Plans of multiple executions for the same SQL statement

Similar Questions

• NPL of multiple execution for the awr sql_id

  Hi Experts,
  
  NPL of multiple execution for the AWR sql_id,
  
  I followed questions
  
  1. What plan using opimizer tcurrently?
  2. make sure optimizer to choose good plans
  
  
  SQL > select * from table (dbms_xplan.display_awr ('fb0p0xv370vmb'));
  
  PLAN_TABLE_OUTPUT
  --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  SQL_ID fb0p0xv370vmb
  --------------------
  
  PLAN_TABLE_OUTPUT
  --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  Hash value of plan: 417907468
  
  ---------------------------------------------------------------------------------------------------------------
  | ID | Operation | Name | Lines | Bytes | TempSpc | Cost (% CPU). Time |
  ---------------------------------------------------------------------------------------------------------------
  | 0 | SELECT STATEMENT | 63353 (100) |
  | 1. UPDATE |
  | 2. SORT ORDER BY | 17133. 2978K | 3136K | 63353 (1) | 00:14:47 |
  | 3. HASH JOIN RIGHT SEMI | 17133. 2978K | 62933 (1) | 00:14:42 |
  | 4. COLLECTION ITERATOR PICKLER FETCH | |
  | 5. HASH JOIN RIGHT SEMI | 68530 | 11 M | 62897 (1) | 00:14:41 |
  
  PLAN_TABLE_OUTPUT
  --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  | 6. VIEW | VW_NSO_1 | 5000 | 35000 | 33087 (1) | 00:07:44 |
  | 7. COUNTY STOPKEY |
  | 8. VIEW | 127K | 868K | 33087 (1) | 00:07:44 |
  | 9. GROUP SORT BY STOPKEY | 127K | 2233K | 46 M | 33087 (1) | 00:07:44 |
  | 10. TABLE ACCESS FULL | ASYNCH_REQUEST | 1741K | 29 M | 29795 (1) | 00:06:58 |
  | 11. TABLE ACCESS FULL | ASYNCH_REQUEST | 1741K | 280 M | 29801 (1) | 00:06:58 |
  ---------------------------------------------------------------------------------------------------------------
  
  SQL_ID fb0p0xv370vmb
  --------------------
  SELECT ASYNCH_REQUEST_ID, REQUEST_STATUS, REQUEST_TYPE, REQUEST_DATA, PRIORITY, SUBMIT_BY, SUBMIT_DATE.
  
  Hash value of plan: 2912273206
  
  --------------------------------------------------------------------------------------------------------------------------
  | ID | Operation | Name | Lines | Bytes | TempSpc | Cost (% CPU). Time |
  
  PLAN_TABLE_OUTPUT
  --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  --------------------------------------------------------------------------------------------------------------------------
  | 0 | SELECT STATEMENT | 45078 (100) |
  | 1. UPDATE |
  | 2. SORT ORDER BY | 1323. 257K | 45078 (1) | 00:10:32 |
  | 3. TABLE ACCESS BY INDEX ROWID | ASYNCH_REQUEST | 1. 190. 3 (0) | 00:00:01 |
  | 4. NESTED LOOPS | 1323. 257K | 45077 (1) | 00:10:32 |
  | 5. THE CARTESIAN MERGE JOIN. 5000 | 45000 | 30069 (1) | 00:07:01 |
  | 6. UNIQUE FATE |
  | 7. COLLECTION ITERATOR PICKLER FETCH | |
  | 8. KIND OF BUFFER. 5000 | 35000 | 30034 (1) | 00:07:01 |
  | 9. VIEW | VW_NSO_1 | 5000 | 35000 | 30033 (1) | 00:07:01 |
  
