Can I specify instance on R/3 side to avoid memory buffer issue

I have one datasource (out of many that work A-OK every day) that likes to throw a TSV_TNEW_PAGE_ALLOC_FAILED short dump intermittently on the source system extract. The short dump occurs in the background job that was created by running the infopackage in BW. The short dump occurs on an internal table as the extractor is building the packets to be sent to BW (at least, that is how it looks).
We have 16 instances on our source system. The short dump has happened on several different ones -- it seems to be whereever the batch job happens to be scheduled.
That brings me to my question. Some of the 16 instances have higher heap total settings than others. That is because some are configured more for batch work than others. On the R/3 side, of course, I can assign batch jobs to instances as I create them. But what about batch jobs that are spawned from the BW-side.
Is there a way to control the instance where the extract batch job runs? It seems you can set an app instance for all extracts if you want -- but, of course, that would not be good for me. In general, I want to use batch processes across many instances.
I want to control where this one specific datasource runs its extraction, so I can point it to a high-memory instance.
Thanks for any help.

Roger Burrows
while scheduling the Job it self you direct specific long running job/Large Volume of Records Job to specific server which has more capicity so that this particular job will run on that specific server only.
Thanks
sat

Similar Messages

How can I specify that a border should have a certain length on a side?

how to customize a border so that length can be specified?
(I know that you can use createMatteBorder() to specify which sides should have a border. But this is different - how do you specify the length on a side).
for example, here is one side: --- is the side and ==== is the border
===========
how do I implement it?
I am stumped.
thanks,
Anil

BorderFactory does dispense some static instances depending on the type of
border being created. But you'll need to subclass MatteBorder to customize it;
I don't think you'll be able to do it via BorderFactory. You can either override
the paintBorder method to not paint the entire top part of the border or paint
over it; the latter approach is taken below:
import java.awt.*;
import javax.swing.border.MatteBorder;
import javax.swing.*;
public class PartiallyPaintedBorder extends MatteBorder {
     private final static int MARGIN = 15;
     public PartiallyPaintedBorder( int top, int left, int bottom, int right, Icon tileIcon ) {
          super( top, left, bottom, right, tileIcon );
     * Paints the matte border.
    public void paintBorder(Component c, Graphics g, int x, int y, int width, int height) {
         super.paintBorder( c, g, x, y, width, height );
        Insets insets = getBorderInsets(c);
        Color oldColor = g.getColor();
        g.translate(x, y);
        color = c.getBackground();
        if ( color == null )
             color = Color.gray;
        // Erase part of top matte edge
        g.setColor(color);
        g.fillRect( width / 2, 0, width, insets.top);
        g.translate(-x, -y);
        g.setColor(oldColor);
     // Example use on a JPanel containing a JButton.
     public static void main( String[] args ) {
          JPanel p = new JPanel();
          p.setBorder( new PartiallyPaintedBorder( MARGIN, MARGIN, MARGIN, MARGIN, UIManager.getIcon( "FileView.computerIcon" ) ) );
          p.add( new javax.swing.JButton( "Press me" ) );
          JFrame f = new JFrame( "Tester" );
          f.setDefaultCloseOperation( JFrame.EXIT_ON_CLOSE );
          f.getContentPane().add( p );
          f.pack();
          f.setVisible( true );
}

How can we specify the order of the predicates execution?

