Problems using BDC and/or HR_INFOTYPE_OPERATION

Hi Gurus.
Our team is developing several inbound HR interfaces but in integrated test in QAS env something really strange is happening during the tests.
We are using BDC data for data input in infotypes 0000 and 0001 through PA30 and for all the other infotypes we use HR_INFOTYPE_OPERATION.
Sometimes data input for 0000 and 0001 works, sometimes no, our table log gives us that the data was processed but when we check on the infotypes there is no data there. The BDC mapping was checked several times and we did some tests in another enviroment, everything is ok, the data is processed but in QAS we are having this problems.
With HR_INFOTYPE_OPERATION, when we run in background mode it give us strange error logs like "a complex application error occurs" but when i did a debug inside of that FM there is nothing missing or nothing wrong.
The enviroment seems to be very instable but basis teams says that there is no log error, but i've never seems errors like these, sometimes works, sometimes no.
The last info that i had is QAS instance that we are using was deleted 3 weeks ago and restored again.
Any suggestions?
Thanks a lot.

Hi,
If your BDC mapping is correct, check the return code after your BDC is excuted with - BDCMSG. Make sure that you give Refresh BDCDATA for every loop.
Having said this, I have also had issues with BDC every there was no error at the end, however the infotype never got updated.
Please double check all the relevant screens. I found that for a particular Inf that I was updating, the subtypes belonged to differnt screens. Can you test with the same data in your Dev or some other box? If that doesnt work, there might be a data issue.
Generally the complex application error occurs when a screen have been modified to prohibit batch processing. It is very difficult to diagnose the error in such cases. Make sure that you are passing all the important data to the FM. You might be able to get more help by searching in this forums as well.
Regards, Liz

Similar Messages

ABAP interfaces build using BDC and BAPIs can be used/replaced with the XI?

Hi,
We have develeloped our interfases using BDC and BAPIs. Now our client has decited to implement XI. Now could you please help in finding whether these interfaces can be used/replaced with the XI.
If yes, please let us know how?
Thanks and Regards,
Shweta

Singh,
Check this threads where they have discussed on how to use them with XI:
import BAPI into XI
Re: interface of xi with bapi
Re: BDC
Regards,
---Satish

Whwn to use BDC and When to use LSMW

Hi all,
Kindly tell me when to use BDC and when to use LSMW.
Please give me some scenarios, so that I can get a clear idea.
Note : Do not give the difference between BDC and LSMW.
Thanks to all in advance.
Regards,
Rajesh P

Hi Rajesh,
LSMW is for standard SAP applications and BDC is for customized applications.
Its worth to check the below link. Its already answered by our Community
Diff between LSmw and Bdc
Regards,
Suresh

Any problem using bseg and bkpf tables

For fico details i using bseg and bkpf tables.
I noted that programming is very difficult because of these tables are cluster table.
plz give me other tables
if i using these tables any problem will come?

>
mysvijai197715 wrote:
> Hi Aniesh,
>
> BSEG and BKPF are cluster tables. It contains transperant tables like BSIS, BSIK etc. For example to take vendor details usr the BSIK.If you use BSEG and BKPF it will take long time search. suppose your concern having lot of data means may be your server will get shutdown. So use only trasnperant tables like BSIS, BSIK etc.
>
>
> Regards
> R.Vijai
Incorrect. BKPF is a transparent table and not a cluster table and you can use it just like any other transparent table. BSEG is a cluster table but there is no problem selecting from it so long as you use the key of BUKRS, BELNR, GJAHR - unless you are selecting a very large amount of data, but then this can cause problems when selecting from any type of table.
The advantage of using BSEG over the other FI line item tables such as BSIS and BSIK is that it holds all the lines of an FI document while the others will hold only a subset eg BSIK will only hold lines that contain a vendor reference and BSAS holds only cleared GL account lines. Though you can only use it when you have the key. If you need to search on vendor, you can use BSIK as a starting point, but since I usually need to get hold of all the lines on an FI document, I then have to select from BSEG anyway.