  PLAN_TABLE_OUTPUT
  --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  | 10. UNIQUE FATE | 5000 | 35000 |
  | 11. COUNTY STOPKEY |
  | 12. VIEW | 81330 | 555K | 30033 (1) | 00:07:01 |
  | 13. GROUP SORT BY STOPKEY | 81330 | 1 429 K | 2384K | 30033 (1) | 00:07:01 |
  | 14. TABLE ACCESS FULL | ASYNCH_REQUEST | 86092 | 1513K | 29731 (1) | 00:06:57 |
  | 15. INDEX RANGE SCAN | ASYNCH_REQUEST_SUB_IDX | 1 | | | 1 (0) | 00:00:01 |
  --------------------------------------------------------------------------------------------------------------------------
  
  Hash value of plan: 3618200564
  
  --------------------------------------------------------------------------------------------------------------------------------
  | ID | Operation | Name | Lines | Bytes | TempSpc | Cost (% CPU). Time |
  
  PLAN_TABLE_OUTPUT
  --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  --------------------------------------------------------------------------------------------------------------------------------
  | 0 | SELECT STATEMENT | 59630 (100) |
  | 1. UPDATE |
  | 2. SORT ORDER BY | 4474 | 777K | 59630 (1) | 00:13:55 |
  | 3. HASH JOIN RIGHT SEMI | 4474 | 777K | 59629 (1) | 00:13:55 |
  | 4. VIEW | VW_NSO_1 | 5000 | 35000 | 30450 (1) | 00:07:07 |
  | 5. COUNTY STOPKEY |
  | 6. VIEW | 79526 | 543K | 30450 (1) | 00:07:07 |
  | 7. GROUP SORT BY STOPKEY | 79526 | 1397K | 7824K | 30450 (1) | 00:07:07 |
  | 8. TABLE ACCESS FULL | ASYNCH_REQUEST | 284K | 5003K | 29804 (1) | 00:06:58 |
  | 9. TABLE ACCESS BY INDEX ROWID | ASYNCH_REQUEST | 71156 | 11 M | 29141 (1) | 00:06:48 |
  
  PLAN_TABLE_OUTPUT
  --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  | 10. NESTED LOOPS | 71156 | 11 M | 29177 (1) | 00:06:49 |
  | 11. UNIQUE FATE |
  | 12. COLLECTION ITERATOR PICKLER FETCH | |
  | 13. INDEX RANGE SCAN | ASYNCH_REQUEST_EFFECTIVE_IDX | 327K | | 392 (1) | 00:00:06 |
  --------------------------------------------------------------------------------------------------------------------------------
  
  Thank you
  -Raj
  
  Published by: tt0008 on August 22, 2012 20:34

  Hello

  (1) you can see what plan has been used lately by running this query:

  select begin_interval_time, plan_hash_value
  from dba_hist_sqlstat st,
          dba_hist_snapshot sn
  where st.snap_id = sn.snap_id
  and sql_id = 'fb0p0xv370vmb'
  order by begin_interval_time desc;
  

  However, there is no guarantee that the next time you run this query, the latest plan will be chosen.
  Periodically, the plan is regenerated (for example when new statistics are collected, is the structure of a table referenced in)
  the query is changed etc.), and you can get 4 plans, or even a new function of many factors
  (statistics, bind variable values, the optimizer, NLS etc settings settings.)

  (2) this question is too large for the answer to fit into a thread, there are books written on the subject. The short answer is:
  If you know which of the 4 plans is right for you, then you can use a stored outline to lock in (it seems that you are not on 11g so SQL Profiler are not an option for you).
  Or you can try to find out why the optimizer generates different plans and address the underlying issue (the most common reason is to bind peeking - but to say
  course, we need to know more, starting with the text of your query).

  Best regards
  Nikolai

• BlackBerry Smartphones why do I get multiple contacts for the same person

  Recently, when I add a new contact the 9700 creates several entries. When I delete these multiple inputs, then it leaves about 10 spaces. If I let multiple entries for the same person, the next person that I add only then indicates once

  Why they appear more than once?

  How can I delete multiple entries so they do not reappear when I create the following entry?

  Help, please

  Try a simple reboot and see if the empty entries remain.

  With the BlackBerry device powered time, remove battery for a few seconds and then reinsert the battery to restart.

• AcriveSync has created multiple numbers for the same contact

  I recently had my transferred to my new pre phone book and set up my corp email.

