Export dump and the Oracle performance

My prod system is 10g r2 on an AIX box with 4 CPU ( oracle only uses two) and 4 G memory. It also installs with a BAAN 5c system. I noticed that when I did the export dump, the Oracle system was slow to other query processes by the BAAN and the message "Memory Notification: Library Cache Object loaded into SGA
Heap size 8356K exceeds notification threshold (2048K)" were shown up in alert log. Although there is a solution in the metalink to avoid the notification, I think it does not solve the problem underneath. Some system maintenance processes also caused that. So How can I track the real cause(s)? Maybe my original configuration need to reconfigure.

Oracle was slow because running export is rather resource intensive (lots and lots of I/O). You generally want to avoid generating an export when the system is being actively used. As an aside, I hope that you're not using the export as your primary backup...
The message you're getting (presumably in Enterprise Manager) would appear to be both reasonable and probably not the source of the slowdown. I'm not shocked to learn that BAAN has some objects which are relatively large, running an export brings them into the library cache, which generates the alert. I would expect that you'd have to get that alert in order to do an export, so it may be reasonable to turn it off if this is a regular thing.
Justin

Similar Messages

  • How Standalone OC4J and the Oracle Application Server linked to each other

    Hi ,
    Please let me know what is the relation between a Standalone OC4J and the Oracle Application Server ?
    In my application The Oracle Application Server is installed in C drive and there is a Stand alone oC4j .
    Please tell me how they are linked to each other .
    Thank you .

    It depends on your need and what features and functionality of Oracle Application Server you are interested in. Note that OC4J is a component of Oracle Application Server that can run as a standalone component or within Oracle Application Server. Question #6 in earlier mentioned FAQ document states:
    OC4J standalone is J2EE 1.3 compatible (1.4 and 1.5 too in later releases of OC4J) and is able to be used in both development and small scale production environments. OC4J standalone provides its own inbuilt HTTP/S listener to allow clients to execute Web applications that it has deployed. Application deployment and server configuration of OC4J standalone is performed by the manual editing of a simple set of XML files.
    For large scale enterprise deployments, the Oracle Application Server product is likely to be more suitable with its comprehensive feature set including Oracle HTTP Server, process monitoring and management capabilities, and its configuration and management console.
    So if you are happy with what you see in standalone OC4J and scalability isn't an issue for you, you can stay with standalone OC4J and don't need to move to Oracle Application Server.
    PS: If you do decide to use OC4J that's embedded with Oracle Application Server, you can't move your standalone OC4J inside Oracle Application Server. You will have to reconfigure embedded OC4J again (as you might have done with standalone OC4J) and redeploy your applications.
    Hope this helps.
    Thanks
    Shail

  • Oracle Workflow Server in a SUN machine; and the Oracle

    Hi All
    I have installed Oracle Workflow Server in a SUN machine; and the Oracle
    Workflow Client (Builder) in a WIN2000 machine.
    In the SUN machine, I already had installed Oracle9i and OMS (Oracle
    Management Server) version 9.2.0.1.0.
    In the WINDOWS 2000 machine, I already had installed the Oracle9i
    Client and the OWB (Oracle Warehouse Builder) version 9.0.3.35.0. Workflow
    Builder
    was installed together with Oracle Client components.
    I have a OWB (Oracle Warehouse Builder) project, and I deploy and schedule
    JOBS in OEM.
    Now I need Oracle Workflow in order to better ordering, manage and schedule
    that jobs.
    I'm reading the following documentation:
    Oracle Workflow Guide (Release 2.6.2)
    http://www.csis.gvsu.edu/GeneralInfo/Oracle/workflow.920/a95265.pdf But I think Oracle Workflow is very complex to learn quickly.
    My OWB Project has several mappings that load tables in a Data Warehouse.
    When I deploy these mappings one by one, in OEM, I can execute them in OEM
    and they work well.
    But when I use the Workflow Deployment Wizard, I create a new Workflow
    Project (Item type and Process).
    But I can't Launch this process. I go to Launch Process in Workflow page
    (http://:/pls//) but when I iniciate the project,
    it stops in the OWB Standard Begin Function. It does not go ahead.
    Well, when I use the Workflow Deployment Wizard inside OWB, it generates
    jobs that can
    be viewed inside OEM (Oracle Enterprise Manager) too. But when I submit
    these jobs, I can't get success too.
    Can somebody help me? May you guide me in order to solve my problem, or send
    me some
    documents or links about workflow, and it's integration with OWB and OEM?
    Best regards