Color palette problems using Toolbar and ImageList ActiveX Controls

I'm using Toolbar and ImageList ActiveX controls to implement toolbar functionality in a top-level VI.
I designed the icons used in the toolbar with a 256 color palette. When I initially added the icon images to the ImageList control and ran the VI, the icons were properly displayed in the toolbar, as expected. After saving VI, exiting LabVIEW, and reloading, it appears that a different palette has been applied to the images in the ImageList control. I'm attaching a screenshot of the problem as well.
Has anyone encountered this type of problem before? Any suggestions you provide will be greatly appreciated.
Thanks,
Zach
Software Engineer
OnWafer Technologies, Inc.
www.onwafer.com
Attachments:
icon.zip ‏3 KB

I'm using Toolbar and ImageList ActiveX controls to implement toolbar functionality in a top-level VI.
I designed the icons used in the toolbar with a 256 color palette. When I initially added the icon images to the ImageList control and ran the VI, the icons were properly displayed in the toolbar, as expected. After saving VI, exiting LabVIEW, and reloading, it appears that a different palette has been applied to the images in the ImageList control. I'm attaching a screenshot of the problem as well.
Has anyone encountered this type of problem before? Any suggestions you provide will be greatly appreciated.
Thanks,
Zach
Software Engineer
OnWafer Technologies, Inc.
www.onwafer.com
Attachments:
icon.zip ‏3 KB

BUG? JDev 10.1.3.1.0 on MacBook Pro - problem using space and delete keys

Is anyone using JDev 10.1.3.1.0 production release on a Macbook Pro? I am experiencing problems using the delete and space keys. The delete key no longer seems to work at all when using any of the key bindings including Default MacOS X. Using the spacebar also seems to have a conflict with the code completion feature. I am finding that the code completion always pops up when the space bar is used and no space is inserted with the editor. The space bar only works when used in conjuction with the shift key. Can anyone verify if they are able to use these keys in JDeveloper 10.1.3.1.0 on a Macbook Pro?
Thanks,
Richard

This was from the dmg file.
I found that in Preferences | Accelerators, Completion Insight lists 'Space' as an accelerator, right after 'CommandAIC'. Removing that fixed the problem, but I still can't fix the Delete issue.
The Delete action is assigned to the 'Delete' key, which is recognizes as the key combination fn-Delete. I tried to assign the Delete Previous Char to just the delete key, but pressing the Delete key while in the New Accelerator field is not recognized - it just pretends as if I had not pressed a key.
I'm looking at the the file:
jdevhome/system/oracle.jdeveloper.10.1.3.39.84/Default_macosx.kdf
trying to find entries for Delete and they are not in there - is there some other place I should be looking?
Thanks,
--ee

Problems using Hotmail and Twitter in Firefox

I'm having problems using both Hotmail and Twitter in Firefox (and Google Chrome) but they both seem to work fine in Internet Explorer. I can log-in to both sites but in Hotmail the drop-down menus and links don't work, and in Twitter the stream does not appear at all.
I've tried everything: deleted cache, history, cookies, disabled add-ons etc... I've also tried re-installing both new and older versions of Firefox and I've still had no joy. If anybody has any ideas for a fix I'd really appreciate it. I can't stand using IE but I am forced to at the moment.

My problem with Firefox & yahoo seems similar - so you are not alone.

Problems using MAX and PXI 8186 RT

Hi all !
I'm facing to problems using MAX
1) We have a 1GB switch in order to connect my PXI 8186 RT controller and the host PC, when i use this switch, and i try to access the cards plugged in the PXI card with MAX, MAX crash! When i use a 10/100 Mb switch everything is ok, i can access to the card without any problem ! I know the PXI 8186 controller is only fitted with a 10/100 Mb Ethernet card, but i think that must not be a problem , am i wrong ? Is this problem a known bug ??
2) I want to install all software used by my application on the host computer (in order to be able to reinstall the PXI8186 controller in case of a crash)! To do that i installed MAX 3.1.1, then all drivers NI-DAQmx 7.4, CAN2.2, VISA 3.3, and finally Labview RT 7.1 Embedded on my gost computer! After that i tried to download these software in the controller using MAX! The problem is these software are on the host computer (there are visible in system/ software) , but when i try to install them on the controller, max don't see them ! I tried to uninstall and re-install all software but no changes ... Is there a procedure or an order to install all these software ??? What should i do ???

