Tm Workflow Fails With An Error (Doc ID 1401730.1)

Last updated on MARCH 02, 2017

Applies to:

Siebel Territory Management - Version 8.1.1.3 SIA[21219] and later
Information in this document applies to any platform.

Symptoms

We are using Territory Management to process accounts, contacts, and opportunities using TM Account, TM Contact, and TM Opportunity objects.


The Load Splitter configuration is the following:

Threshold: 10,000
Minimum Batch Size: 5,000

Server Configuration:

Priority: 1
Server Name: siebsrvr1
Number of Tasks: 20

When data volume needs more tasks than Number of Tasks configured by load splitter, the Run Alignment task failed with the error:

ObjMgrBusServiceLog Error 1 000000304dc21e04:0 2011-05-05 17:55:19 (loadspltrsvc.cpp (463)) SBL-ASG-50403: The number of tasks required for this process exceeds the number of available tasks.

We have loaded additional data into TM Test Environment (about 300 000 accounts and about 250 000 contacts). It's a data for a new market.

The same alignment runs successfully with the smaller subset of data.

We have Solution Support Engineering on site, we did some tuning for attempting to get successful alignment for these large amount of data:

1.

- changing the MaxTasks settings for the following components: SRBroker, AsgnSrvr, AsgnBatch
- Reducing the account records count by 2 times, for this test Accounts number 200,000 - 250,000 and contacts about 200,000 - 250,000.

The Run Alingment succeeded, but when I ran the Activation alignment part it still failed with the message in the AsgnBatch task:

[DataDirect][ODBC Oracle driver][Oracle]ORA-04031: unable to allocate 4024 bytes of shared memory ("large pool","unknown object","kxs-heap-b","bind var buf")

2.

- Add 2nd Siebel Server to Load Splitter
- Load Splitter: Minimum Batch Size: 20,000
- Load Splitter: Number of Tasks: 60

One or more Batch Assignment jobs failed due to Deadlock error:

Additional Message: pfNativeError: 60; szSQLState: S1000; szErrorMsg: [DataDirect][ODBC Oracle driver][Oracle]ORA-00060: deadlock detected while waiting for resource

I was able to get this to work by doing the following tuning:

- Reduced Load Splitter: Minimum Batch Size to 15000.
- Due to reducing the number from 20000 to 15000, I had to increase the Load splitter Number of Tasks from 60 to 80, and MaxTasks for AsgnBatch (Batch Assignment component) to 320 and SRBroker component to 360.
- Ran the alignment again. It ran successfully.

I hope to get an exact instruction on how to calculate the MaxTasks for the components and Load Splitter and how to make sure the Production server can handle that number of tasks. Production environment currently has one Siebel Server for Territory Management.

Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms