Today I have been reported an incidence, S_ESCL_ACTN_REQ table was growing a lot, this is not normal. This table identifies all the rows that are awaiting action execution. In this case before worked fine, but after investigating with the development team is seen throwing a query at a later point a workflow has degraded processes Policies having a worse performance.
The solution is otpimize the query and reduce execution time, of course, new Processes Workflow Policies need to be created to share the load and Improve performance.
Siebel Workflow Policies use three database tables for processing and tracking requests, and these three tables are key in Siebel
It is very important to monitor These tables to Verify That Correctly policies are being processed. The number of tuples to be as low as possible or tend to zero. As DBAs we can monitor these tables in different ways, pl / sql, shell script, Grid Control, third-party tools such as OpenView, BMC Patrol, Nagios.
We are also responsible for maintenance, it is a good idea to cache these tables into memory and ensure quick access, as it can become a bottleneck that chokes entire system. You can cache the memory tables by run the statement;
alter table SIEBEL.S_ESCL_ACTN_REQ storage(buffer_pool keep);
If not using automatic memory, please refer to the DB_KEEP_CACHE_SIZE parameter.
HTH – Antonio NAVARRO