Concept Automatic Undo Tuning

Hi all,
yesterday we exported a large table with datapump. Unfortantly we received an ora-01555. As undo_mangement was set to auto and also undo_tablespace was big enough + autoextensible, I always thought, that Oracle would adjust undo_rentention as needed (longest query + 5 Min) even though rention was set to 900.
Could anyone please put some light on the subject.
Thanks in adavance

http://download.oracle.com/docs/cd/B28359_01/server.111/b28310/undo002.htm#BJFFIACJ
Did you export a table with LOB segments? If so that is the cause. Automatic tuning of undo retention is not supported for LOBs.
Note:
Automatic tuning of undo retention is not supported for LOBs. This is because undo information for LOBs is stored in the segment itself and not in the undo tablespace. >For LOBs, the database attempts to honor the minimum undo retention period specified by UNDO_RETENTION. However, if space becomes low, unexpired LOB undo >information may be overwritten.

Similar Messages

  • Change automatic SQL Tuning Advisor time

    Hi All,
    can anyone help me in changing "automatic SQL Tuning Advisor" job timingsm because currently it is scheduled during peak hours , i want to change it to evening 6 pm daily.Because during peak hours it is taking more cpu and taking one hour.
    Wed May 08 08:00:10 2013
    Begin automatic SQL Tuning Advisor run for special tuning task "SYS_AUTO_SQL_TUNING_TASK"
    Wed May 08 08:05:47 2013
    thanks
    Mapps

    Hello,
    Here's what you can do,
    Create a new window to the time you want it to run:
    EXEC dbms_scheduler.create_window(window_name => 'YOUR_WINDOW_NAME',
    duration => numtodsinterval(1, 'hour'),
    resource_plan => 'DEFAULT_MAINTENANCE_PLAN',
    repeat_interval => 'FREQ=DAILY;BYHOUR=18;BYMINUTE=0;BYSECOND=0');>
    The above window would be in effect from 6 pm to 7 pm.
    Then, disable automatic sql_tuning and then re-enable it to run on the window you just created:
    EXEC DBMS_AUTO_TASK_ADMIN.DISABLE(client_name => 'sql tuning advisor',
    operation => NULL,
    window_name => NULL);>
    and now set it to run to your window:
    EXEC DBMS_AUTO_TASK_ADMIN.ENABLE(client_name => 'sql tuning advisor',
    operation => NULL,
    window_name => 'YOUR_WINDOW_NAME');

  • Automatic Checkpoint Tuning in Oracle 10g

    Hi ,
    How to confirm if the database has been configured for automatic checkpoint tuning...????
    Many thanks...
    Sim

    The log_checkpoint_timeout (default 1800) gives Oracle a time-based target, and the log_checkpoint_interval (default 0, meaning the action is based on the size of the current redo log file) give Oracle a 'volume-based' target.
    The target affects the rate at which the database writer writes dirty blocks to disk. Every three seconds the database writer wakes up and checks the tail-end of the checkpoint queue and writes out any blocks that were attached to the queue too far in the past.
    In this context, "too far in the past" means either "more seconds in the past than the[i] log_checkpoint timeout" or "made dirty by a change described in a redo block more than log_checkpoint_interval blocks back from the current block".
    If you set fast_start_mttr_target to a non-zero number of seconds then Oracle tries to treat that as a request to write out dirty blocks sufficiently rapidly that crash recovery will take no more than that number of seconds. In effect this makes Oracle change the internal value of log_checkpoint_interval and log_checkpoint_timeout dynamically every few seconds.
    Recoverty requires Oracle to apply the redo log from a known point - which means reading and applying a certain amount of redo to a number of "out of synch" blocks that have to be read from disk. So the fast_start_mttr_target code is constantly calculating how much redo would have to be read and applied, and how many data blocks would have to be read and fixed - and writing dirty blocks fast enough to make it possible to do that work within the given time limit.
    Setting the parameter to zero is (as it is with many numerical parameters) equivalent to not setting it at all. If you try to set the parameter to a very small value (say 30 seconds) you may end up finding lots of messages in your alert log saying something like: "I can't honour your request to recover in 30 seconds, the best I can do is XXX".
    Regards
    Jonathan Lewis
    http://jonathanlewis.wordpress.com
    http://www.jlcomp.demon.co.uk

