Auto Termination of BI report jobs causing OOM exception due to complex templates — Oracle Analytics

Oracle Transactional Business Intelligence Idea Lab

Welcome to the Oracle Analytics Community: Please complete your User Profile and upload your Profile Picture

Auto Termination of BI report jobs causing OOM exception due to complex templates

1
Views
0
Comments
Jharna Patel
Jharna Patel Rank 1 - Community Starter

Organization Name (Required - If you are an Oracle Partner, please provide the organization you are logging the idea on behalf of): Metrolinx


Description (Required): We recently encountered an issue that one of our BI job was causing the instance to go down when submitted for larger volume of data (1 year). The sab job completed within minutes when ran for smaller data sets. On investigation by Oracle support team, it was identified that the RTF template logic was having a very high complexity value causing an Out of memory exception for large data volume.

Currently there is no way for developer to calculate the complexity of the RTF templates. Also, as developers usually use smaller data sets to design templates, this issue cannot be identified unless tested with large volumes.

Testing with large volumes of data in this case was that the job kept trying to utilize all the available memory causing the instance to become unavailable for a 3-4 hours.

We therefore need a mechanism to auto terminate the jobs in such cases before it causes issues at the instance level.

Use Case and Business Need (Required):

Extracting a year of data is a very common scenario specially for Payroll. For year end processing we have many custom reports extracting the data for entire year. Reporting this volume of data in a readable format often leads to complex template designs.

Enhancement Request / Service Request: Need a mechanism to auto terminate jobs when it is utilizing too much memory for processing so as to not cause an impact with instance availability.

Need a mechanism to identify the complexity of the templates and to estimate the volume of data it can process with the given complexity without causing memory issues

1
1 votes

Submitted · Last Updated