Service Outage
Credit: Adobe

Bug and tracking system Atlassian Jira has gone through a service outage since being first reported on April 4, with the system taking nearly two full weeks to resolve all its issues and become fully functional again. The disruption, coupled with a lack of communication from Atlassian led to users being frustrated, since the company’s Jira software comprises 84% of the bug and tracking market. Between 50,000 and 800,000 end-users are believed to have been affected by this outage, leading to many being baffled and disappointed with how Atlassian handled the issue.

Why Atlassian experienced the service outages

Per a blog entry written by company CTO Sri Viswanath, the ongoing problems are not the result of a cyberattack, but rather a conflict between platforms that led to some sites being deactivated.

“One of our standalone apps for Jira Service Management and Jira Software, called ‘Insight – Asset Management,’ was fully integrated into our products as native functionality. Because of this, we needed to deactivate the standalone legacy app on customer sites that had it installed,” Viswanath wrote in the blog. “Our engineering teams planned to use an existing script to deactivate instances of this standalone application. However, two critical problems ensued.”

SEE: Feature comparison: Time tracking software and systems (TechRepublic Premium)

The first problem outlined by Atlassian’s CTO included communication issues between the team needing the sites in question to be deactivated, leading to the wrong IDs of the application that should have been disabled to be confused with the IDs for the entire site, causing outages for customers employing the Atlassian software. The other difficulty experienced by the company’s teams came from faulty scripts, having mixed up the script for temporary deletion with the script for items that needed permanent removal. By designating sites for permanent deletion that only needed temporary disabling, sites for approximately 400 organizations were improperly deleted, leading to the extended period of downtime.

On April 12, Atlassian told its customers that it could be another two weeks before sites were up and running again, but as of April 18, the company’s status page indicated that all its customer sites using Jira were once again functional.

Will this outage have lasting effects on Atlassian?

While the issues with the scripts seem to be fixed, the issue may have ripple effects down the road for Atlassian’s customers. With the service outage having been what many believe is a self-inflicted problem, this could lead to some customers having second thoughts when it comes to renewing contracts with the bug and tracking company in the future.

Atlassian tweeted a statement on April 17 that all users’ concerns had been addressed, but the damage may have already been done by that point. Developer and the Pragmatic Engineer newsletter author Gergely Orosz said that from April 4 up until the 12th, one Atlassian user had not received any notice from the company on updates outside of Atlassian’s status page.

This directly contradicted the claim that Atlassian had communicated updates on Jira’s status with each customer, leading to some being frustrated with the lack of information from a service the companies were paying money for. The combination of an unnecessary service outage coupled with the company’s lack of transparency on the issue could lead to some potential customers avoiding the platform as a whole in the future.