I am going to write the following query
select answer, answer_id from surveys s join answers a on s.survey_id = a.survey_seq_id
where question_uid = 206400374 and insertdate = to_date('31/12/2012') and answer > 0;However, when I look at the execution plan, I see that the last predicate (to_number(answer) > 0) has been executed the first. Henceforth, it checks many rows first. Normally, 75 rows belong to 31/12/2012 as you see from the following. Can I specify the execution order?
select count(*) from surveys s join answers a on s.survey_id = a.survey_seq_id
where question_uid = 206400374 and insertdate = to_date('31/12/2012');
COUNT(*)
75If so, how? Because, the type of answer is varchar2 and some of answers contain text characters such as 'Yes' or 'No'. Therefore, before 31/12/2012 some answers contain 'Yes' or 'No'. However in 31/12/2012 all answers are numeric.
Plan hash value: 3217836037
| Id | Operation          | Name    | Rows | Bytes | Cost (%CPU)| Time     |
|   0 | SELECT STATEMENT   |         |     2 |    68 | 9401   (1)| 00:01:53 |
|* 1 | HASH JOIN         |         |     2 |    68 | 9401   (1)| 00:01:53 |
|* 2 |   TABLE ACCESS FULL| SURVEYS |     2 |    26 |   239   (1)| 00:00:03 |
|* 3 |   TABLE ACCESS FULL| ANSWERS |   337 | 7077 | 9162   (1)| 00:01:50 |
Predicate Information (identified by operation id):
   1 - access("S"."SURVEY_ID"="A"."SURVEY_SEQ_ID")
   2 - filter("S"."INSERTDATE"=TO_DATE(' 2012-12-31 00:00:00',
              'syyyy-mm-dd hh24:mi:ss'))
   3 - filter("A"."QUESTION_UID"=206400374 AND
              TO_NUMBER("A"."ANSWER")>0)
select distinct answer from surveys s join answers a on s.survey_id = a.survey_seq_id
where question_uid = 206400374 and insertdate = to_date('31/12/2012');
ANSWER
1
3
0
2And, also I can execute the following query without any error
select to_number(answer) from surveys s join answers a on s.survey_id = a.survey_seq_id
where question_uid = 206400374 and insertdate = to_date('31/12/2012');

In answer to your original question:
970992 wrote:
However, when I look at the execution plan, I see that the last predicate (to_number(answer) > 0) has been executed the first. Henceforth, it checks many rows first. Normally, 75 rows belong to 31/12/2012 as you see from the following. Can I specify the execution order?According to the execution plan, it will do a full scan of surveys using the predicate on insertdate to build the (presumably in-memory) hash table (hash based on survey_id) to do the joins (Step 2). Then, it does a full scan of the answers table using the question_uid and answer predicates (Step 3). For each row it finds that matches those predicate, it will prpobe the hash table created in step 2 using the hashed value of survey_seq_id. So, it is doing the insertdate predicate first.
In answer to your last post
970992 wrote:
>
First of all i would get rid of the implizit type conversion:
TO_NUMBER("A"."ANSWER")>0)it is not implicit conversion, I reckon it is explicit type conversion, isnt it?
No, it is an implicit type conversion. Your code says answer > 0, since the answer column is a varchar2 column, Oracle implicitly converts the answer column to a number to compare against the number on the right side of the comparison. Note that if something like 'A' ever becomes a valid answer, then this query will fail with ORA-01722: invalid number.
>
>
Obviously "A"."ANSWER" is not a number colmun, problably varchar, so use something like
A.ANSWER != "0"
or
A.ANSWER > "0"Yes answer column is varchar2 but can you type A.ANSWER > "0" as a predicate? I mean, you can not varchar > varchar, can you?
Of course you can use inequality predicates on a varcahr column. Is the string A greater than the string B?
Based on the explain plan, your statistics might be a little off, not hugely so. The esitmates are at least in the right order of magnitude based on what you have posted so far.
What indexes, if any, are available on the two tables?
John

No CENTRAL nor DIALOG instance known for system SID

Hi,
We are in the process of setting up our Solution Manager 7.1. Currently we are in the step 'Managed System Configuration'.
We are stuck on the step 7 - "Create Users". There are 4 users to be created which are already there in the systems(Created Manually).
The users are
SAPSUPPORT and SMDAGENT_XXX for ABAP and Java both.
In the ABAP view it gives the error as "No CENTRAL nor DIALOG instance known for system SID".
In the Java view it gives the error as "User Status Cannot be checked".
We are on the below Support Pack status:
SAP_ABA 702 0008 SAPKA70208
SAP_BASIS 702 0008 SAPKB70208
PI_BASIS 702 0008 SAPK-70208INPIBASIS
ST-PI 2008_1_700 0004 SAPKITLRD4
SAP_BS_FND 702 0006 SAPK-70206INSAPBSFND
SAP_BW 702 0008 SAPKW70208
SAP_AP 700 0024 SAPKNA7024
WEBCUIF 701 0005 SAPK-70105INWEBCUIF
BBPCRM 701 0005 SAPKU70105
BI_CONT 706 0003 SAPK-70603INBICONT
CPRXRPM 500_702 0006 SAPK-50006INCPRXRPM
ST 710 0003 SAPKITL703
ST-BCO 710 0001 SAPK-71001INSTBCO
SOCO 101 0000 -
ST-A/PI 01N_700SOL 0000 -
ST-ICO 150_700 0030 SAPK-1507UINSTPL
ST-SER 701_2010_1 0008 SAPKITLOS8
Please suggest a solution to this.
Thanks & Regards,
Ajitabh