  • Illegal rollback Segment operation in Automatic Undo mode

    Hi,
    I use oracle database 9.2.0.1 and I try to install SCM repository but when I run ckcreate I saw an error message in
    the log file that Illegal rollback Segment operation in Automatic Undo mode because oracle 9.2 use automatic undo mode .So how can i install the repository.
    Ghassan Hallaq

    ORA-30019: Illegal rollback Segment operation in Automatic Undo mode
    You have hit bug '2108361'. I think you could safely ignore it in your case.

  • Automatic checkpoint tuning in 10g?

    Does anyone have an idea with the following sentence?
    Automatic checkpoint tuning is enabled when FAST_START_MTTR_TARGET is not explicitly set to zero.
    I think the incremental checkpoint event was introduced in 9i...But I saw this automatic checkpoint tuning feature in the 10g OTN doc. what's new?
    Anyway!! we need to set this parameter with a non-zero value to use the incremental checkpoint event though.

    hi,
    Since Oracle 9i FAST_START_MTTR_TARGET parameter is the preferred method
    of tuning incremental checkpoint target. FAST_START_MTTR_TARGET enables you to specify the number of seconds the database takes to perform crash recovery of a single instance. Based on internal statistics, incremental checkpoint
    automatically adjusts the checkpoint target to meet the requirement of
    FAST_START_MTTR_TARGET.
    V$INSTANCE_RECOVERY.ESTIMATED_MTTR shows the current estimated mean time to
    recover (MTTR) in seconds. This value is shown even if FAST_START_MTTR_TARGET
    is not specified.
    V$INSTANCE_RECOVERY.TARGET_MTTR shows the effective MTTR target in seconds
    enforced by the system.
    V$MTTR_TARGET_ADVICE shows the number of I/Os resulted by the current workload
    under the current MTTR setting and the estimated number of I/Os that would be
    resulted by the current workload under other MTTR settings.
    so I would recommend you run this sql to find out what the system thinks your MTTR_TARGET should be.
    If FAST_START_MTTR_TARGET is specified, LOG_CHECKPOINT_INTERVAL should not
    be set or set to 0.
    rgds
    alan

  • Automatic SQL tuning errors

    Hi,
    Getting the following every time an automatic SQL tuning maintenance task runs... any ideas? The job fails as a result.
    This is 11gR2 RAC on Linux x64.
    ORA-00001: unique constraint (SYS.WRI$_ADV_SQLT_PLANS_PK) violated
    ORA-06512: at "SYS.PRVT_ADVISOR", line 2693
    ORA-06512: at "SYS.DBMS_ADVISOR", line 241
    ORA-06512: at "SYS.DBMS_SQLTUNE", line 772 ORA-06512: at line 4
    Thanks!
    Edited by: user9213912 on Nov 15, 2010 10:41 AM

    Hi,
    check this http://translate.google.co.in/translate?hl=en&sl=ko&u=http://kr.forums.oracle.com/forums/thread.jspa%3FthreadID%3D952867&ei=v3_hTPP4LYmxcbT-vZcM&sa=X&oi=translate&ct=result&resnum=2&ved=0CCYQ7gEwAQ&prev=/search%3Fq%3DSYS.WRI%2524_ADV_SQLT_PLANS_PK%26hl%3Den%26client%3Dfirefox-a%26rls%3Dorg.mozilla:en-US:official%26prmd%3Div

  • [OCP exam] doubt about Automatic SQL Tuning Advisor

    Hi,
    Sorry for my poor english.
    I am currently preparing for the OCP  11g exam.
    For that, I have bought some training tools  to help me prepare the exam.
    But, in the chapter related to the SQL tuning advisor, some explanation really confuse me .
    Extract :
    <<
    the SQL automatic tuning  job runs every night during the maintenance window.
    By default, the job makes SQL profile recommendations, tests them, and automatically implements them if they provide better performance improvement)
    >>
    I was persuaded it was the opposite.
    I check several databases/documentation, and it's seems to be always the opposite.
    By default,new sql profile are tested but not implemented automatically (you have to set a parameter ACCEPT_SQL_PROFILES TO TRUE to activate this option)
    But the exam is validated on oracle 11.2.0.1, and I only have 11.2.0.2/11.2.0.3 db available.
    So may be oracle changed the value of the default parameter in the new releases.
    Can someone confirm if I am right or not ?

