Problem converting static data load mapping to MOLAP

Hi
as a prototyping exercise I am converting some of our ROLAP dimensions and corresponding data load mappings (1 static data i.e. "-1" id to handle unknowns in fact data, and 1 for real data coming from a table) to MOLAP.
The dimension itself converts and deploys correctly and the real data mapping also redeploys and executes correctly.
HOWEVER
my static data mapping will not execute successfully.
The mapping uses constants (ID = -1, NAME 'UNKNOWN' etc), not all attributes are linked (this has been tried). My column WH_ID which was the ROLAP surrogate key gets converted to VARCHAR2 as expected. Mapping does deploy cleanly.
The error i get is below. I have been banging my head on this for a couple of days and tried searching the Net, Metalink to no avail. I'm hoping someone out there can help
LOAD_STATIC_D_TRADER_IU
Warning
ORA-20101: 15:48:04 ***Error Occured in BUILD_DRIVER: In __XML_SEQUENTIAL_LOADER: In __XML_LOAD_ATTRS: Error loading attributes for hierarchy, D_TRADER.AW$NONE.HIERARCHY, level D_TRADER.TRADER.LEVEL, mapping group D_TRADER.TRADER.MAPGROUP1.DIMENSIONMAPGROUP. In __XML_LOAD_ATTRS_ITEM: In ___XML_LOAD_TEMPPRG: The SQL IMPORT command cannot convert from the TEXT type to the DECIMAL type.
TRUNCATE_LOAD=false
AW Execution status: Success
15:48:00 Started Build(Refresh) of MARTS Analytic Workspace.
15:48:00 Attached AW MARTS in RW Mode.
15:48:01 Started Loading Dimensions.
15:48:01 Started Loading Dimension Members.
15:48:01 Started Loading Dimension Members for D_TRADER.DIMENSION (1 out of 1 Dimensions).
15:48:03 Finished Loading Members for D_TRADER.DIMENSION. Added: 1. No Longer Present: 885.
15:48:03 Finished Loading Dimension Members.
15:48:03 Started Loading Hierarchies.
15:48:03 Started Loading Hierarchies for D_TRADER.DIMENSION (1 out of 1 Dimensions).
15:48:03 Finished Loading Hierarchies for D_TRADER.DIMENSION. 1 hierarchy(s) STANDARD Processed.
15:48:03 Finished Loading Hierarchies.
15:48:03 Started Loading Attributes.
15:48:03 Started Loading Attributes for D_TRADER.DIMENSION (1 out of 1 Dimensions).
15:48:04 Failed to Build(Refresh) MARTS Analytic Workspace.
15:48:04 ***Error Occured in BUILD_DRIVER: In __XML_SEQUENTIAL_LOADER: In __XML_LOAD_ATTRS: Error loading attributes for hierarchy, D_TRADER.AW$NONE.HIERARCHY, level D_TRADER.TRADER.LEVEL, mapping group D_TRADER.TRADER.MAPGROUP1.DIMENSIONMAPGROUP. In __XML_LOAD_ATTRS_ITEM: In ___XML_LOAD_TEMPPRG: The SQL IMPORT command cannot convert from the TEXT type to the DECIMAL type.

Hi this looks like a bug in set based mode with using numeric dimension attributes and loading them from a constant. Row based mode is OK, which stages the data before loading the AW, but you probably don't want this.
A workaround is to add an expression operator in the map. You will have to add a link from a source table/constant into the expression operator to satisfy the map analyser. But then you can add expressions such as your numeric attributes in the expression operator's output group, define the values for each expression and map these expression outputs (not the numeric constants) into your dimension. Hopefully this makes sense.
Cheers
David

Similar Messages

ERPi Data load mapping Issue

Hi,
We are facing issue with ERPi data load mappings issue. Mapping file (txt file) has 36k records, whenever we are trying to load mappings, it's taking very long time, nearly 1 hour 30mins. but we want to reduce that time. is there any way to reduce data load mapping time??
Hyperion verion: 11.1.2.2.300
Please help, thanks in advance!!
Thanks.

Any one face the same kind of issue??

