Open failed for update

[kamalshil]

Hi,
i searched for this error but was not able to find solution.

i am running my job on 4 node and its updating 1 million records.
I am using oracle enterprise stage for reading and writing the data.

I am reading same table in source and trying to update after dome transformation in target stage.

Code: Select all

  ORA,3: Open failed for update: (...)
    update is: SQL Statement
    sqlcode is: -60
    esql complaint: ORA-00060: deadlock detected while waiting for resource



   ORA,3: Unable to use a record for update   
   ORA,3: Records inserted: 0 (...)
   ORA,3: the runLocally() of operator [ORA], partition 3 of 4, processID 5460100 on node3 failed.
   ORA,3: Input 0 consumed 53 records.
   ORA,3: Output 0 produced 0 records.
   ORA,3: Operator terminated abnormally: runLocally did not return APT_StatusOk

 

What would be possible cause of job being getting aborted.

[Sainath.Srinivasan]

The deadlock message specifies the issue.

Check whether there is any lock in the underlying table.

Try running your job in 1 node.

If the problem disappears in 1-node, then check the key partitioning.

[mayura]

problem is with table ,,, remove the deadlock of the table and try to run ..
also check with update query..

[chulett]

The problem is not "with the table" but rather with how the data is being sent to the table, something that is causing a deadlock to occur. Talk to your DBA about possible causes and solutions to that error, so you understand what is going on in the database.

Typically in DataStage it involves properly partitioning your data to avoid causing the problem. You should also be able to solve it by running the job on one node.

[kamalshil]

Hi,
it works for me after i implemented sort on the input link for all the keys.