Finding and fixing block corruption in oracle 10g

10.2.0.5.6
OS: Hp-unix
databases files on RAW.
DB Size: 10 TBs+
We had a SAN outage recently. The DB is back online. We want to check for block corruption to be on the safe side. We are planning to take a SAN EMC BCV copy of prod to run this. Looking for opinioms on the best way.Production is up and running. This is just a sanity check.
We are NOT using RMAN for backups. We do a BCV copy and then back that up and archive logs to tape.
RMAN: Can we do this from the control files? Do we have to set anything up? Is this the best way?
DBMS_REPAIR: I have not used this. We don't even have the package installed. I can install it.
DB_VERIFY: not sure if this is a good option or not. Is this current? I know I'll have to do it file by file and grep the logs. I have used this in the past, but its been a while.
Performance issues, CPU, I/O don't matter. We are doing a BCV copy and then mounting the DB on a new server.

Guess2 wrote:
10.2.0.5.6
OS: Hp-unix
databases files on RAW.
DB Size: 10 TBs+
We had a SAN outage recently. The DB is back online. We want to check for block corruption to be on the safe side. We are planning to take a SAN EMC BCV copy of prod to run this. Looking for opinioms on the best way.Production is up and running. This is just a sanity check.
We are NOT using RMAN for backups. We do a BCV copy and then back that up and archive logs to tape.
RMAN: Can we do this from the control files? Do we have to set anything up? Is this the best way?
DBMS_REPAIR: I have not used this. We don't even have the package installed. I can install it.
DB_VERIFY: not sure if this is a good option or not. Is this current? I know I'll have to do it file by file and grep the logs. I have used this in the past, but its been a while.
Performance issues, CPU, I/O don't matter. We are doing a BCV copy and then mounting the DB on a new server.
bcm@bcm-laptop:~$ dbv -help
DBVERIFY: Release 11.2.0.1.0 - Production on Mon May 21 07:29:42 2012
Copyright (c) 1982, 2009, Oracle and/or its affiliates. All rights reserved.
Keyword     Description                    (Default)
FILE        File to Verify                 (NONE)
START       Start Block                    (First Block of File)
END         End Block                      (Last Block of File)
BLOCKSIZE   Logical Block Size             (8192)
LOGFILE     Output Log                     (NONE)
FEEDBACK    Display Progress               (0)
PARFILE     Parameter File                 (NONE)
USERID      Username/Password              (NONE)
SEGMENT_ID Segment ID (tsn.relfile.block) (NONE)
HIGH_SCN    Highest Block SCN To Verify    (NONE)
            (scn_wrap.scn_base OR scn)           can be done against any datafile; open or closed, production or clone
Handle:     Guess2
Status Level:     Newbie
Registered:     Aug 5, 2000
Total Posts:     454
Total Questions:     212 (201 unresolved)
WHY so MANY unanswered questions?
Edited by: sb92075 on May 21, 2012 7:31 AM

Similar Messages

Does oracle import datapump fix block corruption?

Dear all,
I am having a data block corruption on production. I want to export this DB and import it to a test environment using datapump, in order to do some tests on it.
However I am concerned whether impdp will resolve this scenario... and therefore I will lose my test case...
so will oracle import datapump fix block corruption?
Thank you

so will oracle import datapump fix block corruption?.No. It detects Logical corruption (corruption and the corrupted segment will not be exported or imported.
Fix the issue by identifying the segment.
Get the block information from your alert.log
select segment_type,owner.segment_name
from dba_extents
where file_id = [&file_id] and [&block] between block_id and block_id+blocks -1; if it is index then drop and recreate index
if it is table and you have backup of that table then restore backup on another database and exp/imp the table.