Hello Ajitabh,
I'm sure that you will see this error if you expand the error entry:
SPML service failed to process searchRequest
1. If you followed the advices from note 1616058, disabling SPML:
When the SPML is desactivated, the status of users can't be checked.
We are working to provide a note to solve this issue. Note number is 1647267, it is not release yet, but the solution is:
"Enable to flag the user creation to 'manually performed' in solman_setup".
2. If you didn't disable SPML:
Please refer to the steps in this SAP Note : 1647157 which will help you address the issue.
Please let us know the outcome, thanks.
Best regards,
Guilherme

Error--Instance DVEBMGS00 for system SID not configured

i am changing the instance profile in Basic Mainenance.
When i increase teh Dialog instance and click on copy, i am getting the message
Instance DVEBMGS00 for system <SID> not configured
But this system is running for the past one year.
Please let me know what i can do in this regard

Hello Balaji,
Your question is not very clear. Could not understand what you mean by "When i increase teh Dialog instance and click on copy,". Please explain all the steps you are taking.
Regards.
Ruchit.

I want to connect to remote databases which can be specified by URL

Hi,
i'm tinu
I want to connect to remote databases which can be specified by URL
the database is ORACLE 9i
pls help me, how to connect to it
i have the ip address,port address,sid,username and password of the database
is there any difference in the actual code of database connection
plss help

Hi,
There is a particular example with MS SQL 2000 in thread http://forum.java.sun.com/thread.jspa?threadID=608314
In the given example you just need to change the database URL and the JDBC driver. Just examine the code a little bit.
Also you may wish to visit the SUN's JDBC tutorial on http://java.sun.com/docs/books/tutorial/jdbc/
Ferad Zyulkyarov

How can we specify date time under History Tab of Imaging

Hi
I am using Oracle Webcenter Imaging 11.1.1.5, when i search for a document and view it, under History tab i have following information available.
Date and User
The date specifies the date when it was modified and user specifies UserName who modified it.
I want date time over there instead of date.
Can someone help me how to achieve this?
Regards
ACM

In answer to your original question:
970992 wrote:
However, when I look at the execution plan, I see that the last predicate (to_number(answer) > 0) has been executed the first. Henceforth, it checks many rows first. Normally, 75 rows belong to 31/12/2012 as you see from the following. Can I specify the execution order?According to the execution plan, it will do a full scan of surveys using the predicate on insertdate to build the (presumably in-memory) hash table (hash based on survey_id) to do the joins (Step 2). Then, it does a full scan of the answers table using the question_uid and answer predicates (Step 3). For each row it finds that matches those predicate, it will prpobe the hash table created in step 2 using the hashed value of survey_seq_id. So, it is doing the insertdate predicate first.
In answer to your last post
970992 wrote:
>
First of all i would get rid of the implizit type conversion:
TO_NUMBER("A"."ANSWER")>0)it is not implicit conversion, I reckon it is explicit type conversion, isnt it?
No, it is an implicit type conversion. Your code says answer > 0, since the answer column is a varchar2 column, Oracle implicitly converts the answer column to a number to compare against the number on the right side of the comparison. Note that if something like 'A' ever becomes a valid answer, then this query will fail with ORA-01722: invalid number.
>
>
Obviously "A"."ANSWER" is not a number colmun, problably varchar, so use something like
A.ANSWER != "0"
or
A.ANSWER > "0"Yes answer column is varchar2 but can you type A.ANSWER > "0" as a predicate? I mean, you can not varchar > varchar, can you?
Of course you can use inequality predicates on a varcahr column. Is the string A greater than the string B?
Based on the explain plan, your statistics might be a little off, not hugely so. The esitmates are at least in the right order of magnitude based on what you have posted so far.
What indexes, if any, are available on the two tables?
John

I have an iPhone 4 and my lock button is jammed and I also can not touch anything on the left side of my screen well as any pop ups from the phone itself, so i can not get past the set up stage in reset process...how can i fix it/can it be fixed?