  A lot of my contacts are mixed up (last names starting with O come in T, and these records have multiple records for the same phone number.  How can I fix?

  Hey thanks - I thought about it.  Turns out there have a lof of contacts in my constituency of corp I do not even know (or use).  I went in a deleted them and now al is good.  Thank you!

• OAM - map multiple URLS for the same web application

  Is OAM, posible to map multiple URLS to the same web application? When the user access to a different URL, it passes to the same application but set up with different strategy?

  In this case, this is the scenario that describes ROISSAC - as he says OAM cannot protect the same resource with two different strategies.

  Perhaps you could do this at the level of the network - if the two companies access the application through different proxy servers, you can use different identifiers to host for the two WebGates on proxy servers. This solution is easier to apply when there is a physical separation of users (for example, the internal/external users) rather than different users coming through the same path.

  Kind regards
  Colin

• Multiple values for the same column in the columns of diffétent in the same row?

  Hi all
  I wonder how you can display different values for the same column in different columns on the same line. For example, using a CASE statement, I can:
  
  CASE WHEN CODE IN ('1 ', ' 3') THEN COUNT (ID) AS 'Y '.
  CASE WHEN CODE NOT IN ('1 ', am') THEN COUNT (ID) AS "N".
  
  Yes, that will produce two columns but will produce null values to empty and also two separate registers.
  
  
  Any ideas?
  
  Thank you

  Are you sure that this code works for you?

  Select ID
           ,CASE WHEN MODE_CODE IN ('1', '3') THEN COUNT( No) END as "Fulltime"
           ,CASE WHEN MODE_CODE NOT IN ('1', '3') THEN COUNT( No ) END  as "Other"
  From table
  group by ID
  

  I guess the code above fails because MODE_CODE is not in your group by?

  My suggestion would be to put the CASE in the COUNT:

  Select ID
           ,COUNT(CASE WHEN MODE_CODE IN ('1', '3') THEN No END) as "Fulltime"
           ,COUNT(CASE WHEN MODE_CODE NOT IN ('1', '3') THEN No END)  as "Other"
  From table
  group by ID
  

  CASE expressions return no. when the respective conditions are true and NULL otherwise.
  COUNTY will have non-null values.

• Multiple aliases for the same subject?

  Hello world
  
  In OPM, is it possible to use several aliases for the same subject?
  
  I try to maintain the isomorphism of my wording of the policy.
  
  For example, my wording of the policy uses the words "person", hereinafter the "applicant" and "beneficiary" refers to the same person.
  
  I like to use "person", hereinafter the "applicant" and "the beneficiary" in my rules, but have the OPM recognize that whenever I write 'applicant' or 'the beneficiary', I'm actually referring to "the person".
  
  Is there a way to do this in OPM?

  An easy way to do this is to map through the underlying attributes in the rule itself.

  So you you would end up with rules such as (assuming that the entity is called 'the person'):

  the applicant is eligible if
    the applicant is aged over 18 years
       the person's age > 18
  

  It has not built based on synonym, as in most cases, the rules are more transparent without them.

  Published by: Davin Fifield, 12/07/2010 15:48

• Different plan for the same sql id

  Hello

  Our team of application reported recently that a query ran longer than usual.

  Search in tables AWR (dba_hist_active_sess_history & dba_hist_sqlstat), it was a change of plan_hash_value. The next day, without intervention from anyone (collect statistics did not run, no change in the tables involved) the sql used the former plan and completed quickly.

  The differences in the plan was the index that was used to access the table.

  What could be the reason for the optimizer to choose a bad plan and return once more the good thing. ?

  The code sql that has been run twice & three in one day and then it years off cursor cache.

  Regime shifts are normal.

  Flipfloppping plans are quite normal.

  I'd be willing to bet that a large percentage of your SQL shows variations in implementation plans and you never notice.

  And there are several reasons why you might get a different plan.