Problem with Master Data Load

Dear Experts,
If somebody can help me by the following case, please give me some solution. Iu2019m working in a project BI 7.0 were needed to delete master data for an InfoObject material. The way that I took for this was through tcode u201CS14u201D. After that, I have tried to load again the master data, but the process was broken and the load done to half data.
This it is the error:
Second attempt to write record 'YY99993' to /BIC/PYYYY00006 failed
Message no. RSDMD218
Diagnosis
During the master data update, the master data tables are read to determine which records of the data package that was passed have to be inserted, updated, or modified. Some records are inserted in the master data table by a concurrently running request between reading the tables at the start of package processing and the actual record insertion at the end of package processing.
The master data update tries to overwrite the records inserted by the concurrently running process, but the database record modification returns an unexpected error.
Procedure
u2022     Check if the values of the master data record with the key specified in this message are updated correctly.
u2022     Run the RSRV master data test "Time Overlaps of Load Requests" and enter the current request to analyze which requests are running concurrently and may have affected the master data update process.
u2022     Re-schedule the master data load process to avoid such situations in future.
u2022     Read SAP note 668466 to get more information about master data update scheduling.
Other hand, the SID table in the master data product is empty.
Thanks for you well!
Luis

Dear Daya,
Thank for your help, but I was applied your suggesting. I sent to OSS with the following details:
We are on BI 7.0 (system ID DXX)
While loading Master Data for infoobject XXXX00001 (main characteristic in our system u2013 like material) we are facing the following error:
Yellow warning u201CSecond attempt to write record u20182.347.263u2019 to BIC/ XXXX00001 was successfulu201D
We are loading the Master data from data source ZD_BW_XXXXXXX (from APO system) through the DTP ZD_BW_XXXXX / XXX130 -> XXXX00001
The Master Data tables (S, P, X) are not updated properly.
The following reparing actions have been taken so far:
1.     Delete all related transactional and master data, by checking all relation (tcode SLG1 à RSDMD, MD_DEL)
2.     Follow instructions from OSS 632931 (tcode RSRV)
3.     Run report RSDMD_CHECKPRG_ALL from tcode SE38 (using both check and repair options).
After deleting all data, the previous tests were ok, but once we load new master data, the same problem appears again, and the report RSDMD_CHECKPRG_ALL gives the following error.
u201CCharacteristic XXXX00001: error fund during this test.u201D
The RSRV check for u201CCompare sizes of P and X and/or Q and Y tables for characteristic XXXX00001u201D is shown below:
Characteristic XXXX00001: Table /BIC/ PXXXX00001, /BIC/ XXXXX00001 are not consistent 351.196 derivation.
It seems that our problem is described in OSS 1143433 (SP13), even if we already are in SP16.
Could somebody please help us, and let us know how to solve the problem?
Thank for all,
Luis

Problem with the data loads

Hi,
We have a daily data load to ODS and then to the CUBE. Yesterday what happened is that the load to ODS has been taken place for 140 times and the data is not activated in the ODS. We have deleted the requests from the ODS and did the manual load to ODS and activated successfully. My question is that what causes this job to repeat such many number of times. The jobs log is as follows.
Job started
Step 001 started (program ZBIXX_FDA_START_PROCESS_CHAIN, variant ZCUST_DAILY, user ID BWBATCH)
Chain Is OK
Chain ZPC_FDA_CUSTDAILY_TRAN was removed from scheduling
Program RSPROCESS successfully scheduled as job BI_PROCESS_ABAP with ID 05302100
Program RSPROCESS successfully scheduled as job BI_PROCESS_DROPINDEX with ID 05302100
Program RSPROCESS successfully scheduled as job BI_PROCESS_INDEX with ID 05302100
Program RSPROCESS successfully scheduled as job BI_PROCESS_LOADING with ID 05302100
Program RSPROCESS successfully scheduled as job BI_PROCESS_LOADING with ID 05302101
Program RSPROCESS successfully scheduled as job BI_PROCESS_ODSACTIVAT with ID 05302100
Program RSPROCESS successfully scheduled as job BI_PROCESS_TRIGGER with ID 05302100
Chain ZPC_FDA_CUSTDAILY_TRAN Was Activated And Scheduled
Chain Is OK
The same log is repeated for 140 times in the log window.
Please Advise.

hi,
That is not a problem finally the jobs succes right.
and the thing is it is performance problem suppose first time it take error to study the issue and rectify the first instance it is the best way
Regards,
Lakshmi