Diff between logical and physical block corruption

What is the difference between Physical and Logical block corruption.
Dbverify utility, analyze command is used to check the logical block corruption not the physical one am i correct??
When i get
ORA-01578: ORACLE data block corrupted (file # 9, block # 13)
ORA-01110: data file 9: '/oracle/dbs/tbs_91.f'
ORA-01578: ORACLE data block corrupted (file # 2, block # 19)
ORA-01110: data file 2: '/oracle/dbs/tbs_21.f'
How to conform that this a logical or physical block corruption???
please through some light regarding this....
kumaresh

the following link may help u
http://download-east.oracle.com/docs/cd/B19306_01/backup.102/b14191/rcmconc1012.htm

How to Find the Current Instance Id in Oracle 10g

Hi,
how to find the current instance id in oracle 10g studio,
I have the below code :
if (InstanceScreenFlow.processInstance.id.id.indexOf(text : ins.id) < 0) {
/// do something
it is saying that "'processInstance' is not a function"
Thanks,
Brijesh Kumar Singh.

Why not just pass in the "id" predefined variable into your screenflow? Once you do that you'll be able to refer to it as "id.id" anywhere inside your screenflow.
Dan

How to genrate and check wait statistics in oracle 10g

Dear all,
how to genrate and check WAIT STATISTICS in oracle 10g on RHEL4.
Regards,
Ali

Oracle Kernel keeps track of these statistics. Query v$waitstat for overall system level wait statistics or v$session_wait for individual session wait statistics.

Maximim Number of process and session can set in oracle 10g

my os is 32 bit windows and ram size is 10gb and i am using AWE_WINDOW_MEMORY setting and now my sga is 6 gb.what would be Maximim Number of process and session can set in oracle 10g database

mithun wrote:
my os is 32 bit windows and ram size is 10gb and i am using AWE_WINDOW_MEMORY setting and now my sga is 6 gb.what would be Maximim Number of process and session can set in oracle 10g databaseThe maximum number of sessions is controlled by your database's SESSIONS parameter. This value is most derived from the PROCESSES parameter. You cannot exceed this number of sessions unless you raise the parameter value.
If your database is configured to handle enough sessions, then Oracle can literally handle thousands of simultaneous connections, provided your database server also has enough resources to handle this number of connections. You will want to implement Oracle's Shared Server connections so that multiple connections share a pool of processes on the database server. This minimizes the resource consumption used by the connections.
Read the following document as well.
http://download.oracle.com/docs/cd/B19306_01/server.102/b14231/manproc.htm
Thanks

Where can I find the deployed EJB files in Oracle 10g AS(904)

Hello All,
Where can I find the deployed ejb files, stubs and skels in the Oracle 10g (904)
When I try to run my ejb application I m getting following exception:
com.evermind.server.rmi.OrionRemoteException: Error (de-)serializing object: org.apache.log4j.Logger
I am not sure what is causing this error. I am not using any logging in my EJB. But the logging framework is configured in the application server.
Your input would be a great help to me
Thank you

Thanks Carty,
I am not creating Connection from EJB. I have a POJO called DBManager that looks for the DB Connection and return that to either EJB or other POJO that requests for Connection.
I am having a private method that get the connection from the DBManager because while writing UnitTest cases I can mockup getting Connection as my Unit test case will be a POJO again..
I have a question here regarding making DBManager as Singleton class. In my current project, I have configured my database in the application server as getting the connection by lookup for datasource through JNDI. It internally maintains all logical connection and always have one physical connection. In such scenario how implementing singleton will be beneficial?
Thank you for answering all my questions.

Solaris 10 and Hitachi LUN mapping with Oracle 10g RAC and ASM?

