This document serves as a Root Cause Analysis for the Primo service interruption experienced by Ex Libris customers on February 15 and 17, 2017
The goal of this document is to share our findings regarding the event, specify the root cause analysis, outline actions to be taken to solve the downtime event, as well as preventive measures Ex Libris is taking to avoid similar cases in future.
Service interruption was experienced by Ex Libris customers served by the Primo MT EU02 instance at the Europe Data Center during the following hours:
February 15, 2017 from 12:51AM until 1:00AM Amsterdam time zone
February 17, 2017 from 12:32AM until 1:13AM Amsterdam time zone
Event was identified by the 24x7 monitoring. During the event, the service was unavailable for the environment.
Root Cause Analysis
Ex Libris Engineers investigated this event to determine the root cause analysis with the following results:
The service was unavailable due to high CPU load on the Oracle database server.
In this case oracle selected a wrong SQL execution plan that led to full table scan which is a heavy CPU consumption operation.
Our database engineering found out that for several specific SQL queries it is better to use a fixed (static) execution plan instead of letting oracle dynamically select an execution plan.
Technical Action Items and Preventive Measures
Ex Libris has taken the following action and preventive measures to avoid such an occurrence in future:
- Ex Libris engineers have examined multiple SQL queries and have updated their execution plan to a better plan.
- Ex Libirs DBAs are in the process of examining additional configurations that could prevent invalid execution plans.
ExLibris is committed to providing customers with prompt and ongoing updates during Cloud events. Ongoing and prompt updates on service interruptions appear in the system status portal at this address: http://status.exlibrisgroup.com/
These updates are automatically sent as emails to registered customers.
|February 23, 2017||Initial publication|