Abstract
Software unavailability can lead to disastrous consequences ranging from delays and cancellation to a loss of millions of dollars of technology. However, research on the causes that can make systems unavailable is still little. The aim of this paper is to investigate such causes in an industrial context using 2 qualitative approaches ((a) interview and (b) retrospective analysis) using archival data and records from a student information system. As a result, connectivity, human, hardware, storage, and software were found to be the important categories of causes for the unavailability. Besides, the critical lessons to be learned that relate to activities of software management and software business are also discussed. This includes vendor support, systems documentation, health check process, licensing, and software updating or upgrading. To strengthen the claim that our findings are promising, we compare our main findings with those of a previous study. This study can generally assist software engineering people including engineers, developers, project managers, vendors. Additionally, this paper dis cusses the threats to the study's validity and suggests open problems for future research.