Why oracle.exe takes 8 Virtual G; bytes, memroy target may not go beyond 5 G. How to minimize the virtual bytes perfmon

Similar Questions

• How to understand the implementation of the plan in oracle I mean if I see two implementation plans for a single sql_id plans 2 How to determine the best execution plan? Links and answers are much appreciated. Thank you

  How to understand the implementation of the plan in oracle I mean if I see two implementation plans for a single sql_id plans 2 How to determine the best execution plan? Links and answers are much appreciated. Thank you

  How to understand the implementation of the plan in oracle I mean if I see two implementation plans for a single sql_id plans 2 How to determine the best execution plan? Links and answers are much appreciated. Thank you

  After two execution plans that have the same sql_id, so we can see what you're talking about.

  See "Oracle Explain Explain Plan optimizer" by Maria Colgan of the Oracle optimizer group

  http://www.Oracle.com/technetwork/database/bi-Datawarehousing/TWP-explain-the-explain-plan-052011-393674.PDF

  Examine the various aspects of a selectivity to parallel execution plan

  performance and understand what information you should be brilliant

  the plan can be overwhelming even for the most experienced DBA. This document

  offers a detailed explanation on each of the aspects of the execution plan and a

  Overview of what caused the CBO to make the decision, he did.

• Why Autocheck.exe takes 3 + minutes at startup

  Salvation;

  I've recently set up a new build (the end plug) and after installing the win 7 updates that were available, I started having a recurring problem starting sloooooooooow.  I seem to have two options now, and none of them are perfect, that's why I post here.

  1. I can restart and wait 3-4 minutes, and when windows starts, everything seems fine
  2. I can remove autocheck autochk * of the BootExecute registry entry, then it starts fine, but then windows is locked for about 3 minutes after logon. Nothing does.  After that, it works fine.

  In the investigation of this situation, I blame first SMSSinit, but I discovered later that this routine is rather a chaperone for the other processes that are running before Winlogon support functions of escort.  With the help of MS SDK, I found that it is Autochck.exe who is the culprit, taking 200-220 seconds on each boot.  I have done the following (and more) for this problem:

  1. check all readers to forest Sales - none found
  2 DAB check on all connected not discs - no error found
  3 disabled Checkdisk in "task manager".
  4 disabled Checkdisk on the register (when I do that, I get the scenario 2 above)
  5 restart the computer in every way that I can imagine
  6 tried a repair of windows with my stick UEFI (no love - don't say it found no problem)
  7. I tried to go back to a previous restore point (I only had one and it has not resolved)
  8 deleted all the updates of windows that I did (and all Win updates are set to manual only)
  9 autochk.exe caught another install and put it in the folder Windows/system32 (after correct ownership of both) with a new name autochk_safe.exe and made the registry BootExecute entry to this application checkdisk (thought that was a smart test I found on another forum)

  10A ran sfc/scannow - no problems found
  11 set the registry BootExecute entry "" autocheck autochk / k: C / k: M / k: S k:R k:G * '. "  I tried this (which should skip autocheck at startup for C, G, M, R and S readers), but it changed nothing.  And it's a deactivation of all disks in the system.

  If anyone knows what steps I can take here to stop this autocheck scope and not suffer the consequences, it would of course be appreciated.  And Yes, I know that I can do a thorough cleaning and reinstall, but it isn't necessary that something is the cause and needs to be identifiable and solvable IMO.

  See you soon ~
  C

  Update Specs:
  Windows 7 Pro 64-bit SP1 (all the critical updates installed until I removed them to try to solve this problem)
  ASUS X 99 Sabertooth Mobo
  CPU i7 - 5820 k
  M.2 SM951 Samsung 256 GB (drive)
  Corsair ForceGT 128 GB SSD (secondary drive and drive work when running on hand)
  4 TB WD Red
  4x4TB red WD in a RAID5 configuration (fully initialized and works very well)
  Crucial RAM - 2 x 8 GB - 2133 Mhz
  FireGL V8700 GPU (the elder in this system of 2010) Pro
  EVGA Supernova 1000 G2 PSU

  At this stage of installation the installation program has not yet copied all files or written a basic configuration register.  You can see the drive because the drivers are loaded into memory to allow the Setup program see the disks at this time there hadn't even told it the installer where to install Windows, how he knew where to put the files?