    I would rather accept what is in the documentation:
    "If a SQL profile is recommended, the database tests the new profile by executing the SQL statement both with and without the profile. If the performance improvement improves at least threefold, then the database accepts the SQL profile, but only if the ACCEPT_SQL_PROFILES task parameter is set to TRUE. Otherwise, the automatic SQL tuning reports merely report the recommendation to create a SQL profile." Automatic SQL Tuning
    "If automatic implementation of SQL profiles is enabled (the default is disabled), then the database implements any SQL profiles that promise a great performance benefit. The implementation occurs at tuning time so that the database can immediately benefit from the new plan. You can enable or disable automatic implementation by using the SET_AUTO_TUNING_TASK_PARAMETER API to set the ACCEPT_SQL_PROFILES parameter." DBMS_AUTO_SQLTUNE

  • Automatic UNDO

    if we use automatic undo management in oracle database, may we still have a chance to get ERROR"Snapshot too old" or it may be avoided as Oracle manages undo segments itself....
    Regards

    If you are 9i this note is worth a look:Note:301432.1Performance Slowdown During Heavy Undo Segment Onlining
    I have discussed about this topic at my blog:
    http://jaffardba.blogspot.com/
    The question was about 'Sanpshot too old eror', not about the performance related to AUTOMATIC UNDO.
    One more parameter that may cause is undo_retention.
    The other work around to avoid is to run the most long query and use v$undostat view and set the undo tbs and undo_retention parameters.
    Jaffar

  • Automatic Undo Management - Shrink Undo Segments

    Hello guys,
    i have a question regarding the automatic undo management and some large transactions.
    So lets take the following example:
    1 transaction with a lot of updates (round about 20.00000 entries) . After the update a commit was executed.
    Now the corresponding undo segment was blown up to 640 MB.
    Now my questions:
    Will this segment be shrinked after it is not needed anymore (STATUS = EXPIRED)?
    If not and another transaction needs some undo and activates this segment again.. 640 MB are going online again... but the transaction now only needs some KB.
    I know the algorithm if the undo segment needs to be extended (metalinknote #413732.1).. but what is happening to "big" segements if they are not needed anymore ... are the deallocated/shrinked after a time period?
    Regards
    Stefan

    Here is a little example of a smaller undo segment.. which shows my doubt:
    SELECT SEGMENT_NAME,STATUS FROM DBA_ROLLBACK_SEGS WHERE SEGMENT_NAME = '_SYSSMU13$';
    SEGMENT_NAME                   STATUS          
    _SYSSMU13$                     OFFLINE      
    SELECT SEGMENT_NAME, EXTENT_ID, BYTES, STATUS from DBA_UNDO_EXTENTS WHERE SEGMENT_NAME = '_SYSSMU13$';
    SEGMENT_NAME                   EXTENT_ID              BYTES                  STATUS   
    _SYSSMU13$                     0                      65536                  EXPIRED  
    _SYSSMU13$                     1                      65536                  EXPIRED  
    _SYSSMU13$                     2                      8388608                EXPIRED  
    _SYSSMU13$                     3                      8388608                EXPIRED  
    _SYSSMU13$                     4                      8388608                EXPIRED  
    _SYSSMU13$                     5                      8388608                EXPIRED  
    _SYSSMU13$                     6                      8388608                EXPIRED  
    _SYSSMU13$                     7                      8388608                EXPIRED  
    _SYSSMU13$                     8                      8388608                EXPIRED  
    _SYSSMU13$                     9                      8388608                EXPIRED  
    _SYSSMU13$                     10                     8388608                EXPIRED  
    _SYSSMU13$                     11                     67108864               EXPIRED   If the other guys are correct, the extents of undo segment _SYSSMU13$ should be deallocated... but they still persist.
    Regards
    Stefan

  • Undo Block Allocation Algorithm in Automatic Undo Management

    Hi,
    Can someone please explain me the Undo Block allocation algorithm in Automatic Undo Management??
    I need to tune my Undo tablespace in my Prod database.
    regards,
    Arul Sekar

    Hi,
    Thanks for the link provided.....
    But the link which has been provided does not contain the details about the Undo block allocation alogorithm in AUM.
    Can you please help me in this regard??
    regards,
    Arul Sekar

  • About Automatic SQL Tuning.