Hello,
The process to install software on RT controller is :
1/ Install LabVIEW Real-Time 7.1 on your host computer
2/ Install all driveres you need on your RT target
3/ On your host machine, in MAX, go to "remote targets"
4/ Expand "Software"
5/ Clik on "Install" button.
This application note will give you more details about above process.
Our controller is only provided with 10/100 MB network device. You have to check if you can configure your 1GB switch into 10/100 MB.
Regards,
Isabelle
Ingénieur d'applications
National Instruments France

Problems Using BDC ME51 or BAPI_REQUISITION_CREATE

Hello,
Mi problem is the following, I need to load PRs, which items are category: Services with account assignments: Proyects, the problem i'm having is with price conditions of each component...
Using the Bapi, it simply doesn't give me the option of loading this information (Conditions).
Using the a BDC recording of txn me51, it goes wild showing again and again the dynpro of the Account aSSIGNMENT on the WBS associated to the Service. (SAPLKACB 0200)
Regards,
Benjamin

Check the below link:
BAPI_REQUISITION_CREATE
Prakash.

Problem in BDC and a customized screen being called by the SAP Trans

Hello good day,
I recently have a BDC program where I will have to call a SAP STANDARD program (VF02). However, a problem occured because of a customized program being used called by VF02. This would cause the entire BDC proccess to stop running.
My question is, how would be able to pass through this customized screen in VF02. I tried using SHDB but the BDC still wont go ahead with the process. I also used the program name as well as the screen number for it.
For the info of the screen. The source program was created by SE38 as an executable program (i.e. regular report program). That appears to be calling a screen if a certain condition passes. The screen functions as a pop only only used to notify the user. There is push button to continue to the SAP standard screens.
Hope I get help soon, thanks guys and good day.

put a logic like this User exit.
if sy-BINPT ne 'X'.( batch for BackGround Job)
message 'Sdn.SAP.Com'.
endif.
Regards
prabhu
Message was edited by: Prabhu Peram

Problem using JDBC and MS SQL Server

I got a problem trying to acces a database on my SQL Server. I created the table and my connection seems ok, but when i try to get some data from a table, this happens (see below).
java.sql.SQLException: [Microsoft][ODBC SQL Server Driver][SQL Server]Invalid object name 'players'.
at sun.jdbc.odbc.JdbcOdbc.createSQLException(JdbcOdbc.java:6879)
at sun.jdbc.odbc.JdbcOdbc.standardError(JdbcOdbc.java:7036)
at sun.jdbc.odbc.JdbcOdbc.SQLExecDirect(JdbcOdbc.java:3065)
at sun.jdbc.odbc.JdbcOdbcStatement.execute (JdbcOdbcStatement.java:338)
at sun.jdbc.odbc.JdbcOdbcStatement.executeQuery(JdbcOdbcStatement.java:253)
at Team.<init>(Team.java:18)
at Simulator.main(Simulator.java:5)
here's the code that i used to access the db with:
Class.forName("sun.jdbc.odbc.JdbcOdbcDriver");
Connection conn = DriverManager.getConnection
("jdbc:odbc:BRAIN2WINXPPJH","sm","sm");
Statement stm = conn.createStatement();
String query = "Select * from players where teamnr = " + nr +
" order by squadnr";
ResultSet rs = stm.executeQuery(query);
any help is more than welcome
tnx
Jo

If all the tables are in the same database, you can use
Connection.setCatalog("db-name")
to change the current database. But the suggestion to include the database name in the SQL statement is safer, because you do not need to worry about changing the database at the right point.
Thomas