  This can be useful:

  SysInternals LoadOrder v1.0

  (Works on Windows 7)

  John

• Why this system take me round in circles and leaves me not to ask a question?

  I have just updated to Firefox 26.0 27.0 and he lost most of my favorites.
  How can I get back them?

  I was just shopping for the afternoon.
  Missing bookmarks are back!
  I wonder where they were? Still, both pis, problem sorted.
  Thanks for your response

• Why dose Microsoft fix don't library so it dose do not require equipment ir

  Why dose Microsoft fix don't library so it dose do not require equipment ir

  Hello

  On the question of the material on the IR, you can publish your comments. Please visit the link below:

  Microsoft Connect

  http://connect.Microsoft.com/

• Why apps like slate say available on any device but are not compatible on Android?

  Why apps like slate say available on any device but are not compatible on Android? It is today the largest operating system.

  Hi Michael % 20alaska,

  We have a slate Adobe user forum that you can post to. Please go here and post your situation of assistance:

  Adobe_Slate user forum

• How to disable the trace files in the oracle 11g version

  Senario: trace file grow
  How to disable the trace files in the oracle 11g version
  pls guide with best practices

  NATHALIE wrote:
  Senario: trace file grow
  How to disable the trace files in the oracle 11g version
  pls guide with best practices

  11 g, there is an extended tracing which happens for reasons best known only to Oracle. But if you want to disable, Coskan had published a small ticket mentioning a parameter not documented (which means you should think twice before using it) to disable it - disablehealth_check *. Here you can read the full message,
  http://Coskan.WordPress.com/2009/06/03/too-many-trace_file-on-11g/

  Aman...

• Why the GUI vmware.exe resizes when virtual machines?

  Why the GUI vmware.exe resizes when virtual machines?
  For me, it's very annoying - how can I disable this behavior?

  Hi Continuum,

  Have you checked the settings of automatic adjustment under "Preference-> Display? I guess that your "Autofit window" is checked.

• Why oracle does not take into account different types of blocks?

  Hello all;

  I need some information abt 'why oracle does NOT support different block size to create objects?

  I know that we cannot create all objects with different block size.  but do NOT know the reason

  
  

  -----------------------------

  Clear explanation, good for me.

  
  

  Thanks and greetings

  GTS (ASIA PACIFIC)
  

  I need some information abt ' why oracle does NOT support different block size to create objects?

  I know that we cannot create all objects with different block size.  but do NOT know the reason

  If I understand you correctly, this is not true: you can create storage with different blocks and create objects in their breast.

  See CREATE TABLESPACE

• Oracle.exe memory up to 1.9 GB in a week

  Salvation;
  
  We have a database running in Oracle9i Release 9.2.0.7.0, on a windows 2003 32-bit server.
  
  Usually oracle.exe memory does not have 1.2 GB
  Problem is oracle.exe incrises in memory of close size of 1.8 GB, 1.9 GB... the limit of the OS for a process.
  (When it hits this memort size, it is not possible even to make a login)
  
  In the latest version, someone changed init.ora and added:
  session_cached_cursors = 50
  
  This could be the reason? How should we study this?
  
  Thank you.

  PauloSMO wrote:
  (1) select name, sum(value/1024/1024) 'value - Mb '.
  v $ statname n,.
  v$ session s,.
  v$ sesstat t
  where s.sid = t.sid
  and n.statistic # t.statistic = #.
  and s.type = "USER".
  and s.username is not NULL
  and n.name ("pga session in memory", "pga memory max session", "uga session in memory", "uga memory max session")
  Group by name

  PGA session in memory 381 MB
  PGA memory max 507 MB session
  UGA session in 305 MB memory
  UGA memory max 358 MB session

  Nothing close to 1.8 GB!

  (2) is it possible to find the evicence of the use of stateless applications, without closing the session?