I have an iPhone 4 and my lock button is jammed and I also can not touch anything on the left side of my screen well as any pop ups from the phone itself, so i can not get past the set up stage in reset process...how can i fix it/can it be fixed?

Make an appointment at the genius bar and get the phone replaced.

Mozilla crashed yesterday and now bookmarks are not showing along the left side of page as they did before. They are on the menu bar but hubby does not like this location. Anyway I can get it back to the left side of the page?

This spring we changed from IE to Mozilla after upgrading to Win 7. Everything has been fine with Firefox until yesterday when it crashed when I attempted to open an email. Since then the bookmarks (or favorites as my husband calls them) have not been in their usual place which was along the left side of the page. They are located in the Bookmarks bar at the top of the page but husband prefers the old IE way which we had until yesterday PM. Is there any way that we can get them back to the left side of the page other than going back to IE? Now that I have gotten used to Firefox I like it just fine.

*You can open the Bookmarks in the sidebar via View > Sidebar > Bookmarks (Ctrl+B)
*You can open the History in the sidebar via View > Sidebar > History (Ctrl+H)
Press F10 or tap the Alt key to bring up the "Menu Bar" temporarily.
You can find toolbar buttons to open the bookmarks (star) and the history (clock) in the sidebar in the toolbar palette and drag them on a toolbar.<br />
Firefox 4 versions and later have two bookmark buttons with a star in the Customize window.<br />
One star button has a drop marker that open a Bookmark menu.<br />
The other star button without the drop marker opens the bookmarks in the sidebar.<br />
*http://kb.mozillazine.org/Toolbar_customization
*https://support.mozilla.com/kb/Back+and+forward+or+other+toolbar+items+are+missing

How can we specify the Mailbox for Notes?

In Mail.app, how can we specify the Mailbox for Notes (where the new notes will be resided)? It is now 'On My Mac'. I wish to change to my IMAP Mailbox so it sync with my iPhone by just checking email (not by syncing on iTunes).
Thanks in advance.
Message was edited by: Ekapon

You could use the DecimalFormat or NumberFormat to do the job.
double number 1234.567;
DecimalForamt df = new DecimalFormat("###.##");
String s = df.format(number);or
double number 1234.567;
NumberFormat nf = NumberFormat.getNumberInstance();
nf.setMaximumFractionDigits(2);
/* if needed.
   nf.setMaximumIntegerDigits(value);
   nf.setMinimumFractionDigits(value);
   nf.setMinimumIntegerDigits(value); */
String s = nf.format(number);I didn't test the codes but they should work. For more information, please consult the documentations.

How Can i specify multiple server names in rwservlet.properties file?

How Can i specify multiple server names in rwservlet.properties file without clustering?
I am using oracle 10g Application server. we have 3 servers Repsvr1, RepSvr2 and RepSvr3. Now i need to configure rwservlet.properties file to point to these servers based on any running report. i got 3 keymap files with reports info.
Sample entry in the key map file is:
key1: server=Repsvr1 userid=xxx/yyy@dbname report=D:\Web\path1\path2\reports\Report1.rdf destype=cache desformat=PDF %*
key2: server=Repsvr2 userid=xxx/yyy@dbname report=D:\Web\path1\path3\reports\Report2.rdf destype=cache desformat=PDF %*
rwservlet.properties file letting me to enter only one servername. Even though i merged all 3 keymap files into 1, still i have the server name issue. If i leave the server to the default name still i am getting the below error.
REP-51002: Bind to Reports Server Repsvr1 failed. However, i know the default rep_<servername> would be used incase we dont have SERVER=<value> parameter in the rwservlet.properties file.
If i specify the servername in the rwservlet.properties file then only Repsvr1 reports are working fine and other 2 server reports are giving the same error like
REP-51002: Bind to Reports Server <<Server Name>> failed.
how can i configure the info which will work all 3 reports. 2 Port servers are invoking using oracle forms and report server is invoking using ASP pages.
If i specify Server name & Key map file in rwservlet.properties one at a time, all the reports are working without any error, whenever i am trying to integrate all 3 to workable i am getting binding error. if i exclude the server from rwservlet.properties still i am getting the same error.