    I think this might be more of an OWB question, and how OWB uses workflow, I have sent an email internally to try and get someone to responsd. You should also consider contacting Oracle Support if you are having these kinds of problems.
    Hi All
    I have installed Oracle Workflow Server in a SUN machine; and the Oracle
    Workflow Client (Builder) in a WIN2000 machine.
    In the SUN machine, I already had installed Oracle9i and OMS (Oracle
    Management Server) version 9.2.0.1.0.
    In the WINDOWS 2000 machine, I already had installed the Oracle9i
    Client and the OWB (Oracle Warehouse Builder) version 9.0.3.35.0. Workflow
    Builder
    was installed together with Oracle Client components.
    I have a OWB (Oracle Warehouse Builder) project, and I deploy and schedule
    JOBS in OEM.
    Now I need Oracle Workflow in order to better ordering, manage and schedule
    that jobs.
    I'm reading the following documentation:
    Oracle Workflow Guide (Release 2.6.2)
    http://www.csis.gvsu.edu/GeneralInfo/Oracle/workflow.920/a95265.pdf But I think Oracle Workflow is very complex to learn quickly.
    My OWB Project has several mappings that load tables in a Data Warehouse.
    When I deploy these mappings one by one, in OEM, I can execute them in OEM
    and they work well.
    But when I use the Workflow Deployment Wizard, I create a new Workflow
    Project (Item type and Process).
    But I can't Launch this process. I go to Launch Process in Workflow page
    (http://:/pls//) but when I iniciate the project,
    it stops in the OWB Standard Begin Function. It does not go ahead.
    Well, when I use the Workflow Deployment Wizard inside OWB, it generates
    jobs that can
    be viewed inside OEM (Oracle Enterprise Manager) too. But when I submit
    these jobs, I can't get success too.
    Can somebody help me? May you guide me in order to solve my problem, or send
    me some
    documents or links about workflow, and it's integration with OWB and OEM?
    Best regards

  • Howto sync between alsb console and the Oracle Workshop

    I was wondering if it's possible to sync between my project in Oracle Workshop for WebLogic and the project which is deployed on the server.
    When i do some changes in the alsb-console i want to see/sync/import them in my project in the workshop too.
    Is this possible?
    Thanks!
    Eric

    Hi Eric,
    I'm afraid this is not possible. We recommend when using Oracle Service Bus you use either Workshop OR the console, but not make changes in both as things can quickly get out of sync. I would recommend doing your development in Workshop (as it is more developer friendly) and then switch to the console to make changes you cannot perform in Workshop, such as the operational settings (throttling, monitoring, etc.)
    Thanks,
    Chris

  • Using Calendar client version 4.6 on Unix to connect to Calendar Server 4.0, I get a core dump and the client does not launch

    I am using Communicator Pro with Enterprise Calendaring, Calendar client 4.6,
    and when I attempt to launch it I immediately get a core dump and it does not
    launch.
    <P>
    The 4.6 motif client is not able to run against the 4.0 Calendar Server.
    It is strongly recommended that you use the 4.61 motif Calendar client (bundled with the 4.7
    Communicator Pro release) or 4.51/4.52 against Calendar Server 4.0. There are
    known problems when going offline while using the 4.5 client on any platform
    against the 4.0 Server. Both the 4.51 Windows and the 4.52 motif clients have
    repaired this problem.

    You can get Firefox 3.6 from http://www.mozilla.com/en-US/firefox/all-older.html

  • Simple viewer won't export comments and the sequence is wrong

    I am stuck....
    I have added simple viewer on my site, but the comments of every photo are not there.
    - In iphoto i have added the comments with every photo.
    - With iphoto export, i made the simple viewer file. Everything works fine accept from the comments they are just not exported.
    - Also the sequence of the photo's are different in the simple viewer. It seems like the photo's are displayed (or exported) at random.
    Has anyone an idea how I can find a workaround and to get these two issue's solved?

