I have gathered frequency histogram manually on one of my column of a table to provide more information to optimizer for better calculation of cardinality.
Now i have my weekend job runs for gathering stats on schema level with method_opt as 'For all column size repeat'. But i don't want the stats of above column to be overridden by the stats job. I don't want to lock the statistics of whole table, but i just want to lock the column level stats for this table.
I have a transaction table with some custom properties and two status columns. There are 2 different applications(.Net and Pl/SQL Procedure) using the table. Both the process run parallel and fetch records one by one, perform some calculation and update the status column.
There are likely more chances that both applications will fetch the sane record and try to update the same row. This will cause a lock. Can i use row level lock before update by each application? Or is there any other methods/process in which this can be handled. ?
I want to write a sql query which will fetch the data from manual_temp_master and manual_temp_detl.But from manual_temp_detl table, Price_bkt_cds columns should be displayed as columns. Like the should look like as below:
we know we can see lock mode held in session can be analysed using LM column in v$lock.But i confused in seeing LM column it all shows in numbers from 0 to 6.
what is the difference on DBMS_STATS for table level and partition level , which will provide the best optimizer . If the table xxxx is partitioned from 1 to 10 ,then running gather stats on table xxxx as whole table level or partition level which will provide best result on the performance.
Can you take an incremental backup level 1 or level 0 without archivelogs?
syntax would bebackup as compressed backupset cummulative level 1 database.
The reason I ask is because when I run backup as compressed backupset cummulative level 1 database plus archivelogs # it runs fine, but when I run backup as compressed backupset cummulative level 1 database it just hangs.
between statement level or row level trigger, which trigger will execute first.We have BEFORE_UPDATE_ROWLEVEL_TRIGGER and BEFORE_ UPDATE_ STATEMENT LEVEL_TRIGGER triggers on table product.
I want to take the statistics of my Production database and import to my local database, to calculate the Production statistics.I have used the statistics=compute, to export statistics. In the log file, for some tables, there was a waring like "exp-00091 exporting questionable statistics"
Is this dump will be useful for me to calculate production statistics?What option i have to use while importing statistics=recalculate or statistics=safe?
I have large partitioned tables ( 4 partitions are added every month ). Is is possible to collect Incremental Statistics Gathering on these objects ( 9i ). If I collect stats with Ggranularity => ALL and ESTIMATE_PERCENT =100 the stats are accurate but it takes so much time .
One way may be to collect stats as Ggranularity => PARTITION for each new partition ( this quite fast ). but what about the Global Table Stats?
In the article regarding gathering CBO Statistics, it states: QUOTE When an Oracle database is created, a job will be scheduled that will generate the database statistics for you. You will still need to collect system statistics however, as these are not collected by the automatic statistics gathering mechanism.
what is the difference between "database statistics" and "system statistics"? In other words, do I need to run this script for each schema owner in my 10g/11g instance?
variable whoami varchar2(20); begin select user into :whoami from dual; end; exec dbms_stats.gather_schema_stats( - ownname => :whoami, - options => 'GATHER AUTO', - estimate_percent => 15, - cascade => true).
I am using Oracle 11g R2 version.I want to import the DB statistics. But i am getting an exception when i execute the command DBMS_STATS.IMPORT_SCHEMA_STATS ('user1','STATS_INFO', '','', TRUE, FALSE).
The error is ORA-20000: no statistics are imported ORA-06512: at "SYS.DBMS_STATS", line 10603 ORA-06512: at line 1.
The privileges 'ANALYZE ANY' and 'ANALYZE ANY DICTIONARY' is already given to the user.Also i executed this command as sys. But still error occurs.
Same command is successfully executed in Oracle 10g. Is there any difference in importing the statistics in Oracle 10g and 11g ?
updating the statistics for a table (with GATER_TABLE_STATS) and using NUM_ROWS then. This works fine for me as long as I am the owner of the table, but when someone else is, I always get this error: ORA-20000: Table does not exist or insufficient privileges.what privileges do I need to use GATHER_ TABLE_ STATS on all Tables, which were created by Users?
when I tried to use ANALYZE TABLE TEST_TABLE COMPUTE STATISTICS on a certain table I got the following error: a view is not appropriate here. The strange thing is, TEST_TABLE is not a view (at least it is not listed in ALL_VIEWS and is listed in ALL_TABLES, so it cant be a view right?).
Besides, is there another way to gather Table Statistics (not using Analyze Table or Gather_Table_Stats)?
When I launch it from command line it stops with error message: "insufficient privileges" and ask me the user, so I put 'PDMUSER' that is my user, it asks also pwd, I put it and the works.
I have a table which have 300+ columns and have 13 million rows. It is on a 32 kb block size. This is a table in data ware house environment. There no# of rows in the table haven't changed much but I see that the time taken to collect statistics have increased significantly.Initially it took only 15 minutes (with the same 13M rows) now it runs for 4+ hours. The max parallel servers is 4 (which is unchanged). The table is not partitioned.
OS: HP UX Itanium Database: Oracle 11g (11.2.0.2)
Command is: exec dbms_stats.gather_table_stats(ownname=>'ABC',tabname=>'ABC_LOAD',estimate_percent=>dbms_stats.auto_sample_size,cascade=>TRUE,DEGREE=>dbms_stats.auto_degree);
I would like to understand:
1) What could have been the causes of this change in time. 15 minutes to 4+hours ? 2) How can we gather statistics of huge table at a faster rate?
I have to create some indexes in a production database. Do I need to Compute Statistics after creating indexes? Or when I create they automatically are computed?
The version I'm using is:
Oracle Database 10g Release 10.2.0.5.0 - 64bit Production PL/SQL Release 10.2.0.5.0 - Production CORE 10.2.0.5.0 Production TNS for 64-bit Windows: Version 10.2.0.5.0 - Production NLSRTL Version 10.2.0.5.0 - Production
I have created materialized view which hold few million records.Should i have to analyse the view and compute the statistics after i create the materialized view?
Also,just in case i need further indexing,should i have to take the statistics for the table again??
Is it possible to gather stats for a schema which its in use. When i try to analyze the tables of a schema it shows that the statistics for that table is locked. So is it possible that instead of analyzing a table one by one , can i go for gathering the Schema stats while the objects of that Schema is still in use ( like DML or select statements being issued on those schema objects) .
DB version : 10.2.0.4 OS version : RHEL 5.8 DB type : RAC
I don't know, if this is the intent behavior of oracle or not. But i noticed, my queries Execution plan randomly changes after statistics collection. Several tables are truncated after the daily run at 8AM and statistics gathered for all the tables in that schema.
However execution plans for 2-3 sql statements always changes after this and performance is brought back to normal by executing the procedure by explicitly calling it from the command line with arguments instead of bind variables.
I want to get the stale stats for Table resides at APPS schema. Is there is any table or view exists to get the details like DBA_STALE_STATS or anything? Currently I am checking LAST_ANALYZED column from DBA_TABLES?
From the below query i found that there are some stale stats for 3 tables.
================================= select table_name, stale_stats, last_analyzed from dba_tab_statistics where owner= 'SYSADM' and stale_stats='YES' order by last_analyzed desc
I collect stats for those above 3 tables with dbms_stats.gather_table_stats().But no luck.After collection of stats immediately I ran the above query.But still it is showing there are stale stats for 3 tables.
how can I change "STALE-STATS" status, so that optimizer can use the updated stats eficiently.
