Ulrich - RCA - March 20, 2024

Introduction

This document serves as a Root Cause Analysis for the Ulrich service interruption experienced by Ex Libris customers on March 20, 2024.
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.

Event Timeline

Service interruption was experienced by Ex Libris customers served by the Ulrich environment in the Seattle Data Center during the following hours: 

March 20, 2024 from 08:20 until 11:28 Seattle time.

During the event, the services were unavailable.

Root Cause Analysis

An issue was encountered with high CPU usage on one of our UlrichsWeb Database servers. This excessive load disrupted connections to the database, preventing UlrichsWeb application from functioning properly, as the SQL server was unable to process incoming requests.

Our Cloud engineers addressed the situation by aborting resource-intensive CPU jobs and initiating a failover to a SQL server cluster node. As a result, the service was restored.

Technical Action Items and Preventive Measures

• Our engineers have fine-tuned resource-intensive processes, enabling parallel execution and smoother handling of heavy indexing tasks.

• Our DBA engineers have added a monitoring layer to the database, to alerts before CPU usage reaches critical levels, allowing timely intervention.

• Our Cloud engineers have optimized the memory allocation for the databases to ensure efficient performance and avoid resource bottlenecks.