DTP problem on master data loading

Dear Experts
       When we loading 0Employee master data daily, sometime PSA data can not reflect to master table
     For example, data in PSA as following
10001    01/01/2008     01/31/2008    C1
10001    02/01/2008     08/31/2008    C1
10001    09/01/2008     12/31/9999    C2
     After DTP request success running, the data as following
10001    01/01/2008     01/31/2008    C1
10001    02/01/2008     08/31/2008    C1
10001    09/01/2008     12/31/9999    C1
    the C2 in PSA can not update to master table, this problem happened time to time, but it hardly find it until users' complain.
    When I manually deleted DTP request from Master attribute list and re-load from PSA to master table, the problem can be fixed.
    By the way, I have already set DTP Semantic Groups as Employee ID.
     Anyone can advice it, thanks in advance.
Regards
Jie

jie miao,
Previous master data already available in BW. every day load you need to load only changes. So last record is enough.
Eg:
Initial load on 01-01-2008
10001 01/01/2008 12/31/9999 C1
--> Data loaded on 2nd Jan and above data available in bw.
on 03-11-2008 you will get following records
10001 01/01/2008 11/02/2008 C1
10001 09/01/2008 12/31/9999 C2
Load these records on 03 Nov, so system updates accordingly.
Hope it Helps
Srini

Problem in master data loading in BI 7.0

Hi Experts,
I m loading master data in BI 7.0 server.
It is getting failed due to invalid record.
But when i load the same data in 3.5; it is successful.
Can anybody tell me what may be the problem..
Thanks in advance..
Regards,
Ashwin

Hi,
In BI 7.0 we first load data to PSA through Infopacage.Once data comes in PSA table we do transformation between the PSA table which is tied to the datasource and the data target.
Then we load data to the target through DTP.
May be there is some problem in the transformation.Check your mapping in the transformation.
Kindly update me regarding this.
Regards,
Debjani..

Problem Occrued in Data loading (Time Stamp Error)

Hi,
Iam scheduled info package after its showing time stamp error in the monitor menu header status is red and details menu 0 records . time stamp error in the source system.
my problem is i have all genric transaction data sources i checked in R/3 in RSA3 and Rsa6 i excuted in that particular data sources are active and its contain some records also.
before loading the data i replicated all data sources these all active in BI system. but once scheduled the infopackage all data sources going to in active status.
i Tried to Execute T code SE38 in BW give programme name as RS_Transtrucuture_activate .
but i dont have authorisation these program to excute.if i want get authorisation i have to wait 3 days.
please suggestme another solution procedure.
Regards,
Diwa(Devakar tati Reddy)

Hi,
I am not sure about what BW version are you using. I hope you are using 3.5 or low. After Datasource replication you need to activate your transfer structure / infosource. I understand that you dont have authorization to run the SE38 program to activate the transfer structure. But, you can also try to activate each transfer structure/infosource individually through edit mode of the same. This is the only way out if you donot have the authorization for SE38.
Regards
Kishore

Problems converting FILETIME date/time into MM/DD/YYYY format...

Has anyone successfully converted a FILETIME date/time value into a MM/DD/YYYY format using ColdFusion? I am failing drastically, and it seems like an easy conversion.
FILETIME format (details: http://msdn.microsoft.com/en-us/library/windows/desktop/ms724284(v=vs.85).aspx) is a 64-bit value representing the number of 100-nanosecond intervals since January 1, 1601 (UTC).
A database I'm working with has such values, and I want to display them in a user-friendly date format within my application.
For example, one record is:
13003368600
The above number represents the following date and time:
1/22/2013 @ 2:50 PM
Sadly I know this only because the FILETIME value gets written to the database by a third-party application, and within that application I specify it in the MM/DD/YYYY format.
Can anyone offer me some guidance, or better yet has anyone accomplished this already and want to share code?

I did this, on cf9/linux, and it seemed to do the trick:
<cfscript>
Long = createObject("java","java.lang.Long");
Date = createObject("java","java.util.Date");
fileTimeToEpoch =
// take pwdLastSet From Active Directory, it's in filetime
pwdLastSet = JavaCast("long", Long.parseLong("130292682204519505"));
// take filetime and turn it into epoch/java - 1970/1/1
// http://www.silisoftware.com/tools/date.php - converted: jan 1, 1970 00:00 -00
javaTime = JavaCast("long", pwdLastSet - 116444736000000000);
// convert to milliseconds
javaTime = JavaCast("long", javaTime / 10000);
today = JavaCast("string", Date.init(javaTime));
</cfscript>
<cfdump var="#pwdLastSet#">
<cfdump var="#javatime#">
<cfdump var="#today#">

Problem converting XMLTYPE data into database rows.