Hi all,
I am working on an Oracle 10g RAC and ASM installation with Sun E6900 servers attached to a Hitachi SAN for shared storage with Sun Solaris 10 as the server OS. We are using Oracle 10g Release 2 (10.2.0.3) RAC clusterware
for the clustering software and raw devices for shared storage and Veritas VxFs 4.1 filesystem.
My question is this:
How do I map the raw devices and LUNs on the Hitachi SAN to Solaris 10 OS and Oracle 10g RAC ASM?
I am aware that with an Oracle 10g RAC and ASM instance, one needs to configure the ASM instance initialization parameter file to set the asm_diskstring setting to recognize the LUNs that are presented to the host.
I know that Sun Solaris 10 uses /dev/rdsk/CwTxDySz naming convention at the OS level for disks. However, how would I map this to Oracle 10g ASM settings?
I cannot find this critical piece of information ANYWHERE!!!!
Thanks for your help!

You don't seem to state categorically that you are using Solaris Cluster, so I'll assume it since this is mainly a forum about Solaris Cluster (and IMHO, Solaris Cluster with Clusterware is better than Clusterware on its own).
Clusterware has to see the same device names from all cluster nodes. This is why Solaris Cluster (SC) is a positive benefit over Clusterware because SC provides an automatically managed, consistent name space. Clusterware on its own forces you to manage either the symbolic links (or worse mknods) to create a consistent namespace!
So, given the SC consistent namespace you simple add the raw devices into the ASM configuration, i.e. /dev/did/rdsk/dXsY. If you are using Solaris Volume Manager, you would use /dev/md/<setname>/rdsk/dXXX and if you were using CVM/VxVM you would use /dev/vx/rdsk/<dg_name>/<dev_name>.
Of course, if you genuinely are using Clusterware on its own, then you have somewhat of a management issue! ... time to think about installing SC?
Tim
---

Encryptind and decrypting database column in oracle 10g

hi guys...
i am sai sandeep,i got a doubt how to encrypt a database column in oracle 10g..?
i am using a table " emp_uid " ,and strtucture as follows,
create table emp_uid(user_id varchar2(20),pwd varchar2(20));
i need to encrypt a pwd column in the emp_uid.
how to do it..?
thanking u advance.....

Ok, here's a basic example...
SQL> create table myusers (username varchar2(30), password varchar2(40));
Table created.
SQL> create or replace procedure add_user(username in varchar2
2                                      ,password in varchar2) is
3 begin
4    insert into myusers (username, password)
5      values (add_user.username
6             ,dbms_crypto.hash(utl_raw.cast_to_raw(add_user.username||'!'||add_user.password)
7                              ,dbms_crypto.hash_sh1)
8             );
9    commit;
10 end;
11 /
Procedure created.
SQL> exec add_user('Fred','Fr3ddy')
PL/SQL procedure successfully completed.
SQL> select * from myusers
2 /
USERNAME                       PASSWORD
Fred                           E5C975DB4C0A1CF65683E36421A6305F09F4EA9A
SQL> set serverout on;
SQL> create or replace procedure loginuser(username in varchar2
2                                       ,password in varchar2) is
3    v_hash     varchar2(40);
4    v_username varchar2(30);
5 begin
6    v_hash := dbms_crypto.hash(utl_raw.cast_to_raw(loginuser.username||'!'||loginuser.password), dbms_crypto.hash_sh1);
7    select username
8    into   v_username
9    from   myusers
10    where username = loginuser.username
11    and    password = v_hash;
12    dbms_output.put_line('User: '||v_username||' logged in.');
13 exception
14    when no_data_found then
15      dbms_output.put_line('Username/Password is not valid!');
16 end;
17 /
Procedure created.
SQL> exec loginuser('Fred','Freddy');
Username/Password is not valid!
PL/SQL procedure successfully completed.
SQL> exec loginuser('Fred','Fr3ddy');
User: Fred logged in.
PL/SQL procedure successfully completed.
Ideally you would do the hashing of the password inside the client side application so only the Hashed value goes over the network, but the above demonstrates the principle of using hashes to store passwords. Because it's a one way algorithm, only a brute force method can be used to try and determine the original password. There is no way to directly un-hash the value. To check for a valid login, we don't retrieve the password and try to unhash it to compare against what the user has supplied, we actually take what the user has supplied and hash that in the same way and then compare the hashes.
The point of including the username or some other data in the hashing process means that if two users have the same password, they will still have different hash values, so it won't be apparent they are the same passwords. In my example, the point of putting another character between the concatenation of username and password is in case the username and password together would give the same result e.g.
If we had one user "Fred" with password "Fr3ddy" then just concatenating the strings would give "FredFr3ddy".
If we had another user "FredF" and he happened to choose a password "r3ddy" then just concatenating those would also give "FredFr3ddy"
by introducing a known breaking character they would be different e.g. "Fred!Fr3ddy" and "FredF!r3ddy" and hence give different hash values.
That's the basics of how passwords are stored for security.
It would take a lot of processing power and brute force methods just to determine a single password for a single user when using hashing methods of security.
With encryption, a brute force method could be used to find the decryption key, and once found that could be used to decrypt ALL the encyrpted data, hence it is less secure, especially when some clever person will no doubt have written the key down somewhere so they don't forget it. With hashing there's no key to write down.