  I do not think that the SQL statement above shows the real cause of the problem, but it could give an idea. Note that when a session disconnects, statistics for this session disappear from V$ SESSTAT, so the result of your SUM is updating of already disconnected sessions which were previously connected. If you had a situation where the total PGA + SGA total exceeded about 1.8 GB. Watch more closely settings initialization and ask yourself why the settings are set with these values.

  db_cache_size=335544320 = 320MB
  java_pool_size=104857600 = 100MB
  large_pool_size=134217728 = 128MB
  shared_pool_size=134217728 = 128MB
  
  workarea_size_policy=AUTO
  pga_aggregate_target=104857600 = 100MB
  sort_area_retained_size=104857600 = 100MB
  sort_area_size=104857600 = 100MB
  
  HASH_AREA_SIZE defaults to twice the SORT_AREA_SIZE for sessions where WORKAREA_SIZE_POLICY=MANUAL.
  
  optimizer_index_caching=99
  optimizer_index_cost_adj=10
  
  processes=300
  
  _log_simultaneous_copies=4
  _spin_count=10000
  

  Added values for db_cache_size java_pool_size + large_pool_size + shared_pool_size total 676MB (it's probably a OK, if not too low of a value in my opinion, but lifting in this case is not a good idea), but that's not all memory assigned to the SGA. You can run the following in SQL * more:

  SHOW SGA
  

  Your default WORKAREA_SIZE_POLICY is set to AUTO, so unless sessions are explicitly modify this setting at the level of the session or sessions are connected in a shared server configuration, the potential 300 sessions are fighting for their share of the 100MB that is defined for the PGA_AGGREGATE_TARGET - Windows each session will use a minimum of 1 MB of memory just to connect , so if you had 300 sessions connected, which is a minimum of 300 MB consumed before any meeting does all the work. If the sessions were adjust the WORKAREA_SIZE_POLICY or connect with a shared server configuration, the 100 MB / sorting box (it can be more space 1 sort used by only one session at a time) and 200 MB / hash box setting takes effect-, so there may be multiple 100MB memory credits per session , and if I remember correctly, these allocations of memory are not released until the session disconnects.
  A little more information can be found here:
  http://www.jlcomp.demon.co.UK/snark.html

  Then, examine why optimizer_index_caching = 99 and optimizer_index_cost_adj = 10 are defined. The following article can help:
  http://hoopercharles.WordPress.com/2009/12/06/faulty-quotes-1-OPTIMIZER_INDEX_COST_ADJ/

  Finally, examine why you have two parameters hidden defined initialization: logsimultaneous_copies = 4 and spincount = 10000

  Charles Hooper
  Co-author of "Expert Oracle practices: Oracle Database Administration of the Oak Table.
  http://hoopercharles.WordPress.com/
  IT Manager/Oracle DBA
  K & M-making Machine, Inc.

• Explorer.exe takes more than 90% of the CPU when I opened start > menu programs

  IM using Vista Ultimate with Aero theme. I am a single user of this computer. When I open Start-> programs menu, an explorer.exe process takes 90% and more CPU resources and remains to be done (even if I don't move the mouse or press any key on the keyboard) until I close the menu programs. What continues?

  Hello Serresjio I can almost guarantee you that the troubleshooting steps will flip over as soon as you start back. my suggestion is if possible, remove the hard drive and the scanner on another computer with Malware Bytes Anti Malware, Avast! Antivirus and Spybot Search and Destroy, AVG etc. (all updates of course). usually, I use this method when everything has failed. Don't be too surprised boyfriend so often that I have a USB-(e sata, 1394, 2.5"/5.25 IDE) converter for the toughest jobs.

  Everything do you to not lose hope and format as most of the people do. I have corrected problems of windows than most irreparable thought technicians. It's just a matter of removing the cause and repair the damaged windows files. the problem can be solved the difficulty is just to use the proper tools/software. Lately I use Avast and MBAM with 100% recovery. It IS not JUST a single unrecoverable machine!

• Why scvhost.exe uses 75% of the physical memory of my laptop?

  I noticed recently that scvhost.exe a host process for windows, use more than 75% of the physical memory of my computers, usually at about 106, 000 k, it's not normal that my system and it slows down my system, I regularly keep my PC, because I use it regularly at the College and put it on standby saves time as the stopping and starting of the system again Is there a version of the patch that will prevent scvhost.exe to use lots of physical memory.

  Thank you very much.

  Something more about this problem? Still facing this problem where svchost.exe takes too much memory. :(

  Edit: I deactivated "Generator Windows Audio endpoint" in services.msc and the problem disappeared. This servise is to control audio devices but I have a Realtek sound card and I installed the latest drivers for it and the sound works & speakers.
  Hope that helps anyone who has similar problems.
  Edit2: after as I restared, svchost.exe still using a lot of memory, but now it is linked with cscservise. Do not know what is happening right now. Anyone?
  Edit3: I did some research on the Internet and came up with this SuperFetch service. After you disable it in the Manager of tasks, svchost.exe became normal and he did not use this amount of memory. Also, I disable it in services.msc and I hope it will work as good as it was.
  Just do not know why this problem occurred, where. I didn't install anything recently except Battlefield 3.

