parallel expdp on bigfile

Similar Questions

• expdp on cluster - how to determine where is the dumpfile GB

  I am running with parallel expdp = cluster 4 = y and everything works fine except that I have to get on all 4 nodes where the file was put.
  Is there a sql query that would tell me on what node of the file would be written to?

  Hello

  First of all, make sure points of object directory of shared storage that can be seen by all Oracle RAC designated instances to run Data Pump of work processes.
  dba_directories queries and see the directories of dumpfiles,.

  See this post
  http://download.Oracle.com/otndocs/products/database/enterprise_edition/utilities/PDF/datapump11gr2_twp_rac_1009.PDF

• Export data from site 1 site 2

  Hi all
  
  I have Oracle 10GR 1 migration to Oracle 11 GR 2 site 1 site 2. I can't use parallel expdp option (Oracle 10.1 standard edition).
  I had to do an export from site 1 on a NAS and I had to put the SIN on the site 2 and do an import.
  
  You have a way to do it in a much faster way?
  
  Thank you!
  
  Published by: user12064507 on October 19, 2011 08:26

  PL see the steps in the Upgrade Guide - http://download.oracle.com/docs/cd/E11882_01/server.112/e23633/expimp.htm

  HTH
  Srini

• Accelerate the expdp with PARALLEL work and the size of the settings FILE

  Every day we save 6 patterns with a total area of 80 GB.
  Oracle documentation, I think the PARALLELS servers work well when we split the dump file, because each slave process can work with a separate file.
  But I don't know how many parallel processes should be generated and the number of files this dump file must be split?
  
  The command expdp that we plan to use
  

  expdp userid=\'/ as sysdba\' SCHEMAS = schema1,schema2,schema3,schema4,schema5,schema6  DUMPFILE=composite_schemas_expdp.dmp LOGFILE=composite_schemas_expdp.log  DIRECTORY=dpump_dir2 PARALLEL=3

  Related information:
  
  11.2.0.2
  
  Solaris 10 (x86_64) on HP Proliant Machine
  
  8 CPU with 32 GB of RAM
  

  SQL > show parameter parallel
  
  NAME                                 TYPE        VALUE
  ------------------------------------ ----------- ------------------------------
  fast_start_parallel_rollback         string      LOW
  parallel_adaptive_multi_user         boolean     TRUE
  parallel_automatic_tuning            boolean     FALSE
  parallel_degree_limit                string      CPU
  parallel_degree_policy               string      MANUAL
  parallel_execution_message_size      integer     16384
  parallel_force_local                 boolean     TRUE
  parallel_instance_group              string
  parallel_io_cap_enabled              boolean     FALSE
  parallel_max_servers                 integer     32
  parallel_min_percent                 integer     0
  
  NAME                                 TYPE        VALUE
  ------------------------------------ ----------- ------------------------------
  parallel_min_servers                 integer     0
  parallel_min_time_threshold          string      AUTO
  parallel_server                      boolean     TRUE
  parallel_server_instances            integer     2
  parallel_servers_target              integer     32
  parallel_threads_per_cpu             integer     2
  recovery_parallelism                 integer     0

  resistanceIsFruitful wrote:
  But I don't know how many parallel processes should be generated and the number of files this dump file must be split?

  How many parallel processes you need, it's something you can figure out to run the tests against your db, but if you have parallel set to N, then you have at least N dump files in order to use fully all parallel threads spawned. We take backups using parallel = 6 and dumpfile is normally set to dumpfile=dbname.%u.dmp where oracle expands %u necessary if you do not explicitly list individual files.