Find deleted information from user_jobs in Oracle 10g

Hello All,
I had scheduled a job in oracle 10g to run only once using dbms_jobs.
I wanted to find out how long the job took ( total_time) and what SQL it was running ( user_job.what) but the information about the job seem to get deleted from user_jobs once the job is completed.
Is there any other table/ view I can use to find these information?
Also, will it be possible to query the user_job as of previous time when the data was there? if yes, please also let me know what privilidges and setting is required. Currently, I just have basic developer priviledge.
Many thanks in advance for any help.

Information about how long a job took and what SQL it was running is not stored anywhere when you use DBMS_JOB. Once of the nice things about using DBMS_SCHEDULER rather than DBMS_JOB for job scheduling is that DBMS_SCHEDULER automatically logs timing information.
How long ago do you want to look at the data in USER_JOBS? If you are just trying to look back an hour or two ago, you may be able to use a flashback query. How far back you can go depends on your UNDO_RETENTION setting, the size of your UNDO tablespace, and the amount of UNDO that is generated by your system. The theoretical maximum you could flashback is 5 days, realistically, not many production databases can flashback more than a few hours unless someone has specifically identified the need to do longer-term flashback queries. The default UNDO_RETENTION is only 15 minutes. Plus, you're not guaranteed that you'll be able to flashback (that can be changed, but it's a pretty dangerous change because it will cause the database to stop if you run out of UNDO space).
Justin

Simple insert, delete and update not working after oracle 10g installation

Hello,
I made a new installation of Oracle 10g (10.2.0.1.0) and created a schema . Now i try to insert or update data using simple queries, but unfortunately nothing happens. I tried inserting or updating data even via sqlplus, but unfortunately the same result. Query takes infinite time to end, so i had to manually end it every time... No errors... No warnings... but takes long time... I have given all the required permissions to the schema but the same result. Can somebody help me figure out the problem?
Tried creating a new table and the insert and delete are working fine... think the problem is with this table... Imported the rows of the table from another online database. I dont understand how this might be a problem.
Edited by: user13333704 on Mar 15, 2011 1:26 AM

However how do you generate awr reports and how to find the session wait?TO generate the AWR report
sql> ?/rdbms/admin/awrrpt
http://download.oracle.com/docs/cd/B19306_01/server.102/b14211/autostat.htm#BABDHAGF
The event column in v$session , in v$active_session_wait will show you the session wait event.What the session was actually waiting for.
HTH
Anand

Windows and RAC block corruption

Hi,
we're using rac / windows / 9204 with a san and ocfs and we're getting loads of block corruptions (ora-1578).
We're using dbms_repair to isolate the problems and read 'past' the data, but this is obviously not ideal since we loose data.
Besides proprietary SAN disk checking programs, dbms_repair,analyze and dbverify - are there any other mechanisms one could use to check disk integrity?
Regards
Ottar.

user496020 wrote:
Hi,
The other day I was importing a csv file. Not long thereafter that I got an error message:
ORA-01578: ORACLE data block corrupted (file # 28, block # 757)
ORA-01110: data file 28: '/u02/oradata/APEX/APX_APPL01_02.dbf'Your database is corrupt, you need to perform some recovery on the affected data files, etc.

