Unable to read Teradata table from DS parallel JOB

[venkateshkumarg]

Hi,

I am trying to read a Teradata table from the DS parallel job using Teradata Enterprise Stage. Its giving me the following error. Also can't view the data.

TeradataTble: Unrecognized argument: -db_cs

But I could connect thru DS manager to import the metadata definitions.

DataStage Version 7.0
Teradata2 Version 6.1
DataStage Server: Unix.
What is the environment variable to set?. or check it out.

[phanee_k]

Hi ,
What teradata utility are using to read the data from teradata.

Check whether proper Database Name and other properties in Teradata Stage are mentioned appropriately.

If using Fast Export Stage , unselect the " Use Large File AXSMOD of FastExport " on the General tab.


Thanks,
Phani

[venkateshkumarg]

Hi Phanee,

Its a Teradata enterprise stage. even I set environment variables in user defined..thru administrator as

APT_TERA_SYNC_DAEMON_PROG = 'servername'
APT_TERA_SYNC_DATABASE = 'db name'
APT_TERA_SYNC_USER = 'user'
APT_TERA_SYNC_PASSWORD = 'pwd'..

Still the issue is there.

[jking123]

Try this first:
0.Create a copy of job.
1. Use a user which has write privilages to database being read from.
2. Remove all APT_TERA env variables.
3. use unecrypted password.
4. See if the above works.

If the above works then try using encrypted password.

If both work then go back to orignial job and try with unecrypted password for TERA parameter.
Also at one point I believe but am not sure APT_TERA_SYNC_PASSWORD was actually
APT_TER_SYNC_PASSWORD without the A.

[venkateshkumarg]

Still this issue has not been resolved. We are using Teradata API stage instead of Teradata Enterprise Stage. But its perfomance is very poor.

[DSguru2B]

Is this problem having with all the tables/multiple tables or is it just that one table. Also try re-importing the metadate. Sometimes it so happens that the dba decides to change the column name and hence it will create an error doing the select (view data). My dba did that. That is offcourse, if you have a column in that table that has a name close to db_cs.
Also, i believe that in DS 7.0, you could not view data if you have environment variables defined. Someone correct me if i am wrong.

As far as performance goes, that depends upon a number of factors. Like how many bytes of data in a particular row, what sql functionality (update, insert, upserts, delete first and then insert, etc.) is being performed, are there triggers on that table thats slowing it down, how often you pass a commit, whats the transaction size etc.