DatatStage Status

[sherkhan]

Hi,
Can any one help out. I have some maps(mercator) i need to call DataStage Job after the execution I wanted to know that when my DataStage Job is completed. Is there any thing that tells us that Datastage job is completed.

Thanks,

 

[BocaBurger]

Not sure if I understand. Mercator (now DataStage TX) maps normally don't have anything to do with DataStage Server jobs.

Are you trying to integrate DataStage TX and DataStage on the same machine?



BocaBurger
<===========================||////////////////|0
The pen is mightier than the sword, but the sword hurts more!

 

[sherkhan]

hi Boca,
We are using mercator 6.7 and now we need to call a Datastage job for loading of data into the database which is very huge.My question is how can we call Data Stage job from mercator map.
Pls urgent.
thanks

 

[BocaBurger]

Are both programs installed on the same server?
If so, you can use the batch (windows) or shell (UNIX) adapter to issue the dsjob command. If you want the map to wait until the DataStage job is completed, you might want to try the -POLL option of thr adapter. If not, just set the -POLL to 1 and 1.

If not on the same machine, can you NFS mount the drive?



BocaBurger
<===========================||////////////////|0
The pen is mightier than the sword, but the sword hurts more!

 

[sherkhan]

Hi, Boca

Yes Both Datastage and mercator 6.7 are in same server.
I never used Unix Adapter. Can u pls help out.

Thanks

 

[BocaBurger]

Not UNIX adapter, Shell Script adapter.
Create a shell script (or have your UNIX admin do it for you). The script should run the DataStage job stahd alone. You will need to make sure the DataStage TX user has permissions to run DataStage.
In the output card of the map, choose the shell script adapter. Provide the full path to the shell script. Make sure the script is given execute permissions and you (the DS TX user) has permissions to the script.

Build the map for the platform you are using and deply like any other map. Run thye map from the command line to make sure it works. If you are using the Event Server, put the map in a system and deploy.

BTW, if you have DataStage and DataStage TX Event Server on the same machine, I hope you have plenty of CPU's and RAM. They both can be hogs, depending on what you are doing.

Might be a good idea to upgrade both to 7.5.1. Then you already have everything you need to run MapStage. You can create DataStage jobs and use the TX map for all your complex transformations. You also have the use of all the TX adapters. You can use the Event Server to automate error handling in both DataStage and TX.





BocaBurger
<===========================||////////////////|0
The pen is mightier than the sword, but the sword hurts more!

 

[sherkhan]

Hi Boca,
Thanks a lot. I will on that work and get back to you. when i complete.
thanks,
sher..

 

[vmcburney]

Bocaburger is spot on, under the 7.5.1 release DataStage parallel jobs can use the DataStage TX Map stage, it has options such as Target Map Directory, Create map from input and output links, select map and map files etc. A DataStage job can then receive or load flattened data to and from this map stage. For an output link you define the card and card type along with the column metadata.

This will solve your triggering problem and should give greatly improved performance as your data is passed in memory from TX to the parallel job for immediately processing.

 

[BocaBurger]

And if you don't want to flatten the data, you can use MapStage for the entire transformation.
EDI to XML for example.



BocaBurger
<===========================||////////////////|0
The pen is mightier than the sword, but the sword hurts more!