• expdp with parallel writing in a single file on the OS

  Hi friends,
  I am facing a strange problem. Despite giving parallel = x parameter the expdp is written on a single file at the level of the OS at a time, even if it's written in several files in order (not at the same time)
  
  While on other servers, I see expdp is able to start writing in several files simultaneously. Here is the example of newspaper
  of my expdp.
  ++++++++++++++++++++
  Export: Release 10.2.0.3.0 - 64 bit Production on Friday, April 15, 2011 03:06:50
  
  Copyright (c) 2003, 2005, Oracle. All rights reserved.
  ;;;
  Connected to: Oracle Database 10g Enterprise Edition Release 10.2.0.3.0 - 64 bit Production
  With partitioning, OLAP and Data Mining options
  Start "CNVAPPDBO4". "' EXPDP_BL1_DOCUMENT ': table CNVAPPDBO4/***@EXTTKS1 = BL1_DOCUMENT DUMPFILE=DUMP1_S:Expdp_BL1_DOCUMENT_%U.dmp LOGFILE = LOG1_S:Expdp_BL1_DOCUMENT.log CONTENT = DATA_ONLY FILESIZE = 5 G EXCLUDE = INDEX, STATISTICS, CONSTRAINTS, GRANT PARALLEL = 6 JOB_NAME = Expdp_BL1_DOCUMENT
  Current estimation using BLOCKS method...
  Processing object type TABLE_EXPORT/TABLE/TABLE_DATA
  Total estimation using BLOCKS method: 23,93 GB
  . . exported "CNVAPPDBO4." "' BL1_DOCUMENT ' 17,87 GB 150951906 lines
  Table main "CNVAPPDBO4." "' EXPDP_BL1_DOCUMENT ' properly load/unloaded
  ******************************************************************************
  Empty set of files for CNVAPPDBO4. EXPDP_BL1_DOCUMENT is:
  /tksmig/load2/Oracle/postpaidamd/DUMP1_S/Expdp_BL1_DOCUMENT_01.dmp
  /tksmig/load2/Oracle/postpaidamd/DUMP1_S/Expdp_BL1_DOCUMENT_02.dmp
  /tksmig/load2/Oracle/postpaidamd/DUMP1_S/Expdp_BL1_DOCUMENT_03.dmp
  /tksmig/load2/Oracle/postpaidamd/DUMP1_S/Expdp_BL1_DOCUMENT_04.dmp
  Work "CNVAPPDBO4". "' EXPDP_BL1_DOCUMENT ' successfully completed at 03:23:14
  
  ++++++++++++++++++++
  
  

  uname-a

  HP - UX ocsmigbrndapp3 B.11.31 U ia64 3522246036 unlimited-license user
  
  
  He hits a known bug? Please suggest.
  
  regds,
  Malika

  http://download.Oracle.com/docs/CD/E11882_01/server.112/e16536/dp_export.htm#i1006259

• Expdp parallel option

  Hi all
  
  I work with a 10.2.0.3 version RAC on a two node RAC.
  
  Every night I have an export using expdp and I need to reduce the time to spend on it. In fact, I'm with parallel = 4 but, I want to know if I can increase it. The server after I am executing the export job have 4 processors with 4 cores each.
  
  I read I can change the parallel value until the same number of available CPUs * 2.
  
  In my case, how do I count the number of CPU? 4? 16?
  
  Kind regards
  dbajug

  Parallel data pump features,

  http://download.Oracle.com/otndocs/products/database/enterprise_edition/utilities/PDF/datapump11gr2_twp_rac_1009.PDF

  Thank you

  Published by: Cj on January 5, 2011 01:46

• Error in expdp

  Hello

  I'm trying to export a schema with 2 tables with the CLOB (XML file stored) data type. While the expdp am getting error below.

  ORA-39097: Data Pump job encountered the error unexpected-12801

  ORA-39065: exception unexpected master process in HAND

  ORA-12801: error reported in the PZ99, instance XXXXXX parallel query server

  ORA-01460: dead letter or unreasonable conversion requested

  Job 'XXXXXXXX '. "" SYS_EXPORT_SCHEMA_01 "stopped because of the mistake at 03:46:32

  already my parameter PARALLEL_DEGREE_POLICY value is MANUAL. Help, please

  The error that you get a little weird, so if I were you, and if the error is reproducible, I would trace your datapump for more information.  You could follow the datapump job Oracle sessions or add "trace = 1FF0300" to the command line expdp; This will create tracking in your folder of trace files.  Search for 'trace = 1FF0300' to find out how it works.