• Why Oracle ManagedDataAccess (4.121.1.0) throws a System.FormatException "first chance" exception to the debugger when you use OracleCommand.BindByName (on some machines, but not others)?

  Why Oracle ManagedDataAccess (4.121.1.0) throws a System.FormatException "first chance" exception to the debugger when you use OracleCommand.BindByName (on some machines, but not others)?

  This occurs when calling a ref cursor-returning packaged function stored with OracleCommand.BindByName set to true.

  When BindByName undefined, it works as expected.

  I've only been able to recreate this problem on a machine in my control. This issue has also reported to me by a customer.

  I tried to change the locale of the system of EN - UK EN-US, but I was not aware of any change in behavior.

  Similarly, I've tried explicitly overriding the default locale of the system with the OracleGlobalization parameters, and it also had no effect.

  Please see the example below.

  ---

  Environment (default):

  Windows 7 Professional 64-bit SP1

  Visual Studio 12.0.21005.1 REL / / CAN BE SIGNIFICANT

  .NET framework 4.5.51209

  Local system: EN - UK (but also does not, EN-US).

  Oracle.ManagedDataAccess 4.121.1.0

  "< supportedRuntime version ="v4.0"sku =". NETFramework,Version=v4.5.1"/ >

  Environment (base):

  Windows 7 Professional 64-bit SP1

  Updated Visual Studio 12.0.31101.00 //SEE above 4

  .NET framework 4.5.51209

  Local system: EN - UK

  Oracle.ManagedDataAccess 4.121.1.0

  "< supportedRuntime version ="v4.0"sku =". NETFramework,Version=v4.5.1"/ >

  Call stack:

  	mscorlib.dll. Unknown System.Number.StringToNumber(string str, System.Globalization.NumberStyles options, ref System.Number.NumberBuffer number, System.Globalization.NumberFormatInfo info, bool parseDecimal)
  	mscorlib.dll. Unknown System.Number.ParseInt32(string s, System.Globalization.NumberStyles style, System.Globalization.NumberFormatInfo info)
  	Oracle.ManagedDataAccess.dll! Oracle.ManagedDataAccess.Types.OracleRefCursor.OracleRefCursor (connection Oracle.ManagedDataAccess.Client.OracleConnection, OracleInternal.ServiceObjects.OracleRefCursorImpl refCursorImpl, Oracle.ManagedDataAccess.Types.OracleIntervalDS sessionTimeZone, string commandText, string paramPosOrName, long initialLongFS, long initialLobFS, long [] scnFromExecution) unknown
  	Oracle.ManagedDataAccess.dll! OracleInternal.ServiceObjects.OracleParameterImpl.ExtractRefCursorFromAccessor (Oracle.ManagedDataAccess.Client.OracleConnection conn, accessor OracleInternal.TTC.Accessors.Accessor, long fetchSize, Oracle.ManagedDataAccess.Client.PrmEnumType enumType, Oracle.ManagedDataAccess.Types.OracleIntervalDS sessionTimeZone, string commandText, string paramPosOrName, long longFetchSize, long lobFetchSize, long [] scnFromExecution, int currentRow) unknown
  	Oracle.ManagedDataAccess.dll! OracleInternal.ServiceObjects.OracleParameterImpl.GetRefCursorFromBytes unknown (Oracle.ManagedDataAccess.Client.OracleConnection conn, accessor OracleInternal.TTC.Accessors.Accessor, long fetchSize, Oracle.ManagedDataAccess.Client.PrmEnumType enumType, Oracle.ManagedDataAccess.Types.OracleIntervalDS sessionTimeZone, string commandText, string paramPosOrName, long longFetchSize, long lobFetchSize, long [] scnFromExecution)
  	Oracle.ManagedDataAccess.dll! Oracle.ManagedDataAccess.Client.OracleParameter.PostBind_RefCursor (connection Oracle.ManagedDataAccess.Client.OracleConnection, OracleInternal.TTC.Accessors.Accessor bindAccessor, long fetchSize, Oracle.ManagedDataAccess.Types.OracleIntervalDS sessionTimeZone, string commandText, string paramPosOrName, long longFetchSize, long lobFetchSize, long [] scnFromExecution) unknown
  	Oracle.ManagedDataAccess.dll! OracleInternal.ServiceObjects.OracleCommandImpl.ExtractAccessorValuesIntoParam (Oracle.ManagedDataAccess.Client.OracleParameterCollection paramColl, Oracle.ManagedDataAccess.Client.OracleConnection connection, string commandText, long longFetchSize, long lobFetchSize, long [] scnFromExecution) unknown
  	Oracle.ManagedDataAccess.dll! Unknown Oracle.ManagedDataAccess.Client.OracleCommand.ExecuteReader(bool requery, bool fillRequest, System.Data.CommandBehavior behavior)
  	Oracle.ManagedDataAccess.dll! Oracle.ManagedDataAccess.Client.OracleCommand.ExecuteReader (unknown)