PROBLEMS USING WEBUTIL_FILE_TRANSFER AND CLIENT_HOST COMMAND TO VIEW BLOB

I am using webutil_file_transfer.DB_To_Client_with_progress to retrieve a blob
from the database (that was saved in a word format) into a .doc file on the
client.
1st problem: The document that is being saved is not readable. It has all
kind of special characters in it.
Then using the client_host command, I open the document into ms word.
2nd problem: The document opens fine (again not readable in garbage
characters) but when I close the document, notepad pops up blank with nothing
in it.
Below is a sample of the code.:
PROCEDURE DOWNLOAD_DB IS
l_success boolean;
username varchar2(30);
document_name varchar2(20);
wordexe varchar2(20);
BEGIN
     username := webutil_clientinfo.get_user_name;
     document_name := 'wo'||:wod.fk_work_order_number||'.doc';
     :locals.file_name := 'Documents and Settings\'||username||'\my documents\'||document_name;
l_success := webutil_file_transfer.DB_To_Client_with_progress
(clientFile => 'C:\'||:locals.file_name
,tableName => 'DEFECTS'
,columnName => 'DESCRIPTION'
,whereClause => 'FK_WORK_ORDER_DATE_ID = '||:DEF.FK_WO_DATE_ID
,progressTitle => 'Defect document being opened'
,progressSubTitle=> 'Please wait'
if l_success
then
CLIENT_HOST('cmd /c start /wait winword.exe C:\'||'"'||:locals.file_name||'"');
else
message('File download from Database failed');
end if;
exception
     when others
     then
     message('File download failed: '||sqlerrm);
END;
Any ideas?

Hello,
I use this kind of code with 9i and 10g and it works fine with word, excel and any of windows files.
I think that, if the file is not correct, the cause may be the first save of the file in the database.
Which instruction did you use to save first the doc in the BLOB column ?
Francois

