SOA Purging in Efficient Way

Problem Statement

Using Oracle’s out of the box, purging of SOA instances was not deleting all the old and unwanted instances. The out of the box purging script only delete particular instances which are in 1, 3, 17, 19, 31, and 32 states. This problem leads to piling up of old unwanted instances which are not in the state as mentioned earlier and now we are struggling to delete instance in 5, 8, 9, and 10 states.

NOTE: State and its definition

1 – Completed

3 – Completed with faults

5 – Completed with recovery required

8 – Completed with faults and recovery required

9 – Completed with suspended

10 – Running with faults and suspended

17 – Completed with terminated

19 – Completed with fault and terminated

31 – Completed with faulted, recovery required, suspended, and terminated

32 – Unknown

Solution

For Oracle SOA Suite: 11.1.1.7 and Database version: 12c

Our aim is to purge all the older instance and do not worry about the state.

Perform Serial Purging:

In our existing out of the box purging script name – soa,delete_instances procedure has below parameters –

SSE_SOAINFRA.soa.delete_instances (

min_creation_date => MIN_CREATION_DATE,

max_creation_date => MAX_CREATION_DATE,

batch_size => batch_size,

max_runtime => max_runtime,

retention_period => retention_period,

purge_partitioned_component => false);

END;

One needs to modify the parameters to look like –

SSE_SOAINFRA.soa.delete_instances (

min_creation_date => MIN_CREATION_DATE,

max_creation_date => MAX_CREATION_DATE,

batch_size => batch_size,

max_runtime => max_runtime,

retention_period => retention_period,

ignore_state => true,

purge_partitioned_component => false);

END;

In short, we are adding “ignore_state => true,” as it deletes pre-defined state of instance and serves our purpose.

Benefits

1.     Purging all old and unwanted instance

2.     Enterprise manager cannot slow down due to the accumulation of unwanted old instances

Join the Conversation

  1. Ritu Panjwani's avatar
  2. anomaly1984's avatar
  3. Varsha Panjwani's avatar
  4. Unknown's avatar
  5. Dilip Panjwani's avatar
  6. Unknown's avatar

30 Comments

  1. That’s really helpful for maintenance of high volume of data, thanks Dinesh. I will surely inform my BAU teams to take a look at this and resolve our maintenance issues.

    Like

Leave a comment

Design a site like this with WordPress.com
Get started