My RELOAD_KEYMAP setting is YES only.As i said If i specify Server name & Key map file in rwservlet.properties one at a time, all the reports are working without any error.
keymap file entries
key1: server=Repsvr1 userid=xxx/yyy@dbname report=D:\Web\path1\path2\reports\Report1.rdf destype=cache desformat=PDF %*
key2: server=Repsvr2 userid=xxx/yyy@dbname report=D:\Web\path1\path3\reports\Report2.rdf destype=cache desformat=PDF %*
If i use http://server.domain:port/reports/rwservlet? cmdkey = key1 should bring the report from Repsvr1 and http://server.domain:port/reports/rwservlet? cmdkey = key2 should bring the report from Repsvr2, but i am getting an error from Repsvr2 saying that REP-51002: Bind to Reports Server repsvr2 failed.
Only Servername Repsvr1 is in rwservlet.properties file. Now what is the best option to by pass the server from rwservlet.properties file and should be from keymap file. if i comment server name in rwservlet.properties file still i am getting REP-51002: Bind to Reports Server <<Server Name>> failed error for both keys.

Can we delete a single row in SID table?

I am having a problem with conversion exit in SID table.
These are the error messages.
Value in SID table is TPV; correct value is TPV; SID in SID table is 875
Message no. RSRV200
Diagnosis
The following data record either has an incorrect internal format or the characteristic value that is in the correct format appears as a corrected value of another incorrect value:
·     Characteristic value: TPV
·     SID: 875
·     Correct characteristic value: (see below) TPV
·     SID after correction: 875
Value in SID table is TPV 2008; correct value is TPV; SID in SID table is 2887
Message no. RSRV200
Diagnosis
The following data record either has an incorrect internal format or the characteristic value that is in the correct format appears as a corrected value of another incorrect value:
·     Characteristic value: TPV 2008
·     SID: 2887
·     Correct characteristic value: (see below) TPV
·     SID after correction: 875
Now the row with SID 875 is causing the problem. Is it possible that I can delete only this row in the SID table.
Thanks for your help
Subra

Hi.....
Procedure :
RSA1>>>InfoObjects >>Right Click on InfoObject >>delete Master Data >> u can see the option of deleting SIDs............
Otherwise........u can delete value from a SID table......u can use tcode SE14...........to delete the entries.........
Check this......
deleting contents of SID table.
But still I will suggest u not to delete SID...........it may lead to inconsistency of data.........
Try to repair SId using the program : RSDMD_CHECKPRG_ALL or RSRV...........
Regards,
Debjani......

How can I specify the default tab in a CHM Output?

My company uses CHM-based help for some of its products. We build the CHM files from RoboHelp 9, and while these CHM files don't really have a full Index attached to them, the Index tab shows up in the output anyway. Unfortunately, we are suddenly seeing that when the CHM file is launched from our program, the Index tab is displayed by default. We'd rather have the Contents tab be the default look, especially considering that the Index does not exist.
I have poked around into how you can specify a default tab for a CHM file, and the only information I have found suggests that using a CHM file creates a file (HH.dat) that specifies which tab should be displayed on a user-by-user basis, and that the last tab displayed when you close the CHM should be the first one displayed when you re-open it. While this is true if you open the CHM independent from the product, when you launch it from our program, it's all Index, all the time.
So, my question is: How can I specify the default tab for a CHM file? Or, failing that, how can I excise the Index tab from my CHM output.

Hi there
This will be something up to your application developer to resolve. When s/he issues the call to open the CHM, there are parameters that may be used to always open with the desired tab "in front".
Point your developer to the link below and advise that s/he is most likely interested in the section titled: Programming Tips.
Click here to view
Cheers... Rick
Helpful and Handy Links
RoboHelp Wish Form/Bug Reporting Form
Begin learning RoboHelp HTML 7, 8 or 9 within the day!
Adobe Certified RoboHelp HTML Training
SorcerStone Blog
RoboHelp eBooks

Fusion 4.1.2. After install can no longer print on the Mac side.