• Error in database expdp remote

  Hello

  When you attempt to perform an export on a remote database, I get the following errors:

  ORA-39006: internal error

  ORA-39065: exception unexpected master process in SHIPPING

  ORA-06575: Package or function KUPM$ MCP is in an invalid state

  ORA-39097: Data Pump task encountered the unexpected-6575 error

  Local DB running the export

  Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64 bit Production

  PL/SQL Release 11.2.0.4.0 - Production

  CORE Production 11.2.0.4.0

  AMT for Linux: Version 11.2.0.4.0 - Production

  NLSRTL Version 11.2.0.4.0 - Production

  Exported remote DB

  Database Oracle 12 c Enterprise Edition Release 12.1.0.2.0 - 64 bit Production

  PL/SQL Release 12.1.0.2.0 - Production

  CORE Production 12.1.0.2.0

  AMT for Linux: Version 12.1.0.2.0 - Production

  NLSRTL Version 12.1.0.2.0 - Production

  Export command

  expdp / full = directory = network_link db_dump remote_db dumpfile = db.dmp logfile = compression = db.log = all parallel = 3 = db_dump exclude = job_name statistics, Greenbelt:------"IN\ (\'\PLAN_TABLE\',\'TOAD_PLAN_TABLE\',\'OL\$\'\) \".

  Please note that the distance DB is a DR of guard of my production server database, the following sql commands are used to bring the distance DB online for export:

  ALTER DATABASE RECOVER MANAGED STANDBY DATABASE CANCEL;

  ALTER DATABASE OPEN READ-ONLY.

  This process worked when DBs local and remote were 11g, what Miss me here to complete this export with 12 c?

  Thank you

  Hey,.

  I wanted to just follow up on this question in case someone
  else is going through the same problem. To solve this problems, two steps, first,
  install the patch 21252520 then around following works:

  1. Physical conversion standby DB
     on the eve of the snapshot
  2. Open read-write
  3. Convert the eve of the snapshot
     physical standby

• Parallel parameter really reduce time for export?

  Hi all;

  I do realize too fast, if I use a parallel option when exporting.

  As we know, the DIRECT PATH is much faster than the ACCESS EXTERNAL path.

  With EXPDP reference documents.

  When taking export using the parallel option, data access method would be 'EXTERNAL TABLE'

  For large tables data pump always using external path rather than direct path.

  in above case, if the size of the table in popular Go with "n" number of partitions, I can use parallel when the export option?

  The version is 11.2.0.1 on RHEL

  Hello

  In this case, there are two types of parallel

  external table 1) using a parallel query - what is likely to be slower in many cases only a process series using the direct path - query in parallel or single worker process usnig

  (2) extract several tables "at the same time" and each made as direct path. -many workers all do another object at the same time

  Thus, for example, that if you say Parallels 4, it is possible to get 4 datapump work process, each of these processes can extract another table at the same time, each of these excerpts is in the direct path,.

  For most tables that's how it will work, if you use the query or you have some weird datatype or feature, then run the path may not be possible but in almost all cases of Parallels to export must be faster - because of the parallel unloading of objects.

  Parallel queries to a single table for the external table access is generally slower.

  Hope that makes sense...

  See you soon,.

  Rich

• Expdp RAC error

  HI guys

  I tried to expdp in the database of node 2 RAC 11 GR 2.

  expdp dumpfile=testenet_05052014%U.dmp logfile = testenet.log patterns = test_enet = compressing backup directory = all the cluster parallel = N = 2

  Export: Release 11.2.0.4.0 - Production on Mon May 5 08:49:03 2014

  Copyright (c) 1982, 2011, Oracle and/or its affiliates.  All rights reserved.

  Username: test_enet

  Password:

  DEU-01034: operation generated ORACLE error 1034

  ORA-01034: ORACLE not available

  ORA-27101: shared memory realm does not exist

  Linux-x86_64 error: 2: no such file or directory

  UDE 00003: all authorized logon attempts failed

  Thanks in advance

  REDA

  Hello

  When you say sqlplus works how you connect - expdp uses the same process of connection so if we work the other should too.

  you do sqlplus user/pass@db

  and expdp user/pass (without the @db)?