  1 bind variable peeking - it is normal SQL for the age from the cache and then get analyzed once again with different bind variable leading to different cardinality estimates and plans

  2. the statistics change - it is normal that the statistics of change over time causing different cardinality estimates and the various plans, especially if the histograms are involved and changing buckets

  3. data change - especially if you use dynamic sampling it is normal to get a different sample of data leading to different cardinality estimates and plans

  4 SYSDATE - it's normal for queries with SYSDATE in them to recognize that time is never on leave and which, in conjunction with statistics or data, can lead to different cardinality estimates and plans

  5. integrated optimization features - it's normal for features like ACS to kick in and notice that they got forecasts wrong in an execution plan and force the recalculation to a different plan with an adjusted cardinality.

  Using DBMS_XPLAN. DISPLAY_CURSOR or DISPLAY_AWR you can get different plans (provided that they are in memory or AWR). The notes section should indicate cardinality comments have been a factor. You can also get links peeked with format mask of "+ PEEKED_BINDS".

• My iTunes library creates multiple listings for the same artist (spelled identically)

  I use iTunes for years and thought I knew how to do everything that I need in there and solve any problem that I've run across.  But in the last 10-14 days, he began to behave strangely, and I'm completely stumped.  I don't know if it's a problem with a new version, if maybe my library file has become corrupted, or if I'm missing something obvious, but as someone who spends hours meticulously entering all the right data to my iTunes library, that drives me crazy.

  For some reason any recently, iTunes has begun to create several artist registrations for a handful of artists in my library (in only general I have a LOT of material for.)  Thus, for example, when I scroll my iTunes under 'Artists' library I noticed that Bruce Springsteen four times:

  Bruce Springsteen (12 albums, songs of 234)

  Bruce Springsteen (16 albums, 201 songs)

  Bruce Springsteen (2 albums, 22 songs)

  Bruce Springsteen (2 albums, 41 song)

  It would be only incredibly annoying, but it actually divides some albums through multiple "artists" so that securities 1,4,6,7 will be an artist under a listing and the rest on another.

  I created a smart playlist with the criteria ' artist contains 'bruce' "selected all titles of springsteen and changed the artist on EACH of them to ensure that there is no variation in spelling or spacing.  It does not exist. (Springsteen is the worst offender, having four entries of artist, but there's another 3-4 artists who do the same thing)

  It drives me crazy because there seems to be no rhyme or reason to it, and it is debilitating to my efforts to clean and organize my library.  I try to avoid manually recharge all my music in iTunes I have dozens of playlists I've spent hours creating and would not have to start over on those.

  Any suggestions?

  You have already tried the first method, so try the other here:

  Fix split Album or artist listings in iTunes

• How to place multiple orders for the same address on the photos?

  Hello!

  I created several books on Photos and I would like to get printed and shipped to my address.

  Unfortunately, it seems that it is only possible to place an order at the announcement of time (a book both in my case) so pay the shipping costs for each order/book.

  Am I missing something here?

  Thank you in advance for your help!

  You can combine only multiples of the same item in an order.  If you have created several books, you will need to pay the shipping costs for each of them separately.

• Node affinity d ' execution for the PL/SQL Packages?

  Colleagues dbas,

  We have a problem of random performance on one of our database RAC 11 GR 1 material due to waiting for IO.

  In fact, we've tracked down the problem and noticed that the sql packages are executed on the bad weather causing node 1 and were executed on node 2 with good results.

  Now the question is how we apply these pl/sql packages, that are called via batch external, must be running on node 2?

  Affinity of the proceeding came to my mind, but I have no idea whatsoever how to apply for parcels and incorporated the pl/sql code.

  Can you please give me an advice?

  Thanks for your comments!

  Best regards

  Sebastian

  You must define a service to run on node 2 with failover to node 1 and give a your batch jobs a string of connection that connects to this service:

  Add srvctl service d a s batchserv - r node2 - a node1

  and in your tnsnames.ora,.

  batchserv =

  (description =

  (address_list =

  (failover = on) (load_balance = on)

  (address = (node1...) (address = (node2...))