Executing Procedure in SQL SERVER and Storing the results in Oracle 10g

Hello,
I am trying execute SQL SERVER procedure from Oracle 10g and store the results in a oracle table. When I tried executing the procedure i am getting errors.
I am using Oracle Heteroeneous Services.
Steps I followed.
1. Created ODBC DSN on Oracle Server connecting to SQL SERVER database.
2. Created Listner entry and TNSNAMES entries.
3. Created Database link and able to select the data from the SQL SERVER tables.
Here is the code i tried to execute the procedure.
BEGIN
"META"."extract"@abc;
END;
"META"."extract"@abc;
ERROR at line 2:
ORA-06550: line 2, column 1:
PLS-00201: identifier 'META.extract@ABC' must be declared
ORA-06550: line 2, column 1:
PL/SQL: Statement ignored
Please help me.
Thank You,
Seshadri Thope

Hi thopevs,
Can you please tell me the right syntax of calling procedures(on sql server) from oracle?
I am getting following error:
SQL> execute "GetdateSys"@oratosql;
begin "GetdateSys"@oratosql; end;
ORA-06550: line 2, column 7:
PLS-00201: identifier 'GetdateSys@ORATOSQL' must be declared
ORA-06550: line 2, column 7:
PL/SQL: Statement ignored
SQL>
Your help will be highly appreciated.
Thanks & Regards,
M.U.N.A