    Hi Dhara,
    I've been working with Simple Viewer for a couple of weeks now (you can catch my posting and all the replies on the Boards).
    Regarding your issues, my recollections are as follows....but, remember I'm a novice and not an expert on this:
    1. Comments: Make sure that you check 'Use Comments' in iPhoto Export.
    2. Sequence of Photos: I had the same problem with this. I went back to my iPhoto folder and actually changed the file names to 1.jpg, 2.jpg, etc. Then, in iPhoto, I sorted the photos: View, Sort Photos, By Title. I had to move a few manually in the iPhoto display, as numerically sorted some fall out of 'order'. After saving this folder, run iPhoto Export and proceed as you did before.
    Again, I am a newbie...so maybe see if some of the 'gurus' on the Boards post a fix as well. And, let me know if this works!
    Good luck

  • [EWS][FastTransfer][MS-OXCFXICS] Question about paging in the Fast Transfer Stream after exporting message and the meaning of the config in the previous section of the stream.

    Hi, all.
    First, I want to construct a fast transfer stream by programming not using the export item's stream and to import the stream to the folder by posting import request in EWS.
    So, I export the item by EWS and investigate the exporting stream's struct by reading the [MS-OXCFXICS].pdf. And I can parse each MAPI property, marker, meta-property from the stream.
    I find that the stream is made of many pages, most page's length is 0x7BC0, some are less then 0x7BC0. And in the previous section of the stream, there are some configurations. So the questions are:
    1. What is the paging rule? or What is the principle of the paging? Whether a complete propvalue/marker(ref from [MS-OXCFXICS].pdf) must be contained in a page except some binary type?
    2. Why the page's binary length is 0x7BC0? Or where is the configuation about paging length?
    3. The stream is made of the FXOpcodes and the subbuffer about the FXOpcodes, and I find the mean of the FXOpcodes:
    internal enum FxOpcodes
    None = 0,
    Config = 1,
    TransferBuffer = 2,
    IsInterfaceOk = 3,
    TellPartnerVersion = 4,
    StartMdbEventsImport = 11,
    FinishMdbEventsImport = 12,
    AddMdbEvents = 13,
    SetWatermarks = 14,
    SetReceiveFolder = 15,
    SetPerUser = 0x10,
    SetProps = 0x11
    and in the stream there are 4 part(after I parse the stream):
    A. OpCode:Config, value:0000000001000000
    B. OpCode:IsInterfaceOk, value:010000000703020000000000C00000000000004600240080
    C. OpCode:TellPartnerVersion, value:000F91838417
    D and follows is: OpCode:TransferBuffer, Count:31680
    So 
    a. What is the OpCode:Config value meaning?
    b. Some parts of The OpCode:IsInterfaceOk' value is the IID_IMessage("00020307-0000-0000-C000-000000000046"), What is other parts meaning?
    Thanks in advance for any ideas.

    Hi Haiyang,
    Your questions are not covered by the Open Specification documentation and this is documented in [MS-OXWSBTRF] — v20141018 section “1.3 Overview”.
    The upload and export data stream is an opaque format that only needs to be understood by a server implementation. The client only serves as a repository for the opaque data stream so that it can be uploaded to the server at
    a later time.
    You made good progress in interpreting the buffer using other documents. Why do you need to be able to parse the data? Please describe your project in details and I might submit a suggestion to document the layout of the data stream,
    if your project justifies it. You can send the description directly to me.
    Thanks, Vilmos

  • [EWS][FastTransfer] Question about paging in the Fast Transfer Stream after exporting message and the meaning of the config in the previous section of the stream.