  )

  (connect_data = (service_name = batchserv))

  )

  I recorded several demos of this thing of the kind,

  Database Oracle 12 c and g 11 tutorials free for Administration and developers SkillBuilders

  --

  John Watson

  Oracle Certified Master s/n

• DAQmx: Outputs digital multiple tasks for the same card using the on-board clock

  I have a PCI - 6259 DAQ I'm programming with DAQmx. I want to create two tasks of digital output on the same card. A single task using on-board clock, continuous samples. The other task uses an onboard clock, finished samples and an external trigger with trigger Start.Retriggerable property = TRUE. Is this possible with my hardware and DAQmx? There will be a conflict if two tasks use the same clock source?

  This is not possible, you can have only one task timed materially by device.

  Best regards

• Multiple terms for the same key.

  Is there a way you can have several condition (same type of the report).
  
  I have the button I want to hide at different stages.
  
  For example I would like to have a button (Submit) that will be visible when
  
  Status = project
  State value of Type of element in the Expression 1 = Expression 2
  P9_X_STATUS
  Project
  
  and when also when userRole = author
  
  State value of Type of element in the Expression 1 = Expression 2
  P9_X_USER_ROLE
  Author
  
  Robert

  Hi Robert,.

  You can use a SQL query is to treat several conditions:

  SELECT 1
  FROM DUAL
  WHERE :P9_X_STATUS = 'Draft'
  AND :P9_X_USER_ROLE = 'Author'
  

  This returns a record only if two conditions are met - if either a fake, then no records are returned, and the condition is not met, so that the button is not displayed

  Andy

• When I add a bookmark, I get hundreds of multiple entries for the same bookmark in the recently added bookmarks folder and the bookmark tab. Difficulty?

  Adding a results of bookmark in 100's of entries for this bookmark on the Bookmarks tab and in the recently added bookmarks folder. The only way to remove them is to click and delete each separate entry? Any idea what is happening or how to fix it? Thank you.

  Start Firefox in Safe Mode {web link}
  While you are in safe mode;

  Type of topic: preferences #advanced< enter > in the address bar.

  Under Advanced, select General.
  Find and stop using hardware acceleration.

  Search web sites secure. Are there problems?

• Dblink Oracle to sql server, multiple database on the same server sql under a dblink

  Hi, we managed to set up an Oracle dblink to sql server and retrieve data.

  The user of sql server have been using via dblink has access to multiple databases on the same sql server

  But the question is how in oracle (if possible) prepend you the SQL access to this?

  For example:

  Sqlserver_prod has the user sqlserver_user which seems to be set up as default database sqlserver_db1

  But we have select access to sqlserver_db2

  all work well as sqlserver_user

  Select * from table_fromdb1

  Select * from dbo.table_fromdb1

  Select * from sqlserver_db1.dbo.table_fromdb1

  as does

  Select * from sqlserver_db2.dbo.table_fromdb2

  more in Oracle

  Oracle_db a dblink sqlserver_prod. World connection sqlserver_user

  everything works fine

  Select * from 'table_fromdb1"@sqlserver_prod '.

  Select * from 'dbo '. "table_fromdb1"@sqlserver_prod

  But how to (if possible) access from oracle

  sqlserver_db2.dbo.table_fromdb2

  without having to create a new sqlserver_db2_user referenced in a new dblink

  If oracle for oracle would be

  Select * from remote_oracle_schema.table@remote_oracle_db

  Hello

  You cannot select a table in a different SQL * database server from that to which the gateway instance connects.
  As stated in the documentation-

  Oracle® database gateway

  Installation and Configuration Guide

  11g Release 2 (11.2) for AIX 5 L Based Systems (64-bit), HP - UX

  Itanium, Solaris (SPARC 64-Bit), Linux x 86 operating system,

  and Linux x 86-64

  In the section.

  The example SQL Server multiple databases: Configuration of the modem router

  A separate instance of the gateway that is required for each SQL Server database. Each

  instance needs its own Gateway system ID (SID).

  ==========

  You will need to create a new instance of the gateway for the SQL * Server DB2 as well as a link separate db.

  Kind regards

  Mike