DS JOB Continuously running

kmsekhar · Post by **kmsekhar** » Mon Sep 24, 2012 4:23 am

Hi All,
DataStage Parallel job continuously running in prod environment.

Same code is executing fine in QA environment....

Even executed the same job from QA and pointed to PROD Target database, its running fine...

Could you please suggest me how to know the reason for long running jobs?

ArndW · Post by **ArndW** » Mon Sep 24, 2012 6:32 am

What is the job doing and if you monitor it:
- Has the job processed any data?
- If so, then has it process all of the expected data and then "hangs"?
- Does it use any CPU in the "hang" state?

Other questions would depend upon the answers to these.

BI-RMA · Post by **BI-RMA** » Mon Sep 24, 2012 6:39 am

Since You do not provide any information as to what your job is actually doing - except from the fact that You perform some kind of write-operation on some kind of database - it is pretty hard to give any advice whatsoever.

kmsekhar · Post by **kmsekhar** » Mon Sep 24, 2012 6:44 am

Thanks for your reply Arnd...

- No data processed.
- CPU process is 97% idle time.

- In the log job got hanged to invoke startup script: main_program: APT Startup script: /opt/etl/InformationServer/Server/PXEngine/etc/startup.apt

After adding Env variable $APT_STARTUP_STATUS = True, below is the log: and still job is in running state

Code: Select all


   Occurred: 8:28:28 AM      On date: 9/24/2012            Type: Info
   Event: main_program: APT configuration file: /opt/etl/InformationServer/Server/Configurations/prod1x.apt (...)

   Occurred: 8:28:28 AM      On date: 9/24/2012            Type: Info
   Event: main_program: Max inbound connections per node = 3 (...)

   Occurred: 8:28:28 AM      On date: 9/24/2012            Type: Info
   Event: main_program: Defining Section Leaders.

   Occurred: 8:28:28 AM      On date: 9/24/2012            Type: Info
   Event: main_program: Contacting Section Leaders.

   Occurred: 8:28:28 AM      On date: 9/24/2012            Type: Info
   Event: main_program: APT Startup script: /opt/etl/InformationServer/Server/PXEngine/etc/startup.apt

   Occurred: 8:28:28 AM      On date: 9/24/2012            Type: Info
   Event: main_program: Broadcasting score.

   Occurred: 8:28:28 AM      On date: 9/24/2012            Type: Info
   Event: node_node1: broadcastStepIR: score load from /tmp/APTps9380286e3cfd8f on node node1 started.

   Occurred: 8:28:28 AM      On date: 9/24/2012            Type: Info
   Event: main_program: Score (416,891 bytes) sent.

   Occurred: 8:28:28 AM      On date: 9/24/2012            Type: Info
   Event: main_program: Starting Players.

   Occurred: 8:28:28 AM      On date: 9/24/2012            Type: Info
   Event: main_program: Waiting for Players to start.

   Occurred: 8:28:28 AM      On date: 9/24/2012            Type: Info
   Event: main_program: Setting up data connections among Players.

   Occurred: 8:28:28 AM      On date: 9/24/2012            Type: Info
   Event: main_program: Starting step execution.

kmsekhar · Post by **kmsekhar** » Mon Sep 24, 2012 6:52 am

BI-RMA wrote:Since You do not provide any information as to what your job is actually doing - except from the fact that You perform some kind of write-operation on some kind of database - it is pretty hard to give any advice whatsoever.

Oracle Source -> Transformer (1-1 mapping) -> Oracle Target (Upsert mode --> Inserting only new records) --> Peak stage (to reject duplicates/updated records from source)

ArndW · Post by **ArndW** » Mon Sep 24, 2012 7:14 am

Can you make a copy of the job with just the source Oracle stage going to a PEEK and attempt calling that up to see if the problem remains reproduceable?
Also, does that Oracle stage contain SQL or even before/after code?

kmsekhar · Post by **kmsekhar** » Mon Sep 24, 2012 7:29 am

Job finished sucessfully for the above design.

No Before/after code in SQL Query.

ArndW · Post by **ArndW** » Mon Sep 24, 2012 7:41 am

If you add the option "$APT_DISABLE_COMBINATION" and set to "true" in the original job does the error picture remain the same?