Mapping Problem using hibernate and annotations

Hi,
i am German student and new to hibernate. I established a mn conetction between to entities using hbm.xml mapping files. now I try to create the same connection applying annotations. unfortunately it does not work and I do not now why. First my error message followed by my classes and xml's:
Initial SessionFactory creation failed.org.hibernate.AnnotationException: mappedBy reference an unknown target entity property: domain.Termin.person in domain.Person.termine
Exception in thread "main" java.lang.ExceptionInInitializerError
at services.HibernateUtil.sessionFactory(HibernateUtil.java:39)
at services.HibernateUtil.getSessionFactory(HibernateUtil.java:20)
at test.Test.main(Test.java:20)
Caused by: org.hibernate.AnnotationException: mappedBy reference an unknown target entity property: domain.Termin.person in domain.Person.termine
at org.hibernate.cfg.annotations.CollectionBinder.bindStarToManySecondPass(CollectionBinder.java:552)
at org.hibernate.cfg.annotations.CollectionBinder$1.secondPass(CollectionBinder.java:517)
at org.hibernate.cfg.CollectionSecondPass.doSecondPass(CollectionSecondPass.java:43)
at org.hibernate.cfg.Configuration.secondPassCompile(Configuration.java:1130)
at org.hibernate.cfg.AnnotationConfiguration.secondPassCompile(AnnotationConfiguration.java:316)
at org.hibernate.cfg.Configuration.buildSessionFactory(Configuration.java:1286)
at org.hibernate.cfg.AnnotationConfiguration.buildSessionFactory(AnnotationConfiguration.java:915)
at services.HibernateUtil.sessionFactory(HibernateUtil.java:36)
... 2 more
package domain;
import java.util.LinkedList;
import java.util.List;
import domain.Termin;
import javax.persistence.Entity;
import javax.persistence.GeneratedValue;
import javax.persistence.GenerationType;
import javax.persistence.Id;
import javax.persistence.ManyToMany;
import javax.persistence.Table;
@Entity
@Table(name = "PERSON")
public class Person {
   private long id;
   private String vorname;
   private String nachname;
   private List<Termin> termine=new LinkedList<Termin>();
   public Person() {
   @Id @GeneratedValue(strategy=GenerationType.AUTO)
   public long getId() {
      return id;
   public void setId(long id) {
      this.id = id;
   public String getNachname() {
      return nachname;
   public void setNachname(String nachname) {
      this.nachname = nachname;
   public String getVorname() {
      return vorname;
   public void setVorname(String vorname) {
      this.vorname = vorname;
   public void addTermin(Termin termin){
      termine.add(termin);
   @ManyToMany(mappedBy="person")
   public List<Termin> getTermine() {
      return termine;
   public void setTermine(List<Termin> termine) {
      this.termine = termine;
package domain;
import java.util.ArrayList;
import java.util.List;
import javax.persistence.Entity;
import javax.persistence.GeneratedValue;
import javax.persistence.GenerationType;
import javax.persistence.Id;
import javax.persistence.ManyToMany;
import javax.persistence.Table;
@Entity
@Table(name = "TERMIN")
public class Termin {
   private long id;
   private String titel;
   private Person eigentuemer;
   private List<Person> teilnehmer= new ArrayList<Person>();
   public void addTeilnehmer(Person person){
      teilnehmer.add(person);
   @ManyToMany
   public List<Person> getTeilnehmer() {
      return teilnehmer;
   public void setTeilnehmer(List<Person> teilnehmer) {
      this.teilnehmer = teilnehmer;
   public Termin() {
   @Id @GeneratedValue(strategy=GenerationType.AUTO)
   public long getId() {
      return id;
   public void setId(long id) {
      this.id = id;
   public String getTitel() {
      return titel;
   public void setTitel(String titel) {
      this.titel = titel;
   public Person getEigentuemer() {
      return eigentuemer;
   public void setEigentuemer(Person eigentuemer) {
      this.eigentuemer = eigentuemer;
package test;
import org.hibernate.Session;
import org.hibernate.SessionFactory;
import org.hibernate.Transaction;
import org.hibernate.cfg.Configuration;
import org.hibernate.tool.hbm2ddl.SchemaExport;
import services.HibernateUtil;
import domain.Person;
import domain.Termin;
public class Test {
   public static void main(String[] args) {
      Session session = null;
      HibernateUtil.setRecreateDB(true);
      session = HibernateUtil.getSessionFactory().getCurrentSession();
      /*      Person person1 =new Person();
      person1.setNachname("P1");
      Transaction transaction = session.beginTransaction();
      session.save(person1);
      transaction.commit();
      Person person2 =new Person();
      person2.setNachname("P2");
      session = HibernateUtil.getSessionFactory().getCurrentSession();
      transaction = session.beginTransaction();
      session.save(person2);
      transaction.commit();
      Termin termin1 =new Termin();
      termin1.setTitel("T1");
      termin1.setEigentuemer(person1);
      termin1.addTeilnehmer(person1);
      termin1.addTeilnehmer(person2);
      session = HibernateUtil.getSessionFactory().getCurrentSession();
      transaction = session.beginTransaction();
      session.save(termin1);
      transaction.commit();
   Termin termin2 =new Termin();
      termin2.setTitel("t2");
      termin2.setEigentuemer(person1);
      termin2.addTeilnehmer(person1);
      termin2.addTeilnehmer(person2);
      transaction = session.beginTransaction();
      session.save(termin2);
      transaction.commit();
      session.close();
package services;
import org.hibernate.SessionFactory;
import org.hibernate.cfg.AnnotationConfiguration;
import domain.Person;
import domain.Termin;
public class HibernateUtil {
   private static boolean recreateDB = false;
   public static void setRecreateDB(boolean recreateDB) {
      HibernateUtil.recreateDB = recreateDB;
   public static SessionFactory getSessionFactory() {
      if (sessionFactory == null) {
         sessionFactory = sessionFactory("hibernate.cfg.xml");
      return sessionFactory;
   private static SessionFactory sessionFactory = null;
   private static SessionFactory sessionFactory(String configurationFileName) {
      try {
         AnnotationConfiguration annotationConfiguration =
            new AnnotationConfiguration()
            .addAnnotatedClass(Person.class)
            .addAnnotatedClass(Termin.class);
         if (recreateDB) annotationConfiguration.setProperty("hibernate.hbm2ddl.auto", "create");
         annotationConfiguration.configure();
         return annotationConfiguration.buildSessionFactory();
      } catch (Throwable ex){
         System.err.println("Initial SessionFactory creation failed." + ex);
         throw new ExceptionInInitializerError(ex);
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE hibernate-configuration
    PUBLIC "-//Hibernate/Hibernate Configuration DTD//EN"
    "http://hibernate.sourceforge.net/hibernate-configuration-3.0.dtd">
<hibernate-configuration>
    <session-factory >
      <property name="hibernate.connection.driver_class">org.gjt.mm.mysql.Driver</property>
        <property name="hibernate.connection.password">application</property>
        <property name="hibernate.connection.url">jdbc:mysql://localhost/test</property>
        <property name="hibernate.connection.username">application</property>
        <property name="hibernate.dialect">org.hibernate.dialect.MySQLDialect</property>
       <property name="current_session_context_class">thread</property>
      <property name="hibernate.show_sql">true</property>
    </session-factory>
</hibernate-configuration>