Can anyone help with this ?
I�ve got an XML document like the following:
<ROWSET><ROW>><NAME>John</NAME></ROW></ROWSET>
I want to extract the �Name� element and insert it into a table that has one column:
create table employee
(empname varchar2(50));
The following PL/SQL block does the job:
begin
insert into employee
(empname)
select extractvalue(value(xmltab),'/ROW/NAME')
from table(xmlsequence(extract(xmltype('
<ROWSET><ROW>><NAME>Cpty 1</NAME></ROW></ROWSET>')
,'/ROWSET/ROW'))) xmltab;
commit;
end;
However, instead of having the XML string embedded in the actual SQL statement I would like to assign the XML string to a variable (of type XMLTYPE) and refer to the variable in the SQL statement i.e.
declare
v_xml xmltype;
begin
v_xml := xmltype(�<ROWSET><ROW>><NAME>John</NAME></ROW></ROWSET>�)
insert into employee
(empname)
select extractvalue(value(xmltab),'/ROW/NAME')
from table(xmlsequence(extract(v_xml,'/ROWSET/ROW'))) xmltab;
commit;
end;
When I run this I get the following exception:
ORA-22905 Cannot access rows from a non-nested table.
I can�t understand why the first example works but the second example doesn�t. I want to get the second version working because eventually I want to put this code in a stored proc that excepts an xmltype parameter.
Any help would be much appreciated.

Hi
Now I understand why you want to use an insert as select...
The following PL/SQL code should work...
declare
v_xml xmltype;
begin
v_xml := xmltype('<ROWSET><ROW><NAME>John</NAME></ROW></ROWSET>');
insert into employee (empname)
select extractvalue(value(t),'/NAME')
from table(cast(xmlsequence(extract(v_xml, '/ROWSET/ROW/NAME'))as xmlsequencetype)) t;
commit;
end;
Chris

Data load problems

Hello friends
I am facing a problem in the data load. I modified a cube by adding few Characteristics. The characteristics were first added in the communication structure and in the transfer rules. Then I reactivated the update routine. Finally, I deleted any previous data load request for the cube and did a full load. However i wasn't able to find any data for the newly added fields in the Cube.
Did I miss something. Any help will be appreciated in this regard.
Thanks
Rishi

how come ODS came in to picture,this was not mentioned in ur prev post,are u loading it from ODS to CUbe and having problems??
looks u are not using DTP ,in that case check the change log for the newly added fields and then have data flow as ODS>PSA>Cube and check the PSA if those fields are present ..if yes check for update rules in debugging mode of those gets deleted
Hope it Helps
Chetan
@CP..

ERPI: Data Loading problem Hyperion Planning & Oracle EBS

Hi
I am trying to load data from Oracle EBS to Hyperion Planning.
When i push data Zero rows are inserted in Target.
When i look at Table " SELECT * FROM TDATASEG "
It is showing me data but it is not comminting data in Target application.
The reason is Data difference in Source (EBS) and Target.
In my source Year is 2013 but in Target 'FY14' so for Entity Source is '21' but Target is '2143213'
Can you please let me know how to solve this issue?
can i place a Lookup table for this in EPRI.
i am using EPRI and ODI to push data.
Regards
Sher

Have you setup the data load mapping correctly to map the source value to the proper target value? Based on what you are describing it seems that the system generated * to * map is being used, if you are mapping a source to a different target, this needs to be added to the data load mapping.