    Hii all,
                   I have tuned one query by using SQL Tuning  Advisor in SQL Developer Tool. I have seen much improvement in Explain plan in advisor .. After tuning finished i executed following script.
    {code}
    select dbms_sqltune.report_tuning_task('staName47421') as Rems
    from dual;
    {code}
    I seen an explain plan but i am not able to see recommended changes in query .. Is it possible to get the query directly from any view ..
    I tried and checked in many dynamic views by keeping plan_hash_value as reference .. I found records in some dynamic views like dba_sqltune_plans and dba_advisor_sqlstats but i am unable to find exact sqltext .. Kindly advice if possible to get that query .. My query was taking 17 mins but it got reduced to 2 mins by seeing sql tuned explain plan so i want to see as what are the changes been done .
    Regards,
    Nitesh.

    I would rather accept what is in the documentation:
    "If a SQL profile is recommended, the database tests the new profile by executing the SQL statement both with and without the profile. If the performance improvement improves at least threefold, then the database accepts the SQL profile, but only if the ACCEPT_SQL_PROFILES task parameter is set to TRUE. Otherwise, the automatic SQL tuning reports merely report the recommendation to create a SQL profile." Automatic SQL Tuning
    "If automatic implementation of SQL profiles is enabled (the default is disabled), then the database implements any SQL profiles that promise a great performance benefit. The implementation occurs at tuning time so that the database can immediately benefit from the new plan. You can enable or disable automatic implementation by using the SET_AUTO_TUNING_TASK_PARAMETER API to set the ACCEPT_SQL_PROFILES parameter." DBMS_AUTO_SQLTUNE

  • Automatic undo shrink space

    Hello
    1-)
    Suppose undo tbs is 10gb.
    There is no activity in the database and hence Oracle decided to take the rollback segments offline.
    Does it also drop the rollbacksegments?
    I mean, is there a chance to reduce my undo tablespace 10gb to 8gb automatically?
    2-)
    I know that round robin is used in manuel undo, Is it same in AUM?
    Suppose
    transaction1 assigned to rbs1
    transaction2 assigned to rbs2.
    transaction3 assigned to rbs3
    Assume, transaction2 commited. and undo retention period is expired.
    Does transaction4 assigned to rbs2 or rbs1?

    John Carew wrote:
    2-)
    I know that round robin is used in manuel undo, Is it same in AUM?No. I haven't worked out in any detail how Oracle behaves, but it has some algorithm for minimising the number of undo segments used over time. (Essentially it seems to follow a "re-use recent victim" approach - which is why undo segments can end up being automaticallly taken offline and even (I think) dropped.
    Suppose
    transaction1 assigned to rbs1
    transaction2 assigned to rbs2.
    transaction3 assigned to rbs3
    Assume, transaction2 commited. and undo retention period is expired.
    Does transaction4 assigned to rbs2 or rbs1?I think it will go to rbs2 - because part of the selection algorithm aims to avoid having multiple transactions attached to a single undo segment if possible. (But, of course, there may be some other undo segment which is considered a better target for other reasons).
    Regards
    Jonathan Lewis
    http://jonathanlewis.wordpress.com
    http://www.jlcomp.demon.co.uk
    To post code, statspack/AWR report, execution plans or trace files, start and end the section with the tag {noformat}{noformat} (lowercase, curly brackets, no spaces) so that the text appears in fixed format.
    "Science is more than a body of knowledge; it is a way of thinking"
    Carl Sagan                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           

  • I could not set Automatic SQL Tuning enabled

    Hi experts,
    I execute below command by sys user.
    BEGIN
    dbms_auto_task_admin.enable(client_name => 'sql tuning advisor', operation => NULL, window_name => NULL);
    END;
    It occur error:
    ORA-15604: Initialization parameters prevent client from being enabled.
    ORA-06512: at "SYS.DBMS_AUTO_TASK_ADMIN", line 23
    ORA-06512: at "SYS.DBMS_AUTO_TASK_ADMIN", line 122
    Thanks!
    Joseph.