Version:1.0 StartHTML:0000000167 EndHTML:0000003793 StartFragment:0000000475 EndFragment:0000003777
Greetings,
I'm using a MacBook Pro running OSX SL at the most current level.
I upgraded from Fusion 3 to Fusion 4 (uninstalled 3 first then installed 4) after which I did the update to Fusion 4.1.2 which appeared to run fine.
Being I'm new to the Mac I installed Fusion because I need Windows-XP.
Prior to the installation of Fusion I spent 2 weeks trying to get my printers working just on the Mac side without Fusion. I'm using an HP Deskjet 5850 (5800 series) printer which worked just fine. The printer is an IP printer at IP address 192.168.1.11 and can be pinged from both the Mac side and the Windows-XP side. Before Fusion there wasn't any issue printing on the Mac side.
After Fusion installation I didn't have any issues printing on the Windows side.
Now with Fusion installed I can still print on the Windows side but not on the Mac side.
All settings are as normal with printer sharing.
When trying to print on the Mac side I see the printer queue for the HP pop up and it says 'waiting to connect' to 192.168.1.11.
I tried a dozen different settings non of which worked.
I switched from a NAT connection to a Bridged connection for Windows but still the same result, no printing from the Mac side.
Is there a problem with this release of fusion or is it somewhere in my system settings?
Any help will be appreciated.
Thanks

Run through this list of fixes including #16 then stop and report.
Step by Step to fix your Mac

Using External_Stage can we specify a location other than staging dir

While using external stage mode for weblogic(I am trying for weblogic92)we must manually copy the deployment files(web app) to the staging directory of each target server before deployment.
Is there a option to copy the deployment files to another location other than Server's staging directory? To be more specific can I specify a location which is outside the server?
Secondly, with no-stage option of weblogic we can have a external location for the web application, but in case of the cluster,
this location has to be either shared or
the secondary node should have the same location(path to the web application) as that of the primary server.
Here is my config.xml entry of the primary node.
<app-deployment>
<name>TestWebApp</name>
<target>TestCluster</target>
<module-type>war</module-type>
<source-path>C:\TestWebApp\TestWebApp.war</source-path>
<security-dd-model>DDOnly</security-dd-model>
<staging-mode>nostage</staging-mode>
</app-deployment>
Here there is only one source path for the entire Cluster. So if both of the server nodes have a valid location(C:\TestWebApp\TestWebApp.war), then the deployment will be fine on a running server. But on the second server node, if the Web app does not exists on that location, deployment will fail.
When we share the location the concept of cluster is lost. If go with the second option then the user is restricted to have the same path on second node also.
Is there any way to specify differnt locations(of the same web app) for different servers in the Cluster?

A few points of clarity:
1) Session starts in consumer group LOGIN_GROUP. That does not have session queueing since queueing sessions at log on time results in the user experience of a hung session, which we cannot have. Whether service X is used to get it there or something besides services is fine; how is not a concern at this point.
2) Session moves to consumer group MAIN_GROUP after about 5 CPU seconds in the LOGIN_GROUP. We want to do this because the MAIN_GROUP has session queuing, which we want for most sessions after they have the unqueued login experience per point #1.
3) Session moves to consumer group SLOW_GROUP after about 60 CPU seconds in the MAIN_GROUP (or some number of seconds; exact numbers are not the main focus here). We want this because we want long-running queries to be downgraded. However, we don't want to leave the session there since the query is the thing that is running long, so we want the session to get back to MAIN_GROUP when it is done. However, if we use SWITCH_FOR_CALL when going to the SLOW_GROUP, it will switch back to the LOGIN_GROUP. Which is the trouble--LOGIN_GROUP is an unqueued group, so we don't get the desired session queueing.
4) The solution must be fully automated without any DBA intervation or application code. We cannot use appliation code as this solution is intended for BI users working in tools that have direct data access. Thus prevents the use of DBMS_SESSION as you stated in your posts. How would we do that switching in an automated fashion without application code?
5) We cannot put the DBMS_SESSION commands in a login trigger since that would switch groups to MAIN_GROUP and would again run the risk of queueing a session at login, which feels like a hung session to users and results in a bad user experience and help desk calls.
Thanks!

Can I specify instance on R/3 side to avoid memory buffer issue

Similar Messages

Maybe you are looking for