB Node node connectivity so management of network is also includes in DMA's task. General knowledge of netwrok topology required to manage Infiniband switches.

Operating System Administration: Linux or Solaris OS administration. Also manage server availability during patching with high availability management. Setting up users,  applying patches, managing disk space, NFS mount, 

Exadata Storage Administration: Important component of Exadata. In depth knowledge is required to manage the cell servers as this is the main component of Exadata which also configured with Exadata Storage Software. It include monitoring of celldisk, griddisk, cell flash cache, flash disks, OS administration, infiniband connectivity  administration. Maintain storage availability during patching.

Grid Infrastructure Administration: Inlcudes clusterware, ASM disk group, mirrored cell group as Exadata managed mirroring on ASM level not cell level.

Database Administration: Includes single instance, single instance on ASM, one or more RAC node database. Exadata specific parameters.

Here DMA does not need to know full scale Linux , storage or network. Only few relevant concepts are required to manage the Exadata. 

Overall required skill for component wise in percentage as follows.

  • 60% Database Administration
  • 20% Cell Storage Administration
  • 15% Linux Administration 
  • 5% Miscellaneous (Infiniband, network, etc.)
You can follow our Exadata Certification Question Bank to achieve maximum success in your Exadata certification exam.

Exadata Storage Index Part-2 (Demo)

Continue from previous blog.....

In previous blog we have explained storage index and it's features. Now we will check real time performance benefits of storage index.

we have flush shared pool and buffer cache to determine exact value and benefits given by Storage Index.

Below query will show that how much physical IO bytes has been saved by storage Index.

SQL> select name, value from v$mystat s, v$statname n where n.statistic# = s.statistic# and name like '%storage%';

NAME                                                                  VALUE
---------------------------------------------------------------- ----------
cell physical IO bytes saved by storage index                             0

Enabling the storage index

SQL> alter session set "_kcfis_storageidx_disabled"=false;

Session altered.

Executing command on employee table

SQL> select ID from EMPLOYEE where ID is null;

no rows selected

Now we will check how much bytes has been saved by storage index with below query.

SQL> select name, value from v$mystat s, v$statname n where n.statistic# = s.statistic# and name like '%storage%';

NAME                                                                  VALUE
---------------------------------------------------------------- ----------
cell physical IO bytes saved by storage index                      48684323

Above value shows that we have saved 48684323 IO bytes saved with storage index while executing the query.

Exadata Storage Index Part-1

  • It is not like traditional index which is in database. This is the feature of storage software and useful to eliminate IOs.
  • Storage Index filter out data from the consideration. 
  • It works by storing minimum and maximum column values for disk storage units, which are 1 Megabyte (MB) by default.
  • Because SQL predicates are passed to the storage servers when Smart Scans are performed, the storage software can check the predicates against the Storage Index metadata (maximum and minimum values) before doing the requested I/O.
  • Any storage region that cannot possibly have a matching row is skipped.
  • There are no tuning technique or documents for storage index as it maintain by storage software only.
Storage Indexes consist of a minimum and a maximum value for up to eight columns. This structure is maintained for 1MB chunks of storage (storage regions). Storage Indexes are stored in memory only and are never written to disk.

Storage index is not static, whenever we shutdown or reboot storage server it will be removed and once server comes up storage server automatically create index.

Now we will see that how storage index actually work while executing the query on the DB server.

Click on image to enlarge.

Now we are executing below query on this table.

SQL> select * from TABLE where c>7;

In this case first query will go to the storage index and it will check min and max number for rows of first set but 7 is not fitting in Min and Max for rows of first set so, it will eliminate to check the data of that rows and it will move to second set of rows. 

Configure User Equivalence between Database Servers

To configure user equivalence between two servers just follow below command.

On Server 1

Login as Oracle user :

$/usr/bin/ssh-keygen -t rsa

$/usr/bin/ssh-keygen -t dsa
$cd /home/oracle/.ssh/
$touch authorized_keys
$cat >> authorized_keys
$cat >> authorized_keys

on Server 2

Login as Oracle user:

$/usr/bin/ssh-keygen -t rsa

$/usr/bin/ssh-keygen -t dsa
$cd /home/oracle/.ssh/
$scp oracle@server1:/home/oracle/.ssh/authorized_keys  /home/oracle/.ssh/.

$cat >> authorized_keys
$cat >> authorized_keys

$scp authorized_keys  oracle@server1:/home/oracle/.ssh/

Now check user equivalence by doing ssh between both servers by login with oracle user. It will not ask password while login to another server for Oracle user.

              User Equivalence between Cell Storage and Database Servers