  Maybe that's part of it (where my question on ORACLE_SID).

  Be aware that the "directory" is an object of directory oracle rather than a directory on the file system - so something like

  SQL > create directory as ' / tmp';

  expdp directory = test

  See you soon,.

  Rich

• Ordinary Expdp GR 11, 2 DB and Impdp to RAC 11 GR 2

  Hello

  I'll have to expdp single database Instance 11.2.0.3 and Impdp to 11.2.0.4 CARS two database nodes.

  Do you know the guy a link with a good material for the research, I can follow as a tutorial?

  One thing I know that it is good to optimize execution is implemented table_exists_action = replace.

  Do you know other things that I have to perform in order to have a good performance?

  Thank you

  KZ

  consider of PARALLEL and the PARALLEL_EXECUTION_MESSAGE_SIZE parameter.

  validation constraints

  check the permissions and directory size

  Read also: data pump Performance

  HTH

  Tobi

• Accelerate exp/imp or expdp/impdp

  Hello

  Is it possible to speed up and exp/imp or expdp/impdp that already works? Is is possible to speed up a run RMAN backup or restore RMAN process?

  Kind regards

  007

  To accelerate a datapump export market and import you can attach to employment and increase the level of parallelism... impdp attach = job

  I don't know any way to speed up a running RMAN backup.

  To expedite an RMAN restore, you can kill the restoration and re - run using several channels.  The restoration should take up where he left off and can run faster with many channels.  It is relevant only if you have several items from backup.

• expdp crashes after reaching 99% with - wait for a message unread on the broadcast channel

  Here's what I do:

  I export a complete database of PeopleSoft on Solaris 10 using EXPDP.

  I am performing the task as the SYSTEM user and dumping out standard time and labor in a log file.  The $TODAY is just a variable export stamp date.

  
  It's my run line:

  expdp PARFILE DUMPFILE=psptst.$TODAY.f%U.dmp USERID = SYSTEM expdp.par = / * logfile = expdp_$ ORACLE_SID. $TODAY.log

  My parfile looks like this:

  DIRECTORY = data_pump_dir

  FULL = Y

  PARALLEL = 8

  CONTENT = ALL

  exclude = pattern: "IN (select username from dba_users where useridentifier < 54)" "

  I'm running the command on the same server where the database is located, I have plenty of space.  The server is a Solaris M5000 with 64 GB of RAM, a lot of CPU power and it is the only database running here.  I also close the database and he grew up in restricted so mode nothing else can connect to it since this export is for the migration to Linux and 11.2.  The current database is 11.1.0.7.5.

  My short for export until that strikes it full to 99% then crashes just for hours, and never ends.  I checked the TOAD jobs and all the worker nodes are ASSETS, but the dump file, the file log and standard file are all NO more.

  Here is the output of STAT when I connect to the use of data pump:

  Export > stat

  Job: SYS_EXPORT_FULL_01

  Operation: EXPORT

  Mode: FULL

  Status: EXECUTION

  Bytes processed: 145,006,844,064

  Completion percentage: 99

  Current parallelism: 8

  Number of job errors: 0

  Empty the file: /ora_exports/psptst/psptst.2013.Oct.14.1814.f%u.dmp

  Empty the file: /ora_exports/psptst/psptst.2013.Oct.14.1814.f01.dmp

  bytes written: 27,062,280,192

  Empty the file: /ora_exports/psptst/psptst.2013.Oct.14.1814.f02.dmp

  bytes written: 23,638,007,808

  Empty the file: /ora_exports/psptst/psptst.2013.Oct.14.1814.f03.dmp

  bytes written: 10,316,234,752

  Empty the file: /ora_exports/psptst/psptst.2013.Oct.14.1814.f04.dmp

  bytes written: 14,783,823,872

  Empty the file: /ora_exports/psptst/psptst.2013.Oct.14.1814.f05.dmp

  bytes written: 22,908,686,336

  Empty the file: /ora_exports/psptst/psptst.2013.Oct.14.1814.f06.dmp

  bytes written: 23,083,012,096

  Empty the file: /ora_exports/psptst/psptst.2013.Oct.14.1814.f07.dmp

  bytes written: 23,221,575,680

  Empty the file: /ora_exports/psptst/psptst.2013.Oct.14.1814.f08.dmp

  bytes written: 57,970,688

  Worker status 1:

  Process name: DW01

  Status: Pending JOBS

  2 worker status:

  Process name: DW02

  Status: Pending JOBS

  Status of worker 3:

  Process name: DW03

  Status: Pending JOBS

  Worker status 4:

  Process name: DW04

  Status: Pending JOBS

  Worker status 5:

  Process name: DW05

  Status: Pending JOBS

  Worker status 6:

  Process name: DW06

  Status: Pending JOBS

  Worker status 7:

  Process name: DW07

  Status: Pending JOBS

  Status of worker 8:

  Process name: DW08

  Status: Pending JOBS

  The status of all the expectations of workers are "wait message unread on the chain."  The said master worker, "the db file sequential read" as he wait status.

  I have tried this with a value of 4 and 16 parallel and get the same result.  I also tried with no parallel set and get the same result after 2 hours it crashes just indefinitely.

  I noticed that once the + treated bytes: 145,006,844,064 hit that number he stops just whatever the parallel level set.  I checked and the MAX_DUMP_FILE_SIZE database is set to unlimited.

  There is no old jobs in the dba_datapump_jobs and the status of my job is RUNNING.

  
  

  The alerts log is clean just redo log rolls that I expect.

  
  

  This has me scratching my head as it appears that the job runs just forever, but don't go no where.

  With the help of the Support of Oracle solution has proved to be something very simple.

  The statistics of the SYSTEM were stale, once updated the EXPDP stole!

  I ran these DBMS (as SYSDBA) positions:

  EXEC DBMS_STATS. GATHER_FIXED_OBJECTS_STATS (NULL);
  EXEC DBMS_STATS. GATHER_DICTIONARY_STATS;
  EXEC DBMS_STATS. GATHER_SYSTEM_STATS;

  The export not only of finishes, he's even faster than before. It has exported the entire base (450GB) in about 1 hour and 15 minutes (in the past, it took more than 3 hours).  I used a parallel framework 16.

• very slow expdp for CLOB table

  Hello

  I use Oracle 10.2.0.4 on Windows 64-bit.

  I have a table with a CLOB column and expdp tables takes about 90 minutes to complete (I tested the PARALLEL parameter with expdp and I have found no significant difference with different values Parallels or without this parameter). The size of the array is 21G in WHERE user_segments but full dump file size is 11.9 G. Normally if I export a table of this type of size a lot without any BUSINESS, it takes less than 15 minutes to complete.

  Please help if there is a way to increase the speed of export. Thank you

  > CREATE VIEW that do not include the CLOB & export it

  Can you please explain a little more? How do I receive my CLOB data?. Here I am looking for a solution accelerate export (why it is slow if it involves a CLOB), only not to ignore all data to make the Quick export.

  Salman

• Dynamically change the parallel work

  Hello
  
  now, I do pump export data for a data warehouse base,.
  
  Export has been started and I use parallel = 4, it runs now...
  
  1. say if I want to increase parallel to 16 level, how could I change dynamically during the execution of the exp?
  
  (now the exp rate is 5.1/h up to now, do you think this is normal or not...) SA want to change some settings of NFS. I think that increasing the parallel work cannot increase the speed of export, based on my past experience, somehow still feel a bit slower than expected)
  
  Thank you very much.
  
  database is 10.2.0.3 on Linux

  Hello
  You can increase the parallelism by joining the session running and using the parallel option

  for example

  Fix expdp user/pass = your_job_name

  Once logged in, you can then tap x parallel to increase parallelism

  When you say 5.1/hr 5.1 what? (GO?)

  See you soon,.
  Harry