  > ManagedSimpleRefCursor1.exe! NoddyConsole.Program.Command_ExecuteReader_BindByName_Fails (Line 63 c#)

  Example:

  ----

  using System;

  using System.Collections.Generic;

  using System.Linq;

  using System.Text;

  using Oracle.ManagedDataAccess.Client;

  using Oracle.ManagedDataAccess.Types;

  using System.Data;

  namespace NoddyConsole

  {

  public class Program

  {

  public static int Main (string [] args)

  {

  Error int = 0;

  Errors += Command_ExecuteReader_BindByName_Fails();

  Errors += UseReturnValue_Command_ExecuteNonQuery_BindByName_Fails();

  return errors;

  }

  public static int Command_ExecuteReader_BindByName_Fails()

  {

  Error int = 0;

  Try

  {

  var conn = new OracleConnection ("UID = USER; Password = PASS; ("Data Source = / / your-host: 1521 / yours-sid");

  var conn = new OracleConnection ("user Id = UNIT620; Password = UNIT620; ("Data Source = / / ln1apidb01:1521 / fm");

  Conn. Open();

  IRRELEVANT

  OracleGlobalization og is conn. GetSessionInfo();

  og. DateFormat = "HH24:MI:SS MON-DD-YYYY";

  Conn. SetSessionInfo (og);

  var cmd is conn. CreateCommand();

  cmd.CommandType = CommandType.Text;

  BindByName. cmd = true;

  cmd.CommandText = "START: rc: = SHAUN.recreate_issue_1 (in_code = >: in_code); END; « ;

  var prm1 is cmd. CreateParameter();

  PRM1. ParameterName = "rc";

  PRM1. Direction = ParameterDirection.ReturnValue;

  PRM1. OracleDbType = OracleDbType.RefCursor;

  cmd. Parameters.Add (prm1);

  / * SHAUN.recreate_issue_1 * /.

  var prm2 is cmd. CreateParameter();

  PRM2. ParameterName = "in_code";

  PRM2. Direction = ParameterDirection.Input;

  PRM2. OracleDbType = OracleDbType.Varchar2;

  PRM2. Value = "A";

  cmd. Parameters.Add (prm2);

  /* */

  var reader is cmd. ExecuteReader(); FAILS to System.Number.StringToNumber (String str, NumberStyles options, NumberBuffer & number, NumberFormatInfo info, Boolean parseDecimal)

  /* *-/

  OracleRefCursor orc = null;

  cmd ExecuteNonQuery()); must fail here

  ORC = cmd. Parameters ["rc"]. Value as OracleRefCursor;

  If (orc is nothing)

  throws InvalidOperationException new ("Invalid return the ref cursor type");

  var = orc player. GetDataReader();

  /* */

  While (reader. Read

  {

  var values = new object [reader. FieldCount];

  drive. GetOracleValues (values);

  for (int i = 0; i < values.) Length; (++ I)

  {

  Console.WriteLine(values[i]);

  }

  }

  drive. Dispose();

  cmd. Dispose();

  Conn. Dispose();

  }

  catch (Exception ex)

  {

  Console.Error.WriteLine (ex) Message);

  Errors ++;

  }

  Finally

  {

  Return (true);

  }

  return errors;

  }

  public static int UseReturnValue_Command_ExecuteNonQuery_BindByName_Fails()

  {

  Error int = 0;

  Try

  {

  var conn = new OracleConnection ("UID = USER; Password = PASS; ("Data Source = / / your-host: 1521 / yours-sid");