    Hi P. Forstmann ,
    Yes , I have the license.
    SQL> show parameter control_m
    NAME TYPE VALUE
    control_management_pack_access string DIAGNOSTIC+TUNING
    SQL> select * from v$version;
    BANNER
    Oracle Database 11g Enterprise Edition Release 11.2.0.2.0 - Production
    PL/SQL Release 11.2.0.2.0 - Production
    CORE 11.2.0.2.0 Production
    TNS for Linux: Version 11.2.0.2.0 - Production
    NLSRTL Version 11.2.0.2.0 - Production
    Rds,Joseph.

  • Automatic Undo Management, UNDO tablspace, and undo_segment count

    When I created my UNDO tablespace I was given 10 undo_segments:
    SQL> SELECT usn "Rollback Table", GETS, WAITS, xacts "Active Transactions"
    2 FROM V$ROLLSTAT;
    Rollback Table GETS WAITS Active Transactions
    0 1242 0 0
    1 88678 8 0
    2 83884 2 0
    3 82519 9 0
    4 85661 17 0
    5 80613 14 0
    6 82183 0 0
    7 92551 9 0
    8 78371 14 0
    9 89043 1 0
    10 75477 5 0
    The above statistic indicates I need more undo_segments. Do I need to recreate my UNDO tablespace in order to get the number of segments necessary to prevent waits or will more undo_segments be spawned eventually??
    I think I'm suffering Contention? Is there medicine?

    Compared to the number of GETS, the number of WAITS is very small. I doubt that you could ever have enough rollback segments to ensure that there were never any waits. Oracle's read consistency mechanism virtually guarantees that you will at least occasionally get waits.
    Aside from that, the point of UNDO tablespaces is that Oracle will manage the number and size of rollback segments based on the load. If you are really concerned about this, and if you have statistics indicating that waits for rollback segments are a performance bottleneck (something like a statspack report showing large amounts of time for one of the undo waits), then you can simply add more space to your undo tablespace, and Oracle will likely use it to add additional segmetns if it feels the need.
    John

  • Automatic UNDO Management

    Hi there,
    I have a Datawarehouse DB , with two UNDO tablespaces,Last friday we switched UNDO_TABLESPACE to UNDO01 and UNDO02 is not active.While querying DBA_UNDO_EXTENTS , status of all extents in UNDO02 is marked as EXPIRED but there is no free space left in UNDO02 tablespace.
    Now the qustion is
    1. why Oracle is not releasing EXPIRED extents from the NON-ACTIVE Undo tablespace?
    2. AS per Oracle Documentation "Switching Undo Tablespaces" - "The switch operation does not wait for transactions in the old undo tablespace to commit. If there are any pending transactions in the old undo tablespace, the old undo tablespace enters into a PENDING OFFLINE mode (status). In this mode, existing transactions can continue to execute, but undo records for new user transactions cannot be stored in this undo tablespace" ---- But STATUS column in DBA_TABLESPACES in my DB showing both Tablespaces are "ONLINE",
    Any thoughts????
    Thanks in advance

