Atlassian products are frequently found at the heart of the financial systems that many governments and commercial enterprises depend upon. As a result, the consequences can be significant when something goes unexpectedly wrong with their Atlassian stack.
Kristoffer Hansen is a seasoned software developer who has worked for some of Norway's most important financial institutions since 2019. Today, he works for a Norwegian government organization that enables seamless cash flow between 1,300 government entities. Together, those entities account for 2% of the Norwegian government's yearly expenditures. To deliver on this vital mission, the organization employs approximately 150 software developers, of which roughly 1/3 are external consultants.
The IT team within that organization depends on Jira, Jira Service Management, and Bitbucket. One morning in August 2024, Kristoffer came to the office at 8:30 in the morning and was met by the following message:
The entity in question had hit the license ceiling for Bitbucket, Jira, and Jira Service Management. And the message created tremendous stress and anxiety among the 150-FTE-strong developer organization.
For regulatory and practical reasons, the government entity uses Atlassian Data Center hosting, and this morning, they experienced a situation caused by a licensing mechanism that many are unaware of. This organization assumed that when the number of Atlassian users in their instance hit the maximum level of their licensing tier, the only consequence would be the Atlassian administrator's inability to add more users.
In reality, all users were locked out of the Atlassian products on which they depended. For the developers involved, this meant they were suddenly unable to do their job. All software production stopped.
Luckily, the situation was resolved within roughly 30 minutes by having the Atlassian administrators execute a manual "emergency cleanup" of users.
Still, the harm was done. The developers' trust in their employer was diminished, and the manual emergency work created could have been easily avoided.
Any organization relying on Atlassian products can prevent this by configuring automatic, rule-based user cleanups that run regularly, typically weekly or bi-weekly.
In this particular organization, the problem arose from a regular influx of new JSM users from other related government entities. This steady stream of new users happens organically, and by definition, it consists of people who are not internal employees who are covered by structured onboarding and offboarding processes.
In high-turnover situations like this, it is imperative to set up regular user cleanup routines.
This is just one recent example I encountered. Collectively, my team and I have been exposed to dozens more.
How about you? Have you encountered similar situations? How was it overcome? What steps have you taken to prevent a reoccurrence?
I'd love to hear your story in the comments below. You can also reach me privately via a direct message on LinkedIn.
Anna-Karin Østlie _Kantega SSO_
6 comments