  Conn. Open();

  IRRELEVANT

  OracleGlobalization og is conn. GetSessionInfo();

  og. DateFormat = "HH24:MI:SS MON-DD-YYYY";

  Conn. SetSessionInfo (og);

  var cmd is conn. CreateCommand();

  BindByName. cmd = true; // ####

  cmd.CommandText = "START: rc: = SHAUN.recreate_issue_1 (in_code = >: in_code); END; « ;

  cmd.CommandText = "START: rc: = SHAUN.recreate_issue_2; END; « ; No params, works

  var prm1 is cmd. CreateParameter();

  PRM1. ParameterName = "rc";

  PRM1. Direction = ParameterDirection.ReturnValue;

  PRM1. OracleDbType = OracleDbType.RefCursor;

  cmd. Parameters.Add (prm1);

  / * SHAUN.recreate_issue_1 * /.

  var prm2 is cmd. CreateParameter();

  PRM2. ParameterName = "in_code";

  PRM2. Direction = ParameterDirection.Input;

  PRM2. OracleDbType = OracleDbType.Varchar2;

  PRM2. Value = "A";

  cmd. Parameters.Add (prm2);

  /* */

  var reader is cmd. ExecuteReader();

  OracleRefCursor orc = null;

  HERE does NOT cmd. BindByName = true; System.FormatException first-chance exception in at System.Number.StringToNumber (String str, NumberStyles options, NumberBuffer & number, NumberFormatInfo info, Boolean parseDecimal)

  cmd ExecuteNonQuery());

  ORC = cmd. Parameters ["rc"]. Value as OracleRefCursor;

  If (orc is nothing)

  throws InvalidOperationException new ("Invalid return the ref cursor type");

  var = orc player. GetDataReader();

  While (reader. Read

  {

  var values = new object [reader. FieldCount];

  drive. GetOracleValues (values);

  for (int i = 0; i < values.) Length; (++ I)

  {

  Console.WriteLine(values[i]);

  }

  }

  drive. Dispose();

  ORC. Dispose();

  cmd. Dispose();

  Conn. Dispose();

  }

  catch (Exception ex)

  {

  Console.Error.WriteLine (ex) Message);

  Errors ++;

  }

  Finally

  {

  Return (true);

  }

  return errors;

  }

  }

  }

  ---

  create or replace package THAT SHAUN is

  type ref_cur_type is ref cursor;

  function recreate_issue_1)

  in_code varchar2: = null

  ) return ref_cur_type;

  end SHAUN;

  /

  create or replace package body what SHAUN

  -TTS000233

  function recreate_issue_1)

  in_code varchar2: = null

  ) return ref_cur_type is ref_cur ref_cur_type;

  Start

  Open the ref_cur for

  Select UO.object_name from user_objects UO;

  Return ref_cur;

  end recreate_issue_1;

  end SHAUN;

  /

  ---

  Thanks a lot to qualify again for me Alex.  I'll try upgrading to the latest version of the provider.

  The difference in behavior between the two hosts was found to be reduced to differences in configuration VS.

  Checking 'Just my code' prevents the first chance thrown in the debugger exception. As in fact, the exception is handled in the provider, it is enough for me.

• Why can they take default permanent tablespace offline?

  We cannot take the tablespace temporary default offline, as well as of the SYSTEM, UNDOTBS1.
  so, I wonder why we take offline the permanent tablespace default as USERS?
  What happens if there is only a single tablespace 'USERS' in my database?

  PhoenixBai wrote:
  >
  Why do you have to be obsessed to take tablespaces offline?
  >
  not me obsessed to this topic. OCP tends to make a fuss out of it, so I have no choice, but to play the game.

  back to the point, the original question is:

  Your Oracle 10g database has 6 tablespaces in which:
  -TEMP is the default temporary tablespace
  - UNDOTBS1 is the default undo tablespace
  - USERS is the default permanent tablespace
  In this database, which three tablespaces can  be made offline? (Choose three.)
  
  1. USERS
  2. TEMP
  3. UNDOTBS1
  4. SYSTEM
  5. SYSAUX
  6. PROD
  

  responses, are without a doubt, 1,5,6. and I have no problem seen the SYSTEM may not be in offline mode.
  but, do not understand why TEMP and UNDOTBS1 isn't in offline mode while USERS can?
  for me, TEMP & UNDOTBS1 are to serve the manipulation on the data that is contained in the USERS (or PROD) tablespaces in this case.
  Thus, when data are not available, what is the point to keep TEMP & UNDOTBS1?