Troubleshooting 9.3.1 Data Load

I am having a problem with a data load into Essbase 9.3.1 from a flat, pipe-delimited text file, loading via a load rule. I can see an explicit record in the file but the results on that record are not showing up in the database.
* I made a special one-off file with the singular record in question and the data loads properly and is reflected in the database. The record itself seems to parse properly for load.
* I have searched the entire big file (230Mb) for the same member combination, but only come up with this one record, so it does not appear to be a "last value in wins" issue.
* Most other data (610k+ rows) appears to be loading properly, so the fields, in general, are being properly parsed out in the load rule. Additionally, months of a given item are on separate rows, and other rows of the same item are loading properly and being reflected in the database. As well as other items are being loaded properly in the months where this data loads to, so, it is not a metadata-not-existing issue.
* The load is 100% successful according to the non-existent error file. Also, loading the file interactively results in the file showing up under "loaded successfully" (no errors).
NOTE:
The file's last column does contain item descriptions which may include special characters including periods and quotes and other special characters. The load rule moves the description field to the earlier in the columns, but the file itself has it last.
QUESTION:
Is it possible that the a special character (quote??) in a preceding record is causing the field parsing to include the CR/LF, and therefore the next record, into one record? I keep thinking that if the record seems to fine alone, but is not fine where it sits amongst other records, that it may have to do with preceding or subsequent records.
THOUGHTS??

Thanks Glenn. I was too busy looking for explicit members that I neglected thinking through implicit members. I guess I was thinking that implied members don't work if you have a rules file that parses out columns...that a missing member would just error out a record instead of using the last known value. In fact, I thought that (last known value) only worked if you didn't use a load rule.
I would prefer some switch in Essbase that requires keys in all fields in a load rule or allows last known value.

Converting binary data to cluster

Hello,
I have a problem converting binary data to a cluster. I have to create a VI that reads out some parameter from a device, which is connected via RS232. I get an array of 128 Byte from the device. The data has the following (C-)structure:
struct PARAM
char cVersion[64];
float fDeviceID;
float fChannels;
float fSensor;
float fTrigger;
float fConst[5];
char cReserved[28];
How can I convert the binary array to a cluster? I created a cluster (see attachment), but I couldn't connect it with the "Type Cast" VI. Has anyone an idea how to solve this problem?
Thanks, Martin
Attachments:
cluster.jpg ‏16 KB

It is tedious to convert binary data to LabVIEW data when you have arrays. You have to typecast the data with clusters of the same fixed length as in the structure. Large clusters are easily created on the diagram using Array to Cluster primitive. See the attavhed picture.
LabVIEW, C'est LabVIEW
Attachments:
conversion.gif ‏11 KB

Essbase data load process never terminating

Hi
We are using Essbase 11.1.2.1.
We try to load a data file into a BSO Essbase application (essbase export format, so no dataload). We are using "execute in background" option.
If the file contents no unknown member : it is correctly loaded.
If the file contents an unknown member : the data load is never terminated : we have killed it after more than 15 hours, and the request is still in "terminating" so we have to kill the essbase process on the server. No error message or file.
Thanks in advance for your help
Fanny

Hi
I try to start this discussion again, because we still have the same problem, and no idea of why (no idea from the Oracle support too) and my client is waiting for an explanation :-/
So, some more details :
Context :
I have an export (essbase format) from a database.
If I import this file in a database with all the members used in the file => no problem. The dataload takes around 10 seconds.
If I import this file in a database with all the members used in the file, excepted 1 => problem. The data load never ends and I have to kill the Essbase application on the Essbase server.
I have done another test :
I have imported the file in a database with all the members used in the file (so no problem).
I have exported level-0 data in columns.
I have imported this new file in the database with all the members used in the file, using a DLR => no problem. The data load takes around 10 seconds, with a dataload.err generated.
So, John, you are right, the client shoud really use a column format export. But I need to explain to him why he cannot (not should not) use an export format.
If using an export format is possible : I have to identify my problem and make the right modification to solve the problem.
If it not possible : this is a bug, and it is not my responsability anymore and I can close the project!
Thanks in advance for your great ideas!
Fanny

Master Data Load - Number records not match

I am loading the 0PLANT table. In R3, it shows 966 records. After running the InfoPackage, there are 966 records in the PSA, none in error. But I get a message - Number of records requested does not match number transferred (packet 1).
This is a first use of a new source system to this BW. All master data loads give me the same error.
What is setup wrong?

Hello RoberT,
& How r u ?
For the same 0PLANT we load data on daily basis. Not only for this but also for more Master Data Objects. For some of them we used to get this message but not as an Error. On pressing the Update Tree (Refresh) it will change the status.
Anyway u got the records in the PSA so there is no problem with the data load. What signal u find in the Monitor ? Yellow or Red ? If yellow wait for some time, press Update Tree.
Best Regards....
Sankar Kumar
+91 98403 47141

Problem converting static data load mapping to MOLAP

Similar Messages

Maybe you are looking for