Now every time you run a database migration, you have to adjust N tables - and in Atlassian's case, N is 200000. Is that better? It depends. There is no "best" way of doing multitenancy.
That’s just an automation issue. It’s not like you have to write a bespoke database migration script per DB.
Re:Inside the longest Atlassian outage
by robertlagrant on Wednesday April 13, 2022 @04:24PM
(#31019796)
The bug we are mitigating was also just an automation issue.
It’s also pretty easy to foobar up a single DB instance if you don’t have proper guardrails in place.
Automation wasn’t the issue here. It’s the symptom not the cause.
Re:Inside the longest Atlassian outage
by robertlagrant on Wednesday April 13, 2022 @05:23PM
(#31020476)
No, the symptom was the loss of customer data.
The best would be multiple separate database instances, which is not even hard to manage specially for qualified engineers like Atlassian surely has plenty of. The problem are business decisions of ignoring the tech debt, usually...