    SQL> select segment_name, owner, tablespace_name,status from dba_rollback_segs;
    SEGMENT_NAME OWNER TABLESPACE_NAME STATUS
    SYSTEM SYS SYSTEM ONLINE
    SYSROL SYS SYSTEM OFFLINE
    _SYSSMU2$                      PUBLIC UNDO1                          ONLINE
    _SYSSMU3$                      PUBLIC UNDO1                          ONLINE
    _SYSSMU4$                      PUBLIC UNDO1                          ONLINE
    _SYSSMU5$                      PUBLIC UNDO1                          ONLINE
    _SYSSMU6$                      PUBLIC UNDO1                          ONLINE
    _SYSSMU7$                      PUBLIC UNDO1                          ONLINE
    _SYSSMU8$                      PUBLIC UNDO1                          ONLINE
    _SYSSMU9$                      PUBLIC UNDO1                          ONLINE
    _SYSSMU10$                     PUBLIC UNDO1                          ONLINE
    _SYSSMU11$                     PUBLIC UNDO1                          ONLINE
    _SYSSMU12$                     PUBLIC UNDO1                          ONLINE
    _SYSSMU13$                     PUBLIC UNDO1                          ONLINE
    _SYSSMU14$                     PUBLIC UNDO1                          ONLINE
    _SYSSMU15$                     PUBLIC UNDO1                          ONLINE
    _SYSSMU16$                     PUBLIC UNDO1                          ONLINE
    _SYSSMU17$                     PUBLIC UNDO1                          ONLINE
    _SYSSMU18$                     PUBLIC UNDO2                          OFFLINE
    _SYSSMU19$                     PUBLIC UNDO2                          OFFLINE
    _SYSSMU20$                     PUBLIC UNDO2                          OFFLINE
    _SYSSMU21$                     PUBLIC UNDO2                          OFFLINE
    _SYSSMU22$                     PUBLIC UNDO1                          ONLINE
    _SYSSMU23$                     PUBLIC UNDO1                          OFFLINE
    _SYSSMU24$                     PUBLIC UNDO1                          OFFLINE
    _SYSSMU25$                     PUBLIC UNDO1                          OFFLINE
    _SYSSMU26$                     PUBLIC UNDO1                          OFFLINE
    _SYSSMU27$                     PUBLIC UNDO1                          OFFLINE
    _SYSSMU28$                     PUBLIC UNDO2                          OFFLINE
    _SYSSMU29$                     PUBLIC UNDO2                          OFFLINE
    _SYSSMU30$                     PUBLIC UNDO2                          OFFLINE
    _SYSSMU31$                     PUBLIC UNDO2                          OFFLINE
    _SYSSMU32$                     PUBLIC UNDO2                          OFFLINE
    _SYSSMU33$                     PUBLIC UNDO2                          OFFLINE
    34 rows selected.
    SQL> show parameter UNDO
    NAME TYPE VALUE
    undo_management string AUTO
    undo_retention integer 40000
    undo_suppress_errors boolean FALSE
    undo_tablespace string UNDO1
    SQL>
    Thanks

Maybe you are looking for

  • How to use tv as a mac book monitor

    I AM TRYING TO HOW TO WATCH MY MAC ON TV SCREEN. I HAVE ALREADY CONNECT MY TV WITH MAC , WITH  MINI DVI ADAPTER  AND CABLE AND I SAW THAT THE MAC RECOGNIZE THE TV BUT I CAN'T SEE THE ICONS THAT ARE ON MY MAC ON TV. I CAN SEE ONLY DESKTOP BACKGROUND.

  • Help. The audio of my footage goes really quiet when dragged into Premiere Pro timeline

    Hi there, Basically, I conducted an interview and the audio sounds loud when I play it through the top left editing window on Premiere, but when I drag it into the timeline it goes quiet. I go on audio gain and it says the peak amplitude is set to -1

  • Oracle.adf.view.rich.newWindowDetect.OPTIONS ADF_FACES-30163

    Hi, all! I have turned flag oracle.adf.view.rich.newWindowDetect.OPTIONS to ON in web.xml +<context-param>+ +<param-name>oracle.adf.view.rich.newWindowDetect.OPTIONS</param-name>+ +<param-value>on</param-value>+ +</context-param>+ but when I look int

  • How do I find program files on my pro?

    I need to access the program file for my steam account because it is not finding the servers.  Steam suggest that I set my own launch options, but I cannot find where the program file is for steam.

  • Resizing images saved in photoshop

    I'm working as a designer. Whenever I send images to my boss (usually jpgs saved in photoshop), she can't resize them. Is there a certain way I need to save them in order for them to be resized later?