    Hi, all.
    First, I want to construct a fast transfer stream by programming not using the export item's stream and to import the stream to the folder by posting import request in EWS.
    So, I export the item by EWS and investigate the exporting stream's struct by reading the [MS-OXCFXICS].pdf. And I can parse each MAPI property, marker, meta-property from the stream.
    I find that the stream is made of many pages, most page's length is 0x7BC0, some are less then 0x7BC0. And in the previous section of the stream, there are some configurations. So the questions are:
    1. What is the paging rule? or What is the principle of the paging? Whether a complete propvalue/marker(ref from [MS-OXCFXICS].pdf) must be contained in a page except some binary type?
    2. Why the page's binary length is 0x7BC0? Or where is the configuation about paging length?
    3. The stream is made of the FXOpcodes and the subbuffer about the FXOpcodes, and I find the mean of the FXOpcodes:
    internal enum FxOpcodes
    None = 0,
    Config = 1,
    TransferBuffer = 2,
    IsInterfaceOk = 3,
    TellPartnerVersion = 4,
    StartMdbEventsImport = 11,
    FinishMdbEventsImport = 12,
    AddMdbEvents = 13,
    SetWatermarks = 14,
    SetReceiveFolder = 15,
    SetPerUser = 0x10,
    SetProps = 0x11
    and in the stream there are 4 part(after I parse the stream):
    A. OpCode:Config, value:0000000001000000
    B. OpCode:IsInterfaceOk, value:010000000703020000000000C00000000000004600240080
    C. OpCode:TellPartnerVersion, value:000F91838417
    D and follows is: OpCode:TransferBuffer, Count:31680
    So 
    a. What is the OpCode:Config value meaning?
    b. Some parts of The OpCode:IsInterfaceOk' value is the IID_IMessage("00020307-0000-0000-C000-000000000046"), What is other parts meaning?
    Thanks in advance for any ideas.

    OK. Thank you.
    Today I build a stream with a paging rule as below:
    1. Keep each page's byte length is 31680.
    2. Make sure a complete property value in a page except binary type property whose value length is larger
    than 31680.
    And it can work very well when import the stream to the folder by EWS.

  • Using PreparedStatement and the Oracle Error ORA-1000

    Hi,
    I have a question about PreparedStatement objects that is not so simple to explain for me. What I would like to know is: if I use a PreparedStatement following traditional and generic steps:
    1- PreparedStatement pStmt = Connection.prepareStatement(sQuery);
    2- pStmt.setXXX(i,j);
    n - pStmt.setXXX(i,j);
    n+1 - ResultSet rs = pStmt.executeQuery();
    n+2 - while(rs.next()){ ... retrive ResultSet data  ... }
    n+3 - rs.close()
    n+4 - back to point number 2
    and at the end (as you can see in the point numbered n+4), instead of closing the PreparedStatement pStmt using the close() method, I reuse the PreparedStatement pStmt comeing back to the point numebr 2 and setting again all its parameters with new values ... then ... what heppens in the Oracle database ? Has been the cursor (so the mamory area), associated to my PreparedStatement object pStmt, duplicated or is it the same ?
    I know that Java allows you to do this kind of operations with PreparedStatement, and I know that in tha Java Documentation is explained to follow this strategy to optimize the execution time because in this way the same PreparedStatement is precompiled and prepared only once. But if I do a for loop following the steps explained before, after many iterations I have the error "ORA-1000: maximum open cursors exceeded". This error is the reason of my question. Does this error means that it's mandatory to close a PreparedStatement always, otherwise if you reuse it without closing it then the corresponding database cursor will be duplicated ? If it is so, then I think this is a contradiction with official java documentation ...
    I'm using Oracle8i (version 8.1.7) and Oracle JDBC Thin Driver (Windows NT) for use with JDK 1.2.x. Moreover, in my database istance the parameter "maximum open cursor" is equal to 800 ...
    Thank you very much for suggestions :-)