The error message is pretty much telling you the problem:
mappedBy reference an unknown target entity property: domain.Termin.person in domain.Person.termine
This tells you that there's a mappedBy setting on the Person class's termine property annotation, and that the property it's referring to (the person property of the Termin class) doesn't exist.
@ManyToMany(mappedBy="person")
public List<Termin> getTermine() {
   return termine;
}If we have a look at the Termin class, indeed it has the following properties:
id
teilnehmer
titel
eigentuemerAnd no person property. Remember, a property is defined by the existence of a get/set pair for the name. It's unrelated to the types returned by them and the private variables implementing them.
mappedBy has the equivalent effect to the inverse property in a hbm.xml mapping file - it defines which entity's property will cause the foreign key value to be updated (persisting the relationship to the database). From the context of your code, I'm guessing you really want the annotation to read:
@ManyToMany(mappedBy="teilnehmer")

Character encoding problem using XSLT and Tomcat on Linux

Hi,
I have an application running on a Tomcat 4.1.18 application server that applies XSLT transformations to DOM objects using standard calls to javax.xml.transform API. The JDK is J2SE 1.4.1_01.
It is all OK while running on a development enviroment (which is Windows NT 4.0 work station), but when I put it in the production enviroment (which is a red hat linux 8.0), it comes up with some kind of encoding problem: the extended characters (in spanish) are not shown as they should.
The XSL stylesheets are using the ISO-8859-1 encoding, but I have also tried with UTF-8.
These stylesheets are dynamicly generated from a JSP:
// opens a connection to a JSP that generates the XSL
URLConnection urlConn = (new URL( xxxxxx )).openConnection();
Reader readerJsp = new BufferedReader(new InputStreamReader( urlConn.getInputStream() ));
// Gets the object that represents the XSL
Templates translet = tFactory.newTemplates( new StreamSource( readerJsp ));
Transformer transformer = translet.newTransformer();
// applies transformations
// the output is sent to the HttpServletResponse's outputStream object
transformer.transform(myDOMObject, new StreamResult(response.getOutputStream()) );Any help would be appreciated.

Probably you need to set your LANG OS especific environment variable to spanish.
Try adding this line:
export LANG=es_ES
to your tomcat/bin/catalina.sh, and restart tomcat.
It should look like this:
# OS specific support. $var _must_ be set to either true or false.
cygwin=false
case "`uname`" in
CYGWIN*) cygwin=true;;
esac
export LANG=es_ES
# resolve links - $0 may be a softlink
PRG="$0"
.(BTW, keep using ISO-8859-1 encoding for your XSL)
HTH
Un Saludo!

Serious resource utlization problem using SOAP and 8.0.2

Hi,
I'm running into a fairly serious issues whilst trying to download large attachments from a Groupwise 8.0.2 server using SOAP.
Basically, the problem is as follows: at a client site, there's an inbox with an email that has an attached avi file of about 330mb. Our SOAP client tries to retrieve this file in chunks of 1mb in size (if you try to set the size of the chunks any larger, Groupwise will give you a 59922 error, as 1mb is apparently some sort of hard coded limit). So that attachment corresponds to roughly 330 AttachmentRequestMessage's and their responses, each with a 1mb payload.
When we run our client, the cpu utilization of the Groupwise server takes about a dozen seconds to climb to 100%, every single time we try. Since 100% cpu utilization renders all GW clients more or less inoperable (or at least extremely slow) this is a non-starter.
We changed our client to wait 2 secs. in between requesting each chunk of the attachment. This seemed to help initially, but by the time we had transferred 100mb of the file, cpu utilization climbed to 95% and stayed there until the file transfer was completed.
This is a little disconcerting. We're requesting a megabyte of data from an attachment every 2 seconds, and Groupwise seems to have all sorts of problems in trying to service these requests. As you can imagine, client performance isn't that great at 95% server utilization either. I suppose we could increase the pause to something like 5 seconds, but it seems ridiculous that such simple requests are generating this amount of load.
So my question is 2-fold:
1. can I increase the maximum size of an attachment part chunk? 1mb is not very much at all, seeing as our server is on the same gigabit lan segment as the GW server
2. why does cpu utilization shoot up to 95% while servicing what seem to be fairly simple requests, and how do I avoid this by some means other than increasing my pause time?
Thanks in advance,
Maarten
PS I spent some time looking for the 59922 error code in the docs, but turned up nothing

