We know that when you multiplex a control file or redolog member to another physical disk, then you effectively have an active backup of the file. The idea being it is unlikely for physical corruption to occur on multiple physical devices at the same time. So if a physical corruption were to occur you still have a multiplexed file on another physical device, that will be unaffected, and you can use this multiplexed file to recover and open the database again. Also after the physical corruption has been resolved you can copy the uncorrupt multiplexed file(s) to a new physical location and establish the multiplexing again. I already have confirmation from Oracle support that multiplexing only protects against physical corruption of the control and redolog files - and not logical corruption of these files. Oracle support also told me they acknowledge that multiplexing is unnecessary where hardware redundancy (eg SAN RAID5) is already being used, however they still recommend multiplexing to give some protection against accidental dropping of control and redolog files. My question - can there be any other justification for multiplexing control and redologs where hardware redundancy (eg SAN RAID5) is already being used?
A listener.ora file is not required, if the listener is running on the default TCP port of 1521. My question is Without listener.ora file, is it possible for oracle to know which port to listen when multiple listeners present? Any help will be highly appreciated.
i have one query which return million records in 2 mins without group by function. i have to group these records so i have written group by clause on this query ,after writing group by it's taking 15-16 min. i tuned this query in first or second run ,it's take 2-3 min. after that it's taking 8-10 min to run. please tell me what could be the reason...
Sir, Can you please highlight the usage of No_CPU_Costing hint. My system is a large Retail database and in many places we have huge queries. So optimization of queries is the topmost priority in my job. Many times I wonder in my queries if I would have used the hint No_CPU_Costing hint, my plan cost got lowered drastically and in certain cases, even the execution time is very less compared with the original. But till now I resisted myself in using this particular hint, since I don't have a clear idea whether this can be used extensively. Can you please advice about its usage. Thanks, K.Vidhyasagar email@example.com
Hello, there exists a "standard" 10g sql select (...without pl/sql) of the plan_table (or v$sql_plan) table that additionally displays the order of execution (column exord) as part of the output and i have it ! cannot be,yes,no, you have it since long time ? e.g.: ID|EXORD|OPERATION |OPTIONS |OBJECT_NAME ---|-----|------------------------------|---------------|-------------------- 0| 21|SELECT STATEMENT | | 1| 20|SORT |ORDER BY | 2| 19|++NESTED LOOPS | | 3| 17|++++NESTED LOOPS | | 4| 15|++++++NESTED LOOPS | | 5| 12|++++++++NESTED LOOPS | | 6| 9|++++++++++NESTED LOOPS | | 7| 6|++++++++++++NESTED LOOPS | | 8| 2|++++++++++++++TABLE ACCESS |BY INDEX ROWID |REGIONS 9| 1|++++++++++++++++INDEX |UNIQUE SCAN |REG_ID_PK 10| 5|++++++++++++++INLIST ITERATOR | | 11| 4|++++++++++++++++TABLE ACCESS |BY INDEX ROWID |DEPARTMENTS 12| 3|++++++++++++++++++INDEX |UNIQUE SCAN |DEPT_ID_PK 13| 8|++++++++++++TABLE ACCESS |BY INDEX ROWID |LOCATIONS 14| 7|++++++++++++++INDEX |UNIQUE SCAN |LOC_ID_PK 15| 11|++++++++++TABLE ACCESS |BY INDEX ROWID |COUNTRIES 16| 10|++++++++++++INDEX |UNIQUE SCAN |COU_ID_PK 17| 14|++++++++TABLE ACCESS |BY INDEX ROWID |EMPLOYEES 18| 13|++++++++++INDEX |RANGE SCAN |EMP_DEPARTMENT_IX 19| 16|++++++INDEX |UNIQUE SCAN |JOB_ID_PK 20| 18|++++TABLE ACCESS |BY INDEX ROWID |JOBS why isn't exord not a column of the plan_table (v$sql_plan) table, think it could help people in understanding explain plans ? Regards, Joachim
How can you purge ex employees hr related data in order to meet data protection and data retention policies? There is no concurrent request or process to do it. Version 12.0.6. We want to do this in batches with date parameters not by using the delete function. Thanks
oracle provide wrap utility and DBMS_DDL.WRAP Function to encrypt pl/sql code. its not possible to unwrap again. but on following site its easily possible just paste the encrypted code and u got unwrap code on only one click http://www.codecrete.net/UnwrapIt/ please tell me if any one have any other option or method for encrypt ps/sql code.
after i installed oracle linux 6, i discovered that there is no office application on it, is there a way i can install an office suite? if yes, please recommend Thanks Tim.
Hello all, first of all, I wanted to tell you that this is not a challenge. I have tried to find a solution for my below question, but I couldn't. I asked for help on some forums like orafaq and OTN but no one helped me. I am expecting atleast some help here. I have set up a single instance standbys for rac databases. Now, I need to offload the backups from the primary on to the standbys. I learnt a few points to achieve this. 1. cancel managed recovery 2. connect to target(standby) and catalog and backup the standby 3. put standby in managed recovery mode. I think following these steps, I can restore primary(?). Now, My question is, how can I use these standby backups to clone/refresh databases? I tried it by connecting to target(primary), catalog and auxiliary but rman is using the primary backups instead of standby's to refresh the auxiliary database. If anyone can help me with their suggestions that would be great. I have a 11202 oracle on a linux box. Few followups to my question at this link http://www.orafaq.com/forum/t/176696/0/
The Oak Table challenge is not currently live - it goes active when a group of members of the OakTable arrange to meet at a public event - but since that wasn't clear when you raised the question, here are a couple of thoughts.
First - The Oracle version number really matters in this one; there have been bugs with latching on the root block of an index - the most recent that I know of being when an index had been rebuilt. See this blog item
Second - is the latching problem on the root block or on the leaf blocks ? If it's only on the root block then trying to spread the index entries across multiple leaf blocks isn't going to help anyway
Third - Using pctfree to limit the number of rows per block only works when you create or rebuild the index; if you're deleting and inserting rows in the batch then the pattern of activity may simply leave you with well-packed blocks after a very short time. (Possibly this is why you are rebuilding the index every two minutes - but that's likely to cause other, more significant, overheads anyway).
Fourth - your question suggests that you're only using select statements against this table: if that's true I can't see how rebuilding the index every two minutes could help - unless you're on an old version of Oracle that had bugs that generated excessive read-consistent copies of blocks.