    There is no need to close a prepared statement or its resultset for every iteration.
    After the first iteration in a loop, all subsequent executions of it will close the previous resultset. By adding close() method, you are making one extra costly call to the DB for no reason.
    Following is the sample code.I know what you are saying. In fact at the beginning I wrote my code in the same way of your sample (see the code of my first post at the begin of this page).
    But doing so, after thousand iterations of the loop, I had "Oracle Error ORA-1000 : maximun open cursor exeeded" even if in my database istance the parameter "maximum open cursor" is equal to 8000.
    At this moment in my code, for each iteration, I close the PreparedStatement and in this way I don't have anymore the error :-((
    So it seems that only in theory we can reuse a preparedStatement without closing it. In fact if we see the oracle system table "$open_cursor" (as Konrad Pietzka suggest me) we can find that, for each interation,
    at our line code "rs = pstmt.executeQuery();" correspond a new cursor in the database: this means that for each method "pstmt.executeQuery()" the database open a new cursor and do not use the previous one as it should.
    I posted a question two months ago to search if someone had the same problem (it seems that Konrad Pietzka had the same situation) and was able to explain me what is the cause.
    The only reason I found by myself for this problem, is that probably the Oracle JDBC Thin Driver for Windows NT/2000 has some bugs... but I'm not sure ...
    Thank you very much for you time !!
    bye :-))
    Fidalma

  • Multiple Recordsets and the Oracle OleDB Provider

    We are trying to return multiple recordsets from an Oracle SP through the Oracle oledb provider using VB. First of all, is this possible and is there any example code out there. Having trouble finding some info on this.
    Thanks

    After reading install guide I figured out waht might be the problem.
    I put OLEDB install into the same home (ora81).
    It was suppose to go into different home directory (ora92).
    Also, OLEDB is one of few components which can't exist in more then one
    home dir on single machine. Installer knows that, and skips install of
    OLEDB provider itself. So in my case I had to first uninstall OLEDB provider 8.1.7
    and then install OLEDB provider 9.2
    Thanks,
    ...dejan

  • Apex and the Oracle Change Management Pack

    Hi everyone
    I'm looking for some feedback from anyone that has used the Oracle Change Management Pack (CMP) with an Apex development (or is it really only for Oracle Apps)?
    Could the CMP be used against the APEX_04000 schema to sync a Dev environment with Test and Production? If so, this would nicely lead into an Agile approach to development, even perhaps Continuous Deployment :)
    I'm concerned that I can't find anything on the web about Apex and the CMP, so I suspect it can't be done :(
    Any comments gratefully recieved.
    Shunt

    You are going to have to provide a lot more information for anyone to help you from the basic ... like in what version ... to the specific ... what functionality do you want?
    In general none of the "packs" run any code not in the database in the form of built-in packages. For example much of the functionality of OEM Grid is implenented in DBMS_ADVISOR, DBMS_MONITOR, DBMS_SERVER_ALERT, etc.
    But without specific questions no specific answer is possible.
    You can find demos of the above referenced packages here:
    http://www.morganslibrary.org/library.html

  • Dropping Rollback table space in export dump and use the undo table space

    Hi,
    I want to upgrade my DB which is on 9.2.0.1.0 to 10.2.0.1.
    I have done an export.
    In order to import the DB I have created all the table spaces identical that of my export copy. While creating the Blank database I have already created Undo table space. In the exported DB I have a table space called rollback, how can I drop this table space and use undo table space. I believe this is the recommended table space that should be used in new releases
    Thank you

    Very briefly, these are the steps you will need to perform:
    1) Create an UNDO tablespace the same size as the Rollback tablespace, e.g.:
    CREATE UNDO TABLESPACE undo
    DATAFILE ‘/ora/$ORACLE_SID/001/oradata/$ORACLE_SID/undo_01.dbf’ size 1000M;
    2) Add the initialisation parameters UNDO_MANAGEMENT=AUTO, UNDO_TABLESPACE=UNDO. Omit the UNDO_RETENTION parameter unless it is to be set to a value other then the default of 900 (seconds)
    3) Remove the initialisation parameter ROLLBACK_SEGMENTS
    4) Bounce the database to bring the new parameters into effect.
    5) Drop the Rollback tablespace including contents and datafiles
    Cheers,
    Jason

  • XML Publisher and the Oracle Shipping Transactions Form

    I am running 11.5.10.2. I need to print Simplified Chinese characters in a document. We converted to the UTF-8 character set so the server can render these on forms and such. I am using Oracle Reports 6i to produce a report, but due to the limitations with this product it will not produce PDF output with Chinese characters. XML Publisher however works great. The concurrent manager is set so the report produces XML which is then set to the template and out comes a really cool looking layout with Chinese that can be viewed, printed, or e-mailed anywhere.
    Now I have put this report into a document set so it can be printed from the Shipping Transactions form. But it does not work. There is no output. Metalink Note:312194.1 states development only supports XSL-FO and this will not be available until R12.
    Is this really true? Has anyone else come across this? Any suggestions on how to get around this?