The 59922 error is just a warning.
You should use the HTTP GET construct to get large
attachments. It is orders of magnitude faster than
using getAttachmentRequest.
>>> On Friday, August 20, 2010 at 11:06 AM,
mdirkse<[email protected]> wrote:
> Hmm, could you perhaps point me to where in the docs?
>
> Because as far as I know you have to download an attachment using a
> getAttachmentRequest ('Novell Doc: NDK: GroupWise Web Services ‑
> getAttachmentRequest'
>
(http://developer.novell.com/document...a/b7m3i5b.html
> )),
> which takes a length parameter, and if you set that to > 1mb, Groupwise
> returns a 59922 error which, I'm pretty sure, *isn't* in the docs.
>
> Ray;2013273 Wrote:
>> You can download the complete attachment
>> in 1 go over HTTP using a SOAP login session.
>>
>> It's all in the docs...
>>
>> Am 19.08.2010 16:06, schrieb mdirkse:
>> >
>> > Hi,
>> > I'm running into a fairly serious issues whilst trying to download
>> > large attachments from a Groupwise 8.0.2 server using SOAP.
>> >
>> > Basically, the problem is as follows: at a client site, there's an
>> > inbox with an email that has an attached avi file of about 330mb.
>> Our
>> > SOAP client tries to retrieve this file in chunks of 1mb in size (if
>> you
>> > try to set the size of the chunks any larger, Groupwise will give you
>> a
>> > 59922 error, as 1mb is apparently some sort of hard coded limit). So
>> > that attachment corresponds to roughly 330
>> AttachmentRequestMessage's
>> > and their responses, each with a 1mb payload.
>> >
>> > When we run our client, the cpu utilization of the Groupwise server
>> > takes about a dozen seconds to climb to 100%, every single time we
>> try.
>> > Since 100% cpu utilization renders all GW clients more or less
>> > inoperable (or at least extremely slow) this is a non‑starter.
>> >
>> > We changed our client to wait 2 secs. in between requesting each
>> chunk
>> > of the attachment. This seemed to help initially, but by the time we
>> had
>> > transferred 100mb of the file, cpu utilization climbed to 95% and
>> stayed
>> > there until the file transfer was completed.
>> >
>> > This is a little disconcerting. We're requesting a megabyte of data
>> > from an attachment every 2 seconds, and Groupwise seems to have all
>> > sorts of problems in trying to service these requests. As you can
>> > imagine, client performance isn't that great at 95% server
>> utilization
>> > either. I suppose we could increase the pause to something like 5
>> > seconds, but it seems ridiculous that such simple requests are
>> > generating this amount of load.
>> >
>> > So my question is 2‑fold:
>> > 1. can I increase the maximum size of an attachment part chunk? 1mb
>> is
>> > not very much at all, seeing as our server is on the same gigabit
>> lan
>> > segment as the GW server
>> > 2. why does cpu utilization shoot up to 95% while servicing what
>> seem
>> > to be fairly simple requests, and how do I avoid this by some means
>> > other than increasing my pause time?
>> >
>> > Thanks in advance,
>> > Maarten
>> >
>> > PS I spent some time looking for the 59922 error code in the docs,
>> but
>> > turned up nothing
>> >
>> >

Problems using BDC and/or HR_INFOTYPE_OPERATION

Similar Messages

Maybe you are looking for