Ora 600 and ora 7445 error in oracle 10g r2

hi all,
we are using Oracle 10g r2.
os is windows 2003 server,
ram is 4 gb.
front end is java .
we got a below mentioned errors.
i got the RA-00600: internal error code, arguments: [qerghFetch:inc], [], [], [], [], [], [], [],
and ORA-07445: exception encountered: core dump [ACCESS_VIOLATION] [unable_to_trans_pc] [PC:0x61F04AC9] [ADDR:0xC] [UNABLE_TO_READ] []
and Error 0 in kwqmnpartition(), aborting txn errors.
for aug 3 11.42.47 we got a ora 600 error after that through front end (java) users can't able to connect the database.the not able to executed the query session disconnected error.
what is reason for theis error . how it will rectifiy.?
pls advice.Below i attached the alret log message .
Sun Aug 02 04:04:24 2009
Error 0 in kwqmnpartition(), aborting txn
Sun Aug 02 05:48:10 2009
Error 0 in kwqmnpartition(), aborting txn
Sun Aug 02 05:49:10 2009
Error 0 in kwqmnpartition(), aborting txn
Sun Aug 02 07:38:49 2009
Thread 1 advanced to log sequence 1732
Current log# 3 seq# 1732 mem# 0: E:\ORACLE\PRODUCT\10.2.0\ORADATA\ORCL\REDO03.LOG
Sun Aug 02 07:44:14 2009
Error 0 in kwqmnpartition(), aborting txn
Sun Aug 02 07:45:14 2009
Error 0 in kwqmnpartition(), aborting txn
Sun Aug 02 22:00:26 2009
Thread 1 advanced to log sequence 1733
Current log# 1 seq# 1733 mem# 0: E:\ORACLE\PRODUCT\10.2.0\ORADATA\ORCL\REDO01.LOG
Mon Aug 03 07:08:15 2009
Thread 1 advanced to log sequence 1734
Current log# 2 seq# 1734 mem# 0: E:\ORACLE\PRODUCT\10.2.0\ORADATA\ORCL\REDO02.LOG
Mon Aug 03 09:28:23 2009
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Mon Aug 03 09:30:27 2009
Error 0 in kwqmnpartition(), aborting txn
Mon Aug 03 11:42:47 2009
Errors in file e:\oracle\product\10.2.0\admin\orcl\udump\orcl_ora_6012.trc:
ORA-00600: internal error code, arguments: [qerghFetch:inc], [], [], [], [], [], [], []
Mon Aug 03 13:20:00 2009
Thread 1 advanced to log sequence 1735
Current log# 3 seq# 1735 mem# 0: E:\ORACLE\PRODUCT\10.2.0\ORADATA\ORCL\REDO03.LOG
Mon Aug 03 16:07:07 2009
Errors in file e:\oracle\product\10.2.0\em_2\rdbms\trace\orcl_ora_4756.trc:
ORA-07445: exception encountered: core dump [ACCESS_VIOLATION] [unable_to_trans_pc] [PC:0x61F04AC9] [ADDR:0xC] [UNABLE_TO_READ] []
Mon Aug 03 16:07:11 2009
Process m000 died, see its trace file
Mon Aug 03 16:07:11 2009
ksvcreate: Process(m000) creation failed
Mon Aug 03 16:08:37 2009
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Mon Aug 03 16:10:40 2009
Error 0 in kwqmnpartition(), aborting txn
Mon Aug 03 17:25:05 2009
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Mon Aug 03 17:27:11 2009
Error 0 in kwqmnpartition(), aborting txn
Mon Aug 03 17:32:19 2009
Error 0 in kwqmnpartition(), aborting txn
Mon Aug 03 17:33:24 2009
Error 0 in kwqmnpartition(), aborting txn
Mon Aug 03 18:52:08 2009
Error 0 in kwqmnpartition(), aborting txn
Mon Aug 03 18:54:09 2009
Error 0 in kwqmnpartition(), aborting txn
Mon Aug 03 19:05:33 2009
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Mon Aug 03 19:09:02 2009
Errors in file e:\oracle\product\10.2.0\em_2\rdbms\trace\orcl_ora_5132.trc:
ORA-07445: exception encountered: core dump [ACCESS_VIOLATION] [unable_to_trans_pc] [PC:0x7C34126B] [ADDR:0x0] [UNABLE_TO_WRITE] []
Mon Aug 03 19:10:27 2009
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn
Error 0 in kwqmnpartition(), aborting txn

Hi reshumakur
As this is your PRODUCTION system THEN_ The best help to offer in case of ORA-600, and ORA-7445 is to open SR with metalink. Submit with the SR the following
1- Alert log
2- Trace files (in ur case e:\oracle\product\10.2.0\admin\orcl\udump\orcl_ora_6012.trc: and e:\oracle\product\10.2.0\em_2\rdbms\trace\orcl_ora_4756.trc:)
3- Describe what was happening when you had these errors.
Try to avoid free help when you have ORA-600 and ORA-07445. This errors could be becuase of BUGS. Oracle Support will know this from the call stack in the trace file.
Then Do your best to patch 10.2.0.4 and the latest CPU patch.
Good Luck

Exporting database from oracle 11g and then importing it in oracle 10g

Hi! there every one...
please tell me the way to export the data and my data structures i.e the complete database from Oracle 11g to older version(oracle 10g) and how to import it there..
i'm moving the database from one machine to another without lan, preferably via storage device...

EXPORT FROM 11G to 10G.
1) log on to 11G box.
2) cd /<>/product/app/oracle/product/10.2/network/admin/
3) Get the connection string from tnsnames.ora for the respective instances
4) export ORACLE_HOME=/<>/app/oracle/product/10.2.0
5) export LD_LIBRARY_PATH=$ORACLE_HOME/lib
6) cd /<>/app/oracle/product/10.2.0/bin
7) #>./exp file= test.dmp ......
Username: user1@<connection_string>
Password:*******
8) Import as usual

Finding and fixing block corruption in oracle 10g

Similar Messages

Maybe you are looking for