    Hi
    Im assuming you are using the EBusiness Suite becasue of your reference to AR and AP modules. If so getting using an ORacle Report as a datasource for XMLP to format is pretty straightforward. There is a white paper on metalink describe the complete process - Note 295409.1. This will be available on the public site soon http://www.oracle.com/technology/products/applications/publishing/index.html
    As for AR and AP, there is no reason at all why you can not use XMLP with the existing Oracle Reports in those modules. You will start to see more and more modules providing embedded reporting functionality in the EBS based on XMLP technologies.
    Regards
    Tim

  • Oracle 9i   (and the Oracle website!)

    Hello, I've been using this website for many years now and have always found it extremely difficult to navigate. Simple tasks like finding system requirements (e.g. supported operating systems, etc.) is nearly impossible in some cases. Lot's of sales propaganda everywhere though, no shortage of that.
    Just being facetious there. :-) I'm very fond of using Oracle products, and I'm no MS fan. Just don't like how this site is structured, that's all.
    Ok, here's my dilemma. I have: a
    1). Win2K Server machine for database;
    2). SuSE Linux 10.2 machine for database client.
    I figure my best bet is to go with Oracle 9i for both database and client, since SuSE 10.2 isn't supported yet for the newer Oracle 10g. So, where can I download Oracle 9i database and client?

    Thanks, I have also found that link. I go to:
    www.oracle.com ---> database ----> Standard_Edition
    Then there's a FREE DOWNLOAD link at the right for Oracle 10g (but I want Oracle 9i). I press the link and there's a long listing of 10g downloads. At the bottom there's a small link for 'previous database releases.' That's the link you provided. No info about supported systems anywhere. I have to fish through the full documentation index for that. Couldn't this be easier somehow? And shouldn't databases and clients be listed separately?
    Meant constructively of course and nothing against you, B548204BB. I appreciate your response :-)
    Anyway, it seems there is no 'secret link' anywhere for a Linux-based 9i client. But, I just saw something to indicate that Oracle 10g is now supported on SLES 10.2. That might solve my problem! I have openSUSE though. Anyone have success with installing Oracle 10g database or client on openSuSE 10.2?
    I might start a new thread for this.....

  • ASD Export/Import and the Anti-Marshal (kind of like a supervillain)

    Just tried to export a single resource action and import it back into the same environment with a new "prefix" and I get the following errors on validation....  same goes for importing it into another environment.  (which we will call "new environment"... where the Resource Action doesn't exist)  The workflow that the RA points to exists in the current environment because there is an RA that already points to that workflow.  In the new environment, the workflow exists due to a successful vRO sync.  (we can then build a successful form against the workflow in the new environment.)
    It seems like it is complaining about the Resource Type, but none of those were selected for export.. so one would expect that they wouldn't be imported.
    Anybody try out the export/import functionality in ASD and have it be successful?
    When we build a form in our Lab environment, we have to completely rebuild it in two other environments.  Though building forms in ASD is exceedingly fun and intuitive, we would really like to have a working export/import capability!

    It throws same error for exporting/importing ASD blueprints( Nothing else was selected to export like actions etc). In this case ASD blueprint exists in the destination.  Sean, did you make any progress?
    Failed on Can not unmarshal entity:resource-mappings of type: com.vmware.vcac.designer.rest.stubs.CsResourceType
    Warning
    RHEL
    Service Blueprint
    blueprint.name.duplication
    Info
    RHEL
    Service Blueprint
    blueprint.validation.success
    Error
    Can not unmarshal entity:resource-operations
    Resource Action
    resourceoperation.unmarshal.exception
    Error
    Can not unmarshal entity:resource-types
    Custom Resource/Mapping
    resourcetype.unmarshal.exception
    Error
    Can not unmarshal entity:resource-mappings
    Custom Resource/Mapping
    resourcetype.unmarshal.exception

Maybe you are looking for