  Another possibility is: we are allowed to disconnect USERS tablespace, because currently we have an another PROD tablespace to store data?

  want you or anyone can help explain a little.
  I just want one another reasonable explanation that oracle doc said.

  Phonexbai,

  Please see,

  SQL> l
    1  select * from database_properties
    2* where property_name like '%DEFAULT%'
  SQL> /
  
  PROPERTY_NAME                                      PROPERTY_VALUE                                     DESCRIPTION
  -------------------------------------------------- -------------------------------------------------- ----------------------------------------------
  DEFAULT_TEMP_TABLESPACE                            TEMP                                               Name of default temporary tablespace
  DEFAULT_PERMANENT_TABLESPACE                       USERS                                              Name of default permanent tablespace
  DEFAULT_EDITION                                    ORA$BASE                                           Name of the database default edition
  DEFAULT_TBS_TYPE                                   SMALLFILE                                          Default tablespace type
  
  SQL> select * from V$version;
  
  BANNER
  --------------------------------------------------------------------------------
  Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
  PL/SQL Release 11.2.0.1.0 - Production
  CORE    11.2.0.1.0      Production
  TNS for 32-bit Windows: Version 11.2.0.1.0 - Production
  NLSRTL Version 11.2.0.1.0 - Production
  
  SQL> archive log list
  Database log mode              No Archive Mode
  Automatic archival             Disabled
  Archive destination            USE_DB_RECOVERY_FILE_DEST
  Oldest online log sequence     71
  Current log sequence           73
  SQL> alter tablespace users offline ;
  
  Tablespace altered.
  
  SQL> alter tablespace users online;
  
  Tablespace altered.
  

  For the first thing, the declaration this tablespace users may be taken offline is indeed correct . Now the System and Sysaux are containing the data dictionary of the database and without whom, there is no point that this database can work. This is why its mandatory for them to be online and if off line by crushing or something, the recovery of them must also occur in the phase (Mount) offline.

  The Undo tablespace is required to service the mechanism of reading-consistansy of the database. If quoted to be mandatory (i.e. your AUM), it cannot be taken offline and being crushed, his recovery would be as necessary a restoration in offline mode. The Temp tablespace is a special tablespace (he was't it always in all previous versions) for the database and if it's a must for the datbase for read-only and do something that will exceed PGA. So it cannot be disconnected as well.

  Now, I'm sure you mean you know all this, what about users being able to get offline, right? Well, the answer is quite simple. Everything simply because you establish a default tablespace for the user be available as their default tablespace (further a quota to spread on it), it does not change any property of this tablespace. Its the tablespace, unlikely normal Temp, cancel and so on. He did justto do that people do not, by mistake, make objects in the System tablespace. A tablespace of default tablespace of the user database for the creation of objects, does not mean that all of a sudden, its now taking a special meaning. That is why its possible to redo this tablespace still transactions that apply to any other tablespace (by default) with the exception of the system those. I don't think to dig any deeper. There is no other explanation that it is possible for this behavior of the IMO.

  HTH
  Aman...

• Why oracle stores no length for CHAR, VARCHAR2, and other types?

  Hi friends,
  
  I was reading a book by Tom Kyte on the types of character data.

  +For character strings,Oracle will have a leading length field of 1 to 3 bytes, or when they are NULL they+ 
  +will be represented by a single byte value of 0xFE.+
  +If the length of the string is less than or equal to 250, Oracle will use 1 byte for the length.+
  +All strings exceeding 250 bytes will have a flag byte of 0xFE followed by 2 bytes that represent the length.+

  Can someone explain please why Oracle shops length here and where exactly is - it used by Oracle?
  For strings longer than 250 bytes, it sets a flag to 0xFE byte. What is the usefulness of this indicator byte and how is it useful?
  Is there anyway we can see the length bytes that are stored by Oracle? I tried with the DUMP function, but it is to show that the value of the length, but not the bytes it stores for her.
  
  Kind regards
  Anthony

  I think that Oracle chose 0xFF to represent a null value
  (a) Yes
  (b) should be 1 as another column can follow

  Concerning

  Etbin