DSXchange

By way of preparing everyone for the Brave New World of 7.0, Hash Files are NOT always the faster method. It does not make sense to use a hash file in a real-time application where volume is not a factor. If volume is not a factor here then I would explore using a group function on the SQL of TAB2. ...

Amen to that, brother.

We are 5.2.2, Oracle 9.02 on HP-UX 11.11. I imagine it is close enough to your installation for the following to apply. 1. You can use ODBC UNIX with Oracle. Not a problem. 2. Try changing your REJECT link to have a contraint where the DBMSCODE <> 0. Sometimes Oracle will return the ORA- code that c...

Two Comments; 1. ETL is no longer synonymous with the word "BATCH". Staging data is meaningless in a real-time environment. Transaction Control rules here. It is paramount for restartability and recoverability. 2. I wonder how segment processing will evolve with the introduction of 7.0 or ...

Arg1 = 583 TRUE Arg2 is not MARSHAL ISL OR Arg2 is not KWAJALEIN Your code as written appears to be true to me in both cases you've presented. MARSHALISL is not KWAJALEIN AND 583 is one of the Arg1 choices. Likewise for Test. Test is not MARSHALISL and Test is not KWAJALEIN so they are both true AND...

We are realtime so performance is measured differently i.e. how fast to do a single row versus how many rows per second but... XMLReader was faster then XML Input because it has less functionality. Our XML was no more than 5000 bytes long. It took somewhere in the range .1-.5 secs per row. The Oracl...

I have job 1 and job 2. They both run a stored procedure and returns columnID value. in job 1 I set up a transformer constraint that looks like this. "[dslin1].columnID = 0" "[dslink2].columnID = -1" Reject Row Checked. in job2 i just output a sequential file. how can i set up a...

We're running HP-UX so it may be vastly different but... If we shutdown the ETL Engine while netstat shows connections then we get to do an OS reboot! The ETL Engine will NOT start again if it is shutdown while there are active connections. Your only option at that point is an OS re-boot. Never shut...

I doubt it. Datastage has no reason to default a stored procedure call transaction to read only. Did you try the stored procedure from SQLPlus? Were the results the same? It appears to me to be an Oracle issue.

You can use CALL in the OCI Stage as User Defined SQL. You can pass in parameters. It is unsupported. To make sure the parameters are bound correctly make all the columns KEY = YES. You cannot get any values back. If you are hoping for some error checking be aware that not all errors returned from O...

If you are passing the stored procedure correctly, DataStage is oblivious to what happens within the stored procedure. What is the error you are getting? Is it an Oracle error? Depending on your need you could define the stored procedure parameter as INOUT. However, you'll never get anything out. Or...

I am not aware of any. The problem is that DataStage will not recognize the data type. I've got the same problem with CLOB. I have to invoke the to_char() everytime I deal with a CLOB field via DS. Another method I've seen is to make the datatype of the timestamp field a varchar2 since it appears yo...

Sorry, I'm one paren short of a full load. It's the closing paren giving you the very informative missing comma error.
Try
insert into temp (field1) values(to_timestamp('2003-09-24 17:59:00.123456','YYYY-MM-DD HH24:mi:ss.ff6'));

Try this user defined SQL;
insert into table () values (to_timestamp(:1, 'YYYY-MM-DD HH24:mi:ss.ff6').
Make sure the bind variable column (:1) data is in the desired format. You don't have to go all the way to microseconds. It can be any valid date up to ff6.

Put the following in a routine. It works when we upgraded to 5.2.2 running on HP-UX 11.11 so it should work for you in DS 6.0. CurrTime = SYSTEM(12000) Ans=OCONV(DATE(), "D/YMD[4,2,2]") : " " : OCONV(CurrTime, "MTS:") : ".":field(